JP2014027409A - Image selection method, image selection device, encoder, image selection system, image selection program, encoding program and image encoding selection program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To allow a user to easily grasp a decoded image with a low degree of deterioration in image quality of a feature region.SOLUTION: An image selection method includes: generating a plurality of pieces of encoded data obtained by encoding a plurality of frame images included in a moving image; calculating an index value showing a degree of deterioration in image quality, which occurs due to encoding, on at least one type of feature region which is previously set in each frame image; selecting a decoded image whose degree of deterioration which the index value shows is the lowest from among a plurality of decoded images obtained by decoding a plurality of pieces of generated encoded data; and displaying the selected decoded image.

Description

  The disclosed technology relates to an image selection method, an image selection device, an encoding device, an image selection system, an image selection program, an encoding program, and an image encoding selection program.

  There is a case where a user selects an arbitrary frame image from moving images including a plurality of frame images, and saves or prints as a still image.

  In many cases, a moving image is stored in a state where each frame image is encoded. In reproduction, the encoded data is often decoded and the decoded image is displayed. As a technique related to encoding, for example, a technique is known in which image data to be encoded is wavelet transformed to obtain wavelet coefficients, and quantized coefficients obtained by quantizing the wavelet coefficients are encoded.

JP 2011-239066 A

  When an arbitrary image is selected from a plurality of decoded images with priority given to the image quality of a specific area rather than the entire decoded image, the degree of image quality degradation due to coding noise (for example, block noise or mosquito noise) in the area It is desirable to be able to quickly select a decoded image having a small size.

  However, for example, in the method in which the decoded image is displayed as a thumbnail (reduced display) and the user selects an arbitrary image from the reduced image, the display image is too small and the coding noise of the moving image is used. It is difficult for a person to confirm. Further, when the decoded image is displayed at the same magnification and the coding noise is visually compared between the frames, there is a problem that it takes time and effort.

  An object of the disclosed technique is to allow a user to easily grasp a decoded image with a low degree of deterioration in image quality of a feature region.

  The disclosed technique generates a plurality of pieces of encoded data obtained by encoding a plurality of frame images included in a moving image, and sets an index value indicating the degree of image quality degradation caused by the encoding in each frame image. Calculation is performed for at least one type of feature region set in advance. Further, the disclosed technique selects a decoded image with the lowest degree of degradation indicated by the index value from a plurality of decoded images decoded from the plurality of generated encoded data, and selects the selected decoded image. Present.

  As one aspect, the disclosed technology has an effect that a user can easily grasp a decoded image with a low degree of deterioration of the image quality of the characteristic region.

It is a block diagram of the image selection system which combined the encoding apparatus and image selection apparatus which concern on embodiment. It is a block diagram of an encoding apparatus. It is a block diagram of a production | generation part. It is a figure which shows an example of the extraction result of a face area. It is a figure which shows an example of a some fixed area | region. It is a figure which shows an example of the extraction result of a still area. It is a figure which shows an example of the extraction result of a moving region. It is a figure which shows an example of the division area divided | segmented according to the brightness | luminance. It is a figure which shows the data structure example of encoding information. It is a schematic block diagram of the computer which functions as an encoding apparatus. It is a flowchart which shows the flow of a process of an encoding program. It is a block diagram of the image selection apparatus which concerns on embodiment. It is a block diagram of a decoding part. It is a figure which shows an example of a selection screen. It is a figure which shows the example of a display of a selection screen when a user designates a face area on a selection screen. It is a figure which shows the example of a display of a selection screen when a user designates a fixed area | region on a selection screen. It is a figure which shows the example of a display of a selection screen when the user designates the division area according to the brightness | luminance on the selection screen. It is a schematic block diagram of the computer which functions as an image selection apparatus. It is a flowchart which shows the flow of a process of an image selection program. It is another example of the schematic block diagram of the computer which functions as an image selection apparatus. It is a block diagram when the function of an encoding apparatus and an image selection apparatus is implement | achieved by one computer, and the image selection system is comprised.

  Hereinafter, an example of an embodiment of the disclosed technology will be described in detail with reference to the drawings.

  FIG. 1 shows an image selection system 3 according to the present embodiment. The image selection system 3 includes an encoding device 1 and an image selection device 2.

  FIG. 2 shows an encoding apparatus 1 according to this embodiment. The encoding device 1 is, for example, MPEG2 or H.264. It is a device that encodes a moving image by a method such as H.264, and includes a generation unit 10, a storage unit 40, an extraction unit 42, and a calculation unit 44.

  Each functional unit of the encoding device 1 can be realized by, for example, an electronic circuit, and the generation unit 10 may be realized by, for example, a semiconductor integrated circuit, more specifically, an ASIC (Application Specific Integrated Circuit).

  The generation unit 10 performs an encoding process on a frame image included in a moving image, generates encoding information for each frame image, and outputs the encoded information to a predetermined output destination.

  FIG. 3 shows a configuration of the generation unit 10 according to the present embodiment. The generation unit 10 includes a prediction error image generation unit 12, a conversion unit 14, a quantization unit 16, an inverse quantization unit 18, an inverse conversion unit 20, a decoded image generation unit 22, a predicted image generation unit 26, a switching unit 32, and an entropy. An encoding unit 34 is provided.

  The prediction error image generation unit 12 is supplied with the image information of the frame image included in the moving image as an encoding target.

  Further, the prediction error image generation unit 12 is supplied with the image information of the prediction image from the prediction image generation unit 26. The predicted image will be described later. The prediction error image generation unit 12 calculates difference information indicating a difference between the image information to be encoded and the image information of the prediction image, and supplies the difference information to the conversion unit 14 as information of the prediction error image. In this way, the prediction error image generation unit 12 generates a prediction error image.

  The conversion unit 14 converts the supplied prediction error image information, and separates the prediction error image into horizontal and vertical frequency components. Information on each frequency component obtained by the conversion process is input to the quantization unit 16.

  The quantization unit 16 quantizes the information supplied from the conversion unit 14. The quantization unit 16 performs quantization to reduce the code amount of the prediction error image. The quantization unit 16 supplies the quantized information to the entropy encoding unit 34 and the inverse quantization unit 18.

  The inverse quantization unit 18 inversely quantizes the information supplied from the quantization unit 16 and supplies the information to the inverse transform unit 20.

  The inverse transform unit 20 performs inverse transform processing on the information supplied from the inverse quantization unit 18. Here, the reverse conversion process refers to a process of converting in the reverse direction to the conversion process performed by the conversion unit 14.

  By performing the decoding process by the inverse quantization unit 18 and the inverse transform unit 20, information about the same level as the prediction error image before encoding is obtained. This information is called a reproduction prediction error image.

  The decoded image generation unit 22 is supplied with the image information of the reproduction prediction error image that has been subjected to the inverse transform process by the inverse transform unit 20. In addition, the decoded image generation unit 22 is supplied with the image information of the prediction image generated by the prediction image generation unit 26. The decoded image generation unit 22 generates an image (decoded image) obtained by decoding the image to be encoded by adding the image information of the reproduction prediction error image and the image information of the prediction image, and stores the image information in the storage unit Accumulate in 40.

  The accumulation unit 40 is a memory in which each piece of image information of the decoded image supplied from the decoded image generation unit 22 is accumulated. The stored image information of the decoded image is supplied to an inter-frame prediction image generation unit 30 described later, and is referred to for motion vector detection, motion compensation, and the like when encoding other frame images. Accordingly, among the accumulated decoded images, the decoded image referred to when the predicted image is generated by the inter-frame predicted image generation unit 30 is also referred to as a “reference image”. Which decoded image is referred to depends on the prediction mode of the encoding target frame image.

  The predicted image generation unit 26 includes an intra-frame predicted image generation unit 28 and an inter-frame predicted image generation unit 30. The prediction image generation unit 26 generates a prediction image and supplies the prediction image to the prediction error image generation unit 12 and the decoded image generation unit 22 (via the switching unit 32), and side information obtained when the prediction image is generated. Is supplied to the entropy encoding unit 34. The side information includes, for example, motion vector information and a prediction mode.

  The intra-frame prediction image generation unit 28 generates a prediction image when encoding is performed using the intra-frame prediction encoding method. The intra-frame predictive encoding method is a method of encoding / decoding an image using only information of a frame image to be encoded without using another frame image. Specifically, the motion of a neighboring image along a predetermined direction is estimated from an image of a block around the block to be encoded when one frame image is divided into a plurality of blocks, and the object to be encoded The block image is predicted, and the difference from the encoding target block image is encoded.

  The inter-frame prediction image generation unit 30 generates a prediction image when encoding is performed using the inter-frame prediction encoding method. The inter-frame predictive encoding method is a method of generating a prediction image based on frame images at different times and encoding a difference (error) image between the encoding target image and the prediction image.

  Here, the processing content in the inter-frame prediction image generation unit 30 will be specifically described. Here, each encoded block when the frame image to be encoded is divided into a plurality of blocks (hereinafter referred to as encoded blocks) is used as a unit. Then, a plurality of blocks (hereinafter referred to as reference blocks) spatially close to the encoded block are extracted from a decoded image (reference image) temporally close to the frame image to be encoded. Furthermore, a motion vector indicating a relative positional relationship (shift amount and shift direction) between each reference block and the encoded block is generated as a motion vector candidate. Further, an error between each of the images of each reference block and the image of the coding block is calculated, and a motion vector candidate with the smallest error is selected as a motion vector for the coding block to be processed. Known motion compensation is performed using the selected motion vector, and finally a predicted image in units of frame images is generated. The detected motion vector information is also supplied to the entropy coding unit 34 as side information.

  The switching unit 32 selects one of the prediction image generated by the intra-frame prediction image generation unit 28 and the prediction image generated by the inter-frame prediction image generation unit 30 to select the prediction error image generation unit 12 and the decoded image generation unit 22. To supply. The switching unit 32 is controlled and switched by a control unit (not shown).

  The entropy encoding unit 34 entropy-encodes the information on the prediction error image after quantization supplied from the quantization unit 16 and the side information supplied from the prediction image generation unit 26. Entropy coding is a process of assigning and coding a variable-length code according to the appearance frequency of a symbol.

  Further, the entropy encoding unit 34 receives information on index values calculated by the calculation unit 44 described later. The entropy encoding unit 34 generates and outputs encoded information including encoded data generated by entropy encoding and index value information input from the calculation unit 44 in units of frame images (see also FIG. 9). ).

  The extraction unit 42 extracts at least one feature region from each decoded image stored in the storage unit 40. Depending on the type of feature region, information on a motion vector detected at the time of encoding in the generation unit 10 is also used for extracting the feature region.

  Note that, in the extraction unit 42, the types of feature regions extracted from the decoded images stored in the storage unit 40 are set in advance. For example, it may be an area having a predetermined feature or an area having a preset position and size. The extraction unit 42 outputs the pixel position and the pixel value of each pixel of the extracted feature region to the calculation unit 44.

  The calculation unit 44 receives the image information of the pre-encoding frame image that is the source for generating the decoded image, and the pixel position and pixel value of the feature region extracted by the extraction unit 42. . The frame image before the encoding process that is a source for generating the decoded image is referred to as an original image.

  The calculation unit 44 compares the pixel value of the feature region with the pixel value at the same position as the feature region in the original image corresponding to the decoded image from which the feature region has been extracted. 10 is used to calculate an index value indicating the degree of image quality degradation that occurs when encoding is performed. When there is one type of feature region extracted by the extraction unit 42, the calculation unit 44 calculates an index value for the one type of feature region. In addition, when there are a plurality of types of feature regions extracted by the extraction unit 42, an index value is calculated for each of the plurality of types of feature regions.

  In the present embodiment, the calculation unit 44 calculates PSNR (Peak Signal-to-Noise Ratio) as an index value.

  Note that PSNR is calculated by the following equation, where 255 is the maximum pixel value that a pixel can take.

  The PSNR is calculated by substituting the MSE calculated by the equation (1) into the equation (2). Here, MSE calculated by Expression (1) represents a mean square error between the pixel value of the feature region and the pixel value at the same position as the feature region in the original image.

Np is the number of PSNR calculation target pixels.
O (i, j) is a pixel value of the original image.
L (i, j) is a pixel value of the decoded image.
i and j are the x and y coordinates in the PSNR calculation target rectangle.

  The calculation unit 44 outputs the calculated index value to the entropy encoding unit 34.

  Here, a specific example of the feature region extraction processing and the index value calculation processing will be described. In the present embodiment, it is assumed that the following five types of feature regions are extracted by the extraction unit 42, and an index value is calculated for each of the five types of extracted feature regions by the calculation unit 44.

  Note that although PSNR has been described as an example of the index value calculated by the calculation unit 44, the index value calculated by the calculation unit 44 is a value indicating the degree of deterioration of the feature region extracted from the decoded image. There is no particular limitation. For example, a value relating to an error between the pixel value of the feature region extracted from the decoded image and the pixel value of the original image, for example, the MSE expressed by Equation (1) may be calculated as the index value. Further, an average value of errors between the pixel value of the feature region and the pixel value at the same position as the feature region in the original image may be calculated as the index value.

(1) Face region The extraction unit 42 performs face recognition processing on the decoded image, and extracts a region where a face exists from the decoded image. The face recognition process may be performed using a known technique, and the technique used is not particularly limited. For example, the techniques described in P. Viola and M. Jones, “Robust real-time object detection”, Proc. Of IEEE Workshop on Statistical and Computational Theories in Computer Vision, pp.1-25, July 2001. .

  FIG. 4 shows an example of the face area extraction result. In FIG. 4, two face regions are extracted. The extraction unit 42 outputs each pixel position and pixel value of each pixel in the face area to the calculation unit 44.

  The calculation unit 44 compares the pixel value of the face area in the decoded image extracted by the extraction unit 42 with the pixel value at the same position as the face area in the original image corresponding to the decoded image, and the index Calculate the value. For the face area, one index value is calculated for each frame image. In FIG. 4, two face areas are extracted. However, since the two areas are collectively treated as the face area, one index value is calculated here.

(2) Fixed Areas with Fixed Positions and Sizes The positions and sizes of N fixed areas to be extracted as feature areas are set in the encoding device 1 in advance. N is a natural number of 1 or more. This setting may be performed in advance by the user, or may be performed by the manufacturer of the encoding device 1 before shipment.

  Here, the fixed area is a rectangular area, but it may not be a rectangle. If the fixed area is not rectangular, the shape may be set. In the example shown in FIG. 5, N = 15, but the value of N is not limited. The size of the fixed area is arbitrary. In addition, at least one of the size and the shape of each fixed region may be different for each position or the same.

  The extraction unit 42 extracts N fixed regions set in advance as feature regions to be extracted from the decoded image, and outputs the pixel value and the pixel position of each pixel of the extracted fixed region to the calculation unit 44. To do.

  The calculation unit 44 compares the pixel values of the N fixed regions in the decoded image extracted by the extraction unit 42 with the pixel values at the same position as the fixed region in the original image corresponding to the decoded image. And calculate the index value. Here, an index value is calculated for each of the N fixed areas. Accordingly, N index values are calculated for one frame image for a preset region.

(3) Still Area and Moving Area The extraction unit 42 divides the decoded image into Ma rectangular areas (blocks) having a predetermined size, and calculates the size of the motion vector for each block. When the calculated motion vector is less than the threshold value S, the extraction unit 42 determines the block as a still region, and when the calculated magnitude of the motion vector is greater than or equal to the threshold value S, the extraction unit 42 The area is determined as a moving area.

  When a plurality of blocks are determined as still areas, the entire plurality of blocks determined as still areas are extracted as one still area. When a plurality of blocks are determined as a moving area, the entire plurality of blocks determined as the moving area are extracted as one moving area. FIG. 6 shows an example of a still region extraction result, and FIG. 7 shows an example of a motion region extraction result.

  The extraction unit 42 outputs each of the pixel position and the pixel value of the still region to the calculation unit 44 and outputs each of the pixel position and the pixel value of the moving region to the calculation unit 44.

  The calculation unit 44 compares the pixel value of the still region in the decoded image extracted by the extraction unit 42 with the pixel value at the same position as the still region in the original image corresponding to the decoded image, Calculate the value. Further, the calculation unit 44 compares the pixel value of the moving region in the decoded image extracted by the extracting unit 42 with the pixel value at the same position as the moving region in the original image corresponding to the decoded image. The index value is calculated.

  By the above process, the decoded image for one frame can be classified into two, that is, a moving area and a still area, and an index value is calculated in each of the moving area and the still area. That is, an index value for one moving area and one still area is obtained for each frame image.

  Note that a motion vector detected by the generation unit 10 may be used as a motion vector for determining whether the region is a moving region or a still region. When the size of the block divided by the extraction unit 42 is larger than the size when the motion vector is detected by the generation unit 10, the average value of the motion vectors detected by the generation unit 10 may be obtained. Further, instead of using the motion vector detected by the generation unit 10, the calculation may be performed by performing block matching with frames having different times for each block. In addition, there are no restrictions on the number of block divisions Ma and the threshold value S. For example, FIGS. 6 and 7 show examples in which Ma = 16 and S = 1.

(4) Divided area according to luminance When the accuracy is 8 bits per pixel, the luminance value ranges from 0 to 255. For example, as illustrated in the table of FIG. 8, the extraction unit 42 classifies the luminance value into one of a plurality of different luminance ranges (bands), and calculates an index value for each band. More specifically, the extraction unit 42 first divides the decoded image into Mb rectangular regions (blocks), and calculates the average value L of the luminance of the block for each divided block. Then, the extraction unit 42 determines which band the average luminance value L belongs to. For each band, the extraction unit 42 outputs the pixel position and pixel value of each pixel belonging to the band to the calculation unit 44.

  The calculation unit 44 calculates, for each band, the pixel value of the pixel in the decoded image extracted by the extraction unit 42 and the pixel value at the same position as the pixel of the decoded image in the original image corresponding to the decoded image. In comparison, an index value is calculated.

  With the above processing, all blocks when a decoded image for one frame is divided into Mb blocks are classified into any of the eight bands, and an index value is calculated for each band. That is, for the band region (divided region) according to the luminance, eight index values are calculated for one frame image.

  The block division number Mb is not limited. For example, Mb = 66 is given as shown in the left diagram of FIG. Further, the number of divisions of the band is not limited to 8, and may be any number.

  As described above, five types of feature areas, that is, a face area, a fixed area, a stationary area, a moving area, and a divided area corresponding to luminance, are extracted, and an index value is calculated for each.

  When all the index values are calculated by the calculation unit 44, the entropy encoding unit 34 embeds the calculated index value information in the encoding information.

  FIG. 9 shows an example data structure of encoded information. As described above, the entropy encoding unit 34 is calculated by the calculation unit 44 for each frame image with respect to encoded data obtained by entropy encoding the image information of the frame image and each feature region of the frame image. Encoding information including index value information is generated.

  The index value information is used in the image selection device 2 described later.

  In the present embodiment, the index value information is stored in an area in the encoded information that can be decoded by a decoding apparatus that does not use the index value (does not affect the decoding process). For example, the encoding method is H.264. In the case of H.264, the area of User data unregistered SEI message corresponds. Thereby, in the decoding device that does not use the index value, the decoding process is performed by skipping the index value information from the encoded information.

  In addition, the index value information is not embedded in the encoded information, but is saved separately in an arbitrary format (however, the encoded information including the encoded data and the index value correspond to each other), and image selection to be described later The device 2 may be provided.

  Although an example in which the index value is calculated for each of the five types of feature regions has been described here, the present invention is not limited to this example, and may be calculated for at least one type of feature region.

  In addition, each index value may be included in the encoded information in association with the position and size information of the feature region corresponding to the index value.

  Although an example in which the encoding device 1 is realized by an electronic circuit or the like has been described here, it can also be realized by, for example, a computer 50 shown in FIG. The computer 50 includes a CPU (Central Processing Unit) 51, a nonvolatile program storage unit 52 storing a program, a RAM (random access memory) 53, an external IF (Interface) 54, and an image input IF 55. The CPU 51, program storage unit 52, RAM 53, external IF 54, and image input IF 55 are connected to one another via a bus 58.

  The program storage unit 52 stores an encoding program including a process for generating encoded data by encoding each frame image of a moving image, a process for extracting a feature region, and a process for calculating an index value.

  The CPU 51 reads out the encoded program from the program storage unit 52, expands it in the RAM 53, and executes each process of the encoded program. When the CPU 51 executes each process, the functions of the generation unit 10, the extraction unit 42, and the calculation unit 44 illustrated in FIG. 2 are realized.

  The RAM 53 is provided with a storage area corresponding to the storage unit 40.

  The external IF 54 is an interface between the recording medium 56 and the operation panel controller 57. As the recording medium 56, for example, a flash memory connectable via a CD-ROM, a DVD-ROM, or a USB (Universal Serial Bus) can be applied, and the generated encoded information and the like can be stored. The operation panel controller 57 controls an operation panel provided with a display unit including an LCD (liquid crystal display) or the like and a reception unit that includes a numeric keypad, a touch panel, and the like and receives an operation by a user.

  The image input IF 55 is an interface for inputting image information of a moving image. For example, an interface that can be connected to an imaging device that captures moving images may be used, or an interface that can acquire image information from another device via a communication network such as a wide area network. Good.

  FIG. 11 is a flowchart showing the flow of processing of the encoding program.

  In step 100, the generation unit 10 obtains a frame image (original image) to be encoded.

  In step 102, the generation unit 10 generates encoded data obtained by encoding the acquired original image.

  In step 104, the generation unit 10 generates a decoded image (reference image) and stores it in the storage unit 40.

  In step 106, the extraction unit 42 extracts at least one type of feature region.

  In step 108, the calculation unit 44 calculates an index value for each of the extracted feature regions.

  In step 110, the generation unit 10 generates encoded information including the calculated index value.

  As described above, the CPU 51 reads out and executes the program stored in the storage unit such as the program storage unit 52, thereby realizing the function of the encoding device 1 by cooperating the hardware and the program. it can.

  Next, based on the index value information included in the encoded information generated by the encoding apparatus 1, an image for performing processing of selecting and presenting a decoded image with a low degree of image quality degradation from a plurality of decoded images The selection device 2 will be described with a specific example.

  FIG. 12 shows an image selection device 2 according to this embodiment. The image selection device 2 includes a decoding unit 71, a first storage unit 80, a second storage unit 81, a selection unit 82, a reception unit 83, and a presentation unit 84.

  Each functional unit of the image selection device 2 can be realized by, for example, an electronic circuit, and the generation unit 10 may be realized by, for example, a semiconductor integrated circuit, more specifically, an ASIC (Application Specific Integrated Circuit).

  The decoding unit 71 decodes encoded data included in the encoded information and generates image information of a decoded image. The image information of the decoded image is stored in the first storage unit 80. In addition, the decoding unit 71 extracts an index value included in the encoded information and stores it in the second storage unit 81.

  FIG. 13 shows the configuration of the decoding unit 71 according to the present embodiment. The decoding unit 71 includes an entropy decoding unit 72, an inverse quantization unit 73, an inverse transformation unit 74, a decoded image generation unit 75, a predicted image generation unit 76, and a switching unit 79.

  The entropy decoding unit 72 performs entropy decoding (variable length decoding) on the encoded data (bit stream) included in the encoding information generated by the encoding device 1 to obtain the quantization coefficient of the prediction error image and the side information. Output. Entropy decoding refers to processing for returning a variable-length code to the original signal in accordance with the appearance frequency of symbols. The decoded quantization coefficient is supplied to the inverse quantization unit 73, and the decoded side information is supplied to the predicted image generation unit 76.

  Furthermore, the entropy decoding unit 72 takes out index value information included in the encoded information and stores it in the second storage unit 81.

  The inverse quantization unit 73 performs inverse quantization (scaling) on the quantization coefficient decoded by the entropy decoding unit 72 and outputs a reproduction frequency coefficient. The inverse quantization unit 73 generates a reproduction frequency coefficient obtained by returning the quantized coefficient of the compressed frequency to the original scale by scaling, and supplies the reproduction frequency coefficient to the inverse transform unit 74.

  The inverse transform unit 74 performs an inverse transform process on the reproduction frequency coefficient supplied from the inverse quantization unit 73. The inverse conversion unit 74 returns the frequency component of the reproduction frequency coefficient to the original prediction error image (reproduction prediction error image) by inverse conversion processing. Then, the reproduction prediction error image is supplied to the decoded image generation unit 75.

  The decoded image generation unit 75 adds the prediction image generated by the prediction image generation unit 76 and the reproduction prediction error image generated by the inverse quantization unit 73 and the inverse transformation unit 74. The image information of the decoded image added and decoded (reproduced) is accumulated in the first accumulation unit 80. The accumulated decoded image is supplied to an inter-frame prediction image generation unit 78, which will be described later, and is referred to in motion compensation when other frame images are decoded. Therefore, among the accumulated decoded images, the decoded image that is referred to when the predicted image is generated by the inter-frame predicted image generation unit 78 is referred to as a “reference image”. Which decoded image is referred to depends on the prediction mode of the decoding target frame image.

  The predicted image generation unit 76 includes an intra-frame predicted image generation unit 77 and an inter-frame predicted image generation unit 78. The predicted image generation unit 76 generates a predicted image based on the side information supplied from the entropy decoding unit 72 and, if necessary, the decoded image, and supplies the predicted image to the decoded image generation unit 75 via the switching unit 79.

  Based on the side information, the intra-frame prediction image generation unit 77 generates a prediction image for decoding the encoded data encoded by the intra-frame prediction encoding method in units of blocks.

  The inter-frame prediction image generation unit 78 generates a prediction image when decoding encoded data encoded by the inter-frame prediction encoding method, in units of blocks, based on the side information and the decoded image. Specifically, the predicted image is generated with reference to the pixel value in the decoded image indicated by the motion vector information included in the side information.

  The switching unit 79 selects one of the prediction image generated by the intra-frame prediction image generation unit 77 and the prediction image generated by the prediction image generation unit 76 and supplies the selected image to the decoded image generation unit 75. The switching unit 79 is controlled and switched by a control unit (not shown).

  The image information of the decoded image stored in the first storage unit 80 is input to the presentation unit 84. The presentation unit 84 has a function of extracting a plurality of frame images (decoded images stored in the first storage unit 80) continuous on the time axis from the moving image and displaying them on the display. For example, the presentation unit 84 generates a reduced image (hereinafter referred to as a thumbnail image) obtained by reducing the decoded image, and displays the reduced image by incorporating it into the selection screen for the decoded image.

  FIG. 14 shows an example of the selection screen 300.

  An area 303 is a thumbnail image display area. The user can designate a thumbnail image of a decoded image (hereinafter referred to as a reference image) that serves as a selection criterion for the decoded image from a plurality of thumbnail images displayed in the area 303. The specification result of the reference image is received by the receiving unit 83 and input to the selection unit 82. In the area 303, the selection result of the decoded image selected by the selection unit 82 of the image selection device 2 is displayed. Thereby, the selection result is presented to the user.

  An area 301 is an area for designating the type of feature area that the user has priority. When the user selects a button shown by operating an operation unit (not shown), a feature region shown on the right side of the selected button is designated. The result of specifying the feature area is received by the receiving unit 83 and input to the selection unit 82. In this embodiment, five types of feature areas, that is, a face area, N fixed areas (hereinafter referred to as N points), a moving area, a stationary area, and a divided area corresponding to luminance (hereinafter referred to as luminance distribution). Let the user select (specify) one of them.

  In the area 302, an image obtained by superimposing the position image indicating the position of the feature region designated in the area 301 on the thumbnail image of the reference image is displayed. Furthermore, when the user designates N points in the area 301 or the feature region of the luminance distribution, the user can perform more detailed designation in the area 302. Here, the designation result specifying the details of the feature region is received by the receiving unit 83 and input to the selection unit 82.

  For example, as described above, when the position image is stored in the encoded information in association with each index value and the position and size information of the feature region corresponding to the index value, It can be superimposed on the reference image based on the position and size of the feature region. Further, the feature region may be extracted each time by image processing, and the position image may be superimposed on the extracted position.

  The selection unit 82 is supplied with the image information of the decoded image accumulated in the first accumulation unit 80, the index value information accumulated in the second accumulation unit 81, and the designation result received by the reception unit 83. Based on the designation result accepted by the accepting unit 83 and the index value information accumulated in the second accumulating unit 81, the selecting unit 82 deteriorates the image quality from the decoded image accumulated in the first accumulating unit 80. The image having the lowest degree is selected, and the selection result is output to the presentation unit 84.

  Specifically, the degree of deterioration in image quality indicated by the index values of the feature areas designated by the user from the area 301 and the area 302 from a plurality of decoded images within a predetermined range including the reference image designated by the user from the area 303 The decoded image with the lowest is selected.

  As a specific example of the decoded image within a predetermined range including the reference image, for example, the reference image and the decoded image for n frames before and after the reference image (n is an integer of 1 or more) may be used. Also, a plurality of decoded images from a frame of the reference image to a frame after s seconds of the reference image may be used. The selection range may be set in advance or may be set in advance by the user.

  When PSNR is used as the index value, the selection unit 82 selects a decoded image that maximizes the PSNR of the feature region designated by the user.

  The presentation unit 84 presents the decoded image indicated by the selection result output from the selection unit 82 to the user. For example, the thumbnail image displayed in the area 303 of the selection screen 300 may be presented by making the display format of the thumbnail image of the decoded image selected by the selection unit 82 different from other thumbnail images.

  Hereinafter, a specific example of the operation of the image selection device 2 of the present embodiment will be described.

  First, the decoding unit 71 decodes encoded data included in the encoded information, and generates a decoded image. Furthermore, the presentation unit 84 displays the selection screen 300 on the display. At this time, for example, a plurality of decoded images included in a predetermined time interval are extracted from the start point in the area 303 of the selection screen, starting from the first decoded image, as a start point. A thumbnail image of the decoded image may be displayed on the display.

  Furthermore, when a plurality of thumbnail images are displayed as described above, for example, when the user performs a predetermined operation such as when the user presses the “forward button” by an operation unit (not shown), the presentation unit 84 may switch the display of thumbnail images.

  More specifically, for example, a thumbnail image of the decoded image immediately after the decoded image represented by the last displayed thumbnail image among the currently displayed thumbnail images is generated and additionally displayed in the area 303. . Then, the thumbnail image of the decoded image displayed at the head of the displayed thumbnail images is deleted from the area 303.

  Alternatively, a decoded image that is displayed last among the displayed thumbnail images, or a decoded image that is one after the decoded image, is used as a starting point, and a plurality of decoded images that are included between the starting point and a predetermined time interval later The display may be switched to the thumbnail image.

  Further, for example, when a predetermined operation such as when the user presses the “return button” is performed, display switching opposite to the above can be performed. That is, a thumbnail image of the decoded image immediately before the decoded image represented by the thumbnail image displayed at the top of the currently displayed thumbnail images is generated and additionally displayed in the area 303. Then, the thumbnail image of the decoded image displayed last among the displayed thumbnail images is deleted from the area 303.

  Alternatively, a decoded image displayed at the head of the thumbnail images being displayed, or a decoded image immediately before the decoded image, and a plurality of decoded images included before the predetermined time interval from the end point The display may be switched to the thumbnail image.

  Thereby, the user can display a desired scene in the area 303 of the selection screen 300.

  Furthermore, the user can designate a reference image as a reference for selection from thumbnail images of a desired scene while operating the operation unit to change the display of the area 303.

  The user designates the type of feature area from the area 301 on the selection screen 300. The designation result is accepted by the accepting unit 83 and output to the selecting unit 82. FIG. 15 shows a display example of the selection screen when the user designates a face area in the area 301. In the area 303, thumbnail images corresponding to a plurality of decoded images are displayed. A thumbnail image A clearly shown in an area 303 in FIG. 15 is a thumbnail image of a decoded image selected as a reference image by the user.

  The presentation unit 84 displays in the area 302 an image obtained by superimposing the position image indicating the position of the feature region designated in the area 301 on the thumbnail image of the reference image. In the example shown in FIG. 15, a rectangular frame image is superimposed on the position of the face area.

  When the designation result designating the face area from the selection screen 300 is input from the receiving unit 83, the selection unit 82 converts the second storage unit 81 into a decoded image within a predetermined range including the reference image designated by the user. The index value of the corresponding face area is acquired. The selection unit 82 selects a decoded image having the lowest degree of deterioration in image quality indicated by the acquired index value of the face area from the plurality of decoded images within the predetermined range. For example, when the index value is PSNR, it is determined that the degree of deterioration is the lowest if the PSNR value of the face area is the maximum among the plurality of decoded images within the predetermined range. When the index value is MSE, if the MSE value of the face area is the smallest among the plurality of decoded images within the predetermined range, it is determined that the degree of deterioration is the lowest. The selection result is output to the presentation unit 84.

  Upon receiving the selection result, the presentation unit 84 presents the decoded image selected by the selection unit 82. For example, the thumbnail image of the selected decoded image is displayed together with the thumbnail image of the decoded image that is temporally continuous in the area 303, and the display state of the thumbnail image corresponding to the selected decoded image is displayed as another thumbnail image. You may make it differ. In FIG. 15, a thumbnail image B is a thumbnail image of a decoded image selected by the selection unit 82, and a double-line frame image is superimposed on the thumbnail image B.

  Next, FIG. 16 shows a display example of the selection screen when the user designates N feature images. The thumbnail image A displayed in the area 303 of the selection screen 300 is a thumbnail image of a reference image designated as a selection reference by the user. The presentation unit 84 displays the thumbnail image A of the reference image in the area 302, and allows the user to further specify an area from the N fixed areas in the screen. For this selection, a plurality of N fixed areas may be selected. When the user designates a fixed area, the selection result is input to the selection unit 82 via the reception unit 83. The selection unit 82 acquires an index value of the designated fixed area corresponding to a decoded image within a predetermined range including the reference image designated by the user. The selection unit 82 selects a decoded image having the lowest degree of image quality degradation indicated by the acquired index value from a plurality of decoded images within the predetermined range. For example, when the index value is PSNR, it is determined that the degree of deterioration is the lowest if the PSNR value of the fixed area designated by the user is the maximum among the plurality of decoded images within the predetermined range. . When the index value is MSE, if the MSE value of the fixed area designated by the user is the smallest among the plurality of decoded images within the predetermined range, it is determined that the degree of deterioration is the lowest. . When there are a plurality of fixed areas designated by the user, an average value of a plurality of index values may be used, or a maximum value or a minimum value of a plurality of index values may be selected as a representative value. May be. The selection result of the selection unit 82 is output to the presentation unit 84.

  Upon receiving the selection result, the presentation unit 84 presents the decoded image selected by the selection unit 82. In FIG. 16, a thumbnail image B is a thumbnail image of a decoded image selected by the selection unit 82, and a double-line frame image is superimposed on the thumbnail image B.

  Next, FIG. 17 shows a display example of the selection screen when the user designates a characteristic image of the luminance distribution. The thumbnail image A displayed in the area 303 of the selection screen 300 is a thumbnail image of a reference image designated as a selection reference by the user. The presentation unit 84 displays the thumbnail image A of the reference image in the area 302 and allows the user to specify an arbitrary position in the screen. When the selection unit 82 receives the information on the designated position via the reception unit 83, the selection unit 82 calculates the average luminance value L of the rectangular area (block) corresponding to the designated position, and to which band the average luminance value L is. Determine if it belongs. The determination result is output to the presentation unit 84, and the presentation unit 84 displays the determination result in the area 302. For example, when the user designates a mountain portion, the band 2 corresponding to the designated position is designated (see also the area 302 in FIG. 17).

  In addition, the selection unit 82 acquires, from the second storage unit 81, the index value of the band specified by the user among the index values of the decoded image within the predetermined range including the reference image specified by the user. The selection unit 82 selects a decoded image with the lowest degree of deterioration in image quality indicated by the obtained index value of the band from the plurality of decoded images within the predetermined range. For example, when the index value is PSNR, it is determined that the degree of deterioration is the lowest if the PSNR value of the band designated by the user is the maximum among the plurality of decoded images within the predetermined range. When the index value is MSE, if the MSE value of the band designated by the user among the plurality of decoded images within the predetermined range is the smallest, it is determined that the degree of deterioration is the lowest. The selection result is output to the presentation unit 84.

  Although an example in which a band is specified by specifying a position in an image has been described here, the user may directly specify a band. There may be a plurality of bands to be specified. In this case, a decoded image is selected using the average value of the index values of each band.

  Upon receiving the selection result, the presentation unit 84 presents the decoded image selected by the selection unit 82. In FIG. 17, a thumbnail image B is a thumbnail image of a decoded image selected by the selection unit 82, and a double-line frame image is superimposed on the thumbnail image B.

  In addition, although illustration is abbreviate | omitted here, when a user designates a moving area | region or a still area | region, it can select and show a decoding image similarly to the above using the index value of the area | region.

  In the above description, an example in which the image selection device 2 is realized by an electronic circuit or the like has been described. The computer 60 includes a CPU (Central Processing Unit) 61, a nonvolatile program storage unit 62 storing a program, a RAM (random access memory) 63, an image processing unit 64, an external IF (Interface) 66, and an image output IF 67. Yes. The CPU 61, the program storage unit 62, the RAM 63, the image processing unit 64, the external IF 66, and the image output unit IF 67 are connected to each other via a bus 70.

  The program storage unit 62 includes an image including a selection process for selecting an image with a low degree of deterioration in image quality from a plurality of decoded images, a reception process for receiving a user designation, and a process for presenting the selected decoded image. A selection program is stored. The image selection program may include a decoding process for decoding encoded data obtained by encoding each frame image of the moving image, or the decoding program is stored as an independent program different from the image selection program. It may be.

  The CPU 61 controls the entire image selection device 2. Further, the CPU 61 reads out the image selection program from the program storage unit 62 and develops it in the RAM 63, and executes each process of the image selection program. When the CPU 61 executes each process, the functions of the decryption unit 71, the selection unit 82, the reception unit 83, and the presentation unit 84 illustrated in FIG. 12 are realized.

  The RAM 63 is provided with storage areas corresponding to the first accumulation unit 80 and the second accumulation unit 81. Also provided is a buffer area for buffering the encoded information generated by the encoding device 1 and the image information of the selection screen 300.

  The image processing unit 64 performs various types of image processing such as generating image information of thumbnail images by reducing the image information of the decoded image and generating image information for displaying the selection screen 300.

  The external IF 66 is an interface between the recording medium 68 and the operation panel controller 69. The recording medium 68 can be, for example, a CD-RW or a flash memory that can be connected via a USB (Universal Serial Bus). When the encoded information generated by the encoding device 1 is stored in the recording medium 68, the encoded information can be acquired from the recording medium 68 via the external IF 66. The operation panel controller 69 controls an operation panel provided with a display unit including an LCD (liquid crystal display) and the like and a reception unit configured with a numeric keypad, a touch panel, and the like to receive an operation by a user.

  The image output IF 67 is an interface that is connected to the display and outputs image information of the decoded image read from the RAM 63 and image information of the selection screen 300 (including image information of the thumbnail image) to the display. Thereby, the decoded image and the thumbnail image are displayed on the display.

  FIG. 19 is a flowchart showing the flow of processing of the image selection program.

  In step 200, the decoding unit 71 reads the encoded information from the recording medium 68 and buffers it in the RAM 63.

  In step 202, the decoding unit 71 reads the encoded information from the RAM 63, and decodes the encoded data included in the encoded information. The image information of the decoded image is stored in a recording area corresponding to the first accumulation unit 80 of the RAM 63.

  In step 204, the decoding unit 71 reads the encoded information from the RAM 63, extracts an index value included in the encoded information, and stores it in a storage area corresponding to the second accumulation unit 81 of the RAM 63.

  In step 206, the presentation unit 84 controls the image processing unit 64 to cause the image processing unit 64 to perform a reduction process on the image information of the decoded image and generate image information of the thumbnail image. The image information of the thumbnail image is stored in the RAM 63.

  In step 208, the presentation unit 84 combines the image information of the thumbnail image with the image information of the selection screen 300, and displays the final display image on the display.

  In step 210, the accepting unit 83 accepts designation of the type of the reference image and the feature area from the user.

  In step 212, the selection unit 82 uses the index value of the feature region designated by the user from a plurality of decoded images within a predetermined range including the received reference image, and the image quality degradation indicated by the index value is determined. The decoded image with the lowest degree is selected.

  In step 214, the presentation unit 84 presents the selection result.

  In addition, although the example which implement | achieves the function of the decoding part 71 when CPU61 runs a program was demonstrated here, it is not limited to the example. For example, the decoding unit 71 may be realized by hardware such as an electronic circuit, and the functions of the selection unit 82, the reception unit 83, and the presentation unit 84 may be realized by the CPU 61 executing a program. A hardware configuration diagram in this case is shown in FIG. 20, the same parts as those in FIGS. 12 and 18 are denoted by the same reference numerals. As shown in FIG. 20, by connecting the decoding unit 71 to a bus 70, the function of the decoding unit 71 is realized by hardware processing instead of software processing. Such a configuration also acts in the same manner as described above. Alternatively, the image selection device 2 may not be provided with the function of the decoding unit 71 but may be decoded by another device. The image selection device 2 may acquire the image information and index value of the decoded image decoded from another device, and perform the image selection process.

  In the above description, an example in which a user designates a reference image and selects a decoded image with a low degree of image quality degradation from decoded images within a predetermined range including the reference image has been described. However, the present invention is not limited to this example. For example, you may make it select from the decoded image corresponding to the thumbnail image currently displayed on the display (For example, it is contained in the area 303 of the selection screen 300 displayed on the display above.). In this case, the user does not have to specify the reference image.

  Here, an example has been described in which each of the index values of a plurality of types of feature areas is included in the encoded information. However, when the index values of one type of feature areas are included in the encoded information. In this case, the specification of the feature region type in the area 301 is omitted.

  Also, the specification of the type of the feature area by the user in the area 301 and the area 302 and the default value for the detailed specification may be set in the image selection device 2 in advance. In this case, designation of the user can be omitted, and reception of designation from the user by the reception unit 83 is omitted.

  In the above description, an example has been described in which the selection unit 82 selects a decoded image with the lowest degree of image quality degradation from a plurality of decoded images within a predetermined range including a reference image designated by the user. It is not limited to. For example, the user may specify a selection range instead of a reference image, and the selection unit 82 may select a decoded image with the lowest degree of image quality degradation from a plurality of decoded images in the selection range. Good.

  In the above description, an example in which a decoded image having the lowest degree of image quality degradation is selected has been described, but the present invention is not limited to this example. For example, not only the decoded image with the lowest degree of image quality degradation but also the top P decoded images (P is an integer of 2 or more) may be selected in the order of the lowest degree of degradation. In this case, the selected decoded images can be presented so that the rank order of the degree of deterioration is clearly indicated. For example, the higher the ranking, the more conspicuous it is displayed, and the numerical value indicating the ranking is displayed.

  As described above, according to the present embodiment, a decoded image with a low degree of image quality degradation can be selected and presented to the user based on the index value of the feature region. As a result, the user can easily grasp a decoded image with good image quality in the feature region.

  Furthermore, as shown in FIG. 21, the functions of the encoding device 1 and the image selection device 2 can be realized by a single computer 90 to provide the image selection system 3. For example, the aspect which provides each function of the encoding apparatus 1 and the image selection apparatus 2 in the portable terminal provided with the imaging function and the display is mentioned.

  The computer 90 includes a CPU (Central Processing Unit) 91, a nonvolatile program storage unit 92 that stores programs, a RAM (random access memory) 93, an image processing unit 94, an external IF (Interface) 95, an image input IF 96, and an image output IF 97. It has. The CPU 91, program storage unit 92, RAM 93, image processing unit 94, external IF 95, image input IF 96, and image output IF 97 are connected to each other via a bus 98.

  The program storage unit 92 stores an image encoding selection program configured by combining an encoding program and an image selection program.

  The CPU 91 reads out the image encoding selection program from the program storage unit 92, expands it in the RAM 93, and executes each process of the encoding program and the image selection program. When the CPU 91 executes each process, the functions of the generation unit 10, the extraction unit 42, the calculation unit 44, the selection unit 82, the reception unit 83, and the presentation unit 84 illustrated in FIGS. 2 and 12 are realized.

  The RAM 93 is provided with storage areas corresponding to the storage unit 40, the first storage unit 80, and the second storage unit 81.

  The external IF 95 is an interface between the recording medium 99 and the operation panel controller 89. The image input IF 96 is an interface for inputting image information of a moving image. The image output IF 97 is connected to the display and is an interface for outputting the image information of the decoded image read from the RAM 93 and the image information of the selection screen 300 to the display.

  Such a configuration also acts in the same manner as described above.

  In the above description, the encoding program is stored in the program storage units 52 and 92. However, the encoding program is a portable recording medium such as a CD-ROM or a flash memory that can be connected via USB. It is also possible to provide it in a form recorded on a medium. Further, although the mode in which the image selection program is stored in the program storage units 62 and 92 has been described, the image data storage program is a portable recording medium such as a CD-ROM or a flash memory connectable via USB. It is also possible to provide it in the form recorded in

  All documents, patent applications and technical standards mentioned in this specification are to the same extent as if each individual document, patent application and technical standard were specifically and individually stated to be incorporated by reference. Incorporated by reference in the book.

  Regarding the above embodiment, the following additional notes are disclosed.

(Appendix 1)
A plurality of pieces of encoded data obtained by encoding a plurality of frame images included in a moving image are generated, and at least one index value indicating the degree of image quality degradation caused by the encoding is set in advance in each frame image. Calculate for different types of feature areas,
Image selection including selecting a decoded image having the lowest degree of degradation indicated by the index value from a plurality of decoded images decoded from the plurality of generated encoded data, and presenting the selected decoded image Method.

(Appendix 2)
In calculating the index value, the index value is calculated for each of a plurality of preset feature areas in each frame image,
When selecting the decoded image, an index value corresponding to a feature region designated in advance by a user among the plurality of index values is selected from the plurality of decoded images decoded from the plurality of encoded data. The image selection method according to attachment 1, wherein a decoded image having the lowest degree of degradation is selected.

(Appendix 3)
When the decoded image is selected, the index value indicates the degradation indicated by the decoded image within the range specified in advance by the user or the decoded image within the predetermined range including the decoded image specified by the user in advance. The image selection method according to Supplementary Note 1 or Supplementary Note 2, wherein a decoded image having the lowest degree is selected.

(Appendix 4)
The preset feature area includes a face area, a fixed area with a predetermined position and size, a still area with a motion vector less than a threshold, a motion area with a motion vector greater than a threshold, a divided area divided according to luminance, The image selection method according to any one of supplementary notes 1 to 3, including at least one of the above.

(Appendix 5)
When selecting the decoded image, the top N decoded images (N is an integer equal to or greater than 2) are selected in the order of the degree of deterioration. Image selection method according to any one of appendix 1 to appendix 4 .

(Appendix 6)
A plurality of encoded data generated by encoding a plurality of frame images included in a moving image, and at the time of the encoding, the code for at least one type of feature region set in advance in each frame image A decoded image having the lowest degree of degradation indicated by the index value is selected from a plurality of decoded images decoded from a plurality of pieces of encoded data for which an index value indicating the degree of image quality degradation caused by the conversion is calculated. A selection section;
A presentation unit for presenting the decoded image selected by the selection unit;
An image selection device comprising:

(Appendix 7)
For each of the encoded data, the index value is calculated for each of a plurality of types of feature regions set in advance in each frame image,
The selection unit has a degree of deterioration indicated by an index value corresponding to a feature region designated in advance by a user among the plurality of index values from among a plurality of decoded images decoded from the plurality of encoded data. The image selection device according to appendix 6, wherein the lowest decoded image is selected.

(Appendix 8)
The selection unit has the lowest degree of deterioration indicated by the index value from among a decoded image within a range specified in advance by a user or a decoded image within a predetermined range including a decoded image specified in advance by a user. The image selection device according to appendix 6 or appendix 7, which selects a decoded image.

(Appendix 9)
The preset feature area includes a face area, a fixed area with a predetermined position and size, a still area with a motion vector less than a threshold, a motion area with a motion vector greater than a threshold, a divided area divided according to luminance, The image selection device according to any one of appendix 6 to appendix 8, including at least one of the above.

(Appendix 10)
The image selection device according to any one of appendix 6 to appendix 9, wherein the selection unit selects the top N decoded images (N is an integer equal to or greater than 2) in descending order of the degree of deterioration.

(Appendix 11)
A generating unit that generates a plurality of encoded data obtained by encoding a plurality of frame images included in a moving image;
A calculation unit that calculates an index value indicating a degree of image quality degradation caused by the encoding for at least one type of feature region set in advance in each frame image;
An encoding device comprising:

(Appendix 12)
The encoding device according to claim 11, wherein the calculation unit calculates the index value for each of a plurality of types of feature regions set in advance in the frame images.

(Appendix 13)
The preset feature area includes a face area, a fixed area with a predetermined position and size, a still area with a motion vector less than a threshold, a motion area with a motion vector greater than a threshold, a divided area divided according to luminance, The encoding device according to appendix 11 or appendix 12, including at least one of the above.

(Appendix 14)
A generating unit that generates a plurality of encoded data obtained by encoding a plurality of frame images included in a moving image, and an index value indicating a degree of image quality degradation caused by the encoding is set at least in advance in each frame image An encoding device including a calculation unit for calculating one type of feature region;
A selection unit that selects a decoded image with the lowest degree of degradation indicated by the index value from among a plurality of decoded images decoded from a plurality of encoded data generated by the encoding device, and selected by the selection unit An image selection device comprising a presentation unit for presenting the decoded image,
An image selection system.

(Appendix 15)
The calculation unit calculates the index value for each of a plurality of preset feature areas in each frame image,
The selection unit has a degree of deterioration indicated by an index value corresponding to a feature region designated in advance by a user among the plurality of index values from among a plurality of decoded images decoded from the plurality of encoded data. The image selection system according to appendix 14, wherein the lowest decoded image is selected.

(Appendix 16)
The selection unit has the lowest degree of deterioration indicated by the index value from among a decoded image within a range specified in advance by a user or a decoded image within a predetermined range including a decoded image specified in advance by a user. The image selection system according to appendix 14 or appendix 15, wherein the decoded image is selected.

(Appendix 17)
The preset feature area includes a face area, a fixed area with a predetermined position and size, a still area with a motion vector less than a threshold, a motion area with a motion vector greater than a threshold, a divided area divided according to luminance, The image selection system according to any one of supplementary notes 14 to 16, including at least one of the above.

(Appendix 18)
The image selection system according to any one of appendix 14 to appendix 17, wherein the selection unit selects the top N decoded images (N is an integer equal to or greater than 2) in descending order of the degree of deterioration.

(Appendix 19)
On the computer,
Encoded data generated by encoding a plurality of frame images included in a moving image, and at the time of encoding, at least one type of feature region set in advance in each frame image is encoded From among a plurality of decoded images decoded from a plurality of encoded data for which an index value indicating the degree of image quality degradation that has occurred is selected, a decoded image having the lowest degree of degradation indicated by the index value is selected,
An image selection program for executing a process of presenting the selected decoded image or a recording medium on which an image selection program is recorded.

(Appendix 20)
For each of the encoded data, the index value is calculated for each of a plurality of types of feature regions set in advance in each frame image,
When selecting the decoded image, an index value corresponding to a feature region designated in advance by a user among the plurality of index values is selected from the plurality of decoded images decoded from the plurality of encoded data. The image selection program according to appendix 19, or a recording medium on which the image selection program is recorded.

(Appendix 21)
When the decoded image is selected, the index value indicates the degradation indicated by the decoded image within the range specified in advance by the user or the decoded image within the predetermined range including the decoded image specified by the user in advance. A recording medium on which the image selection program or the image selection program according to attachment 19 or 20 is recorded.

(Appendix 22)
The preset feature area includes a face area, a fixed area with a predetermined position and size, a still area with a motion vector less than a threshold, a motion area with a motion vector greater than a threshold, a divided area divided according to luminance, An image selection program according to any one of Supplementary Note 19 to Supplementary Note 21, or a recording medium on which the image selection program is recorded.

(Appendix 23)
23. The image selection program according to any one of appendix 19 to appendix 22, wherein when selecting the decoded image, the top N decoded images (N is an integer equal to or greater than 2) are selected in descending order of the degree of deterioration. Or the recording medium which recorded the image selection program.

(Appendix 24)
On the computer,
Generating a plurality of encoded data obtained by encoding a plurality of frame images included in a moving image;
An encoding program or an encoding program for executing a process of calculating an index value indicating the degree of image quality degradation caused by the encoding for at least one feature region set in advance in each frame image is recorded. recoding media.

(Appendix 25)
25. The encoding program according to appendix 24, or a recording medium on which the encoding program is recorded, when calculating the index value, calculates the index value for each of a plurality of preset feature areas in each frame image.

(Appendix 26)
The preset feature area includes a face area, a fixed area with a predetermined position and size, a still area with a motion vector less than a threshold, a motion area with a motion vector greater than a threshold, a divided area divided according to luminance, An encoding program according to appendix 24 or appendix 25 or a recording medium on which the encoding program is recorded.

(Appendix 27)
On the computer,
Generating a plurality of encoded data obtained by encoding a plurality of frame images included in a moving image;
An index value indicating the degree of image quality degradation caused by the encoding is calculated for at least one type of feature region set in advance in each frame image;
From among a plurality of decoded images decoded from the generated plurality of encoded data, select a decoded image having the lowest degree of degradation indicated by the index value,
An image encoding selection program or a recording medium on which an image encoding selection program for executing the process of presenting the selected decoded image is recorded.

(Appendix 28)
In calculating the index value, the index value is calculated for each of a plurality of preset feature areas in each frame image,
When selecting the decoded image, an index value corresponding to a feature region designated in advance by a user among the plurality of index values is selected from the plurality of decoded images decoded from the plurality of encoded data. The image encoding selection program according to attachment 27 or a recording medium on which the image encoding selection program is recorded.

(Appendix 29)
When the decoded image is selected, the index value indicates the degradation indicated by the decoded image within the range specified in advance by the user or the decoded image within the predetermined range including the decoded image specified by the user in advance. A recording medium on which the image encoding selection program or the image encoding selection program according to Supplementary Note 27 or 28 is selected.

(Appendix 30)
The preset feature area includes a face area, a fixed area with a predetermined position and size, a still area with a motion vector less than a threshold, a motion area with a motion vector greater than a threshold, a divided area divided according to luminance, A recording medium on which the image encoding selection program or the image encoding selection program according to any one of appendix 27 to appendix 29 is recorded.

(Appendix 31)
31. When selecting the decoded image, the top N decoded images (N is an integer equal to or greater than 2) are selected in descending order of the degree of deterioration. Image coding according to any one of appendix 27 to appendix 30 A recording medium on which a selection program or an image encoding selection program is recorded.

DESCRIPTION OF SYMBOLS 1 Encoding apparatus 2 Image selection apparatus 3 Image selection system 10 Generation | occurrence | production part 40 Storage part 42 Extraction part 44 Calculation part 50, 60, 90 Computer 71 Decoding part 80 1st storage part 81 2nd storage part 82 Selection part 83 Reception part 84 Presenter

Claims (17)

A plurality of pieces of encoded data obtained by encoding a plurality of frame images included in a moving image are generated, and at least one index value indicating the degree of image quality degradation caused by the encoding is set in advance in each frame image. Calculate for different types of feature areas,
Image selection including selecting a decoded image having the lowest degree of degradation indicated by the index value from a plurality of decoded images decoded from the plurality of generated encoded data, and presenting the selected decoded image Method. In calculating the index value, the index value is calculated for each of a plurality of preset feature areas in each frame image,
When selecting the decoded image, an index value corresponding to a feature region designated in advance by a user among the plurality of index values is selected from the plurality of decoded images decoded from the plurality of encoded data. The image selection method according to claim 1, wherein a decoded image having the lowest degree of degradation is selected. When the decoded image is selected, the index value indicates the degradation indicated by the decoded image within the range specified in advance by the user or the decoded image within the predetermined range including the decoded image specified by the user in advance. The image selection method according to claim 1 or 2, wherein a decoded image having the lowest degree is selected. The preset feature area includes a face area, a fixed area with a predetermined position and size, a still area with a motion vector less than a threshold, a motion area with a motion vector greater than a threshold, a divided area divided according to luminance, The image selection method according to claim 1, comprising at least one of the above. The image according to any one of claims 1 to 4, wherein, when selecting the decoded image, the top N decoded images (N is an integer of 2 or more) are selected in descending order of the degree of deterioration. Selection method. A plurality of encoded data generated by encoding a plurality of frame images included in a moving image, and at the time of the encoding, the code for at least one type of feature region set in advance in each frame image A decoded image having the lowest degree of degradation indicated by the index value is selected from a plurality of decoded images decoded from a plurality of pieces of encoded data for which an index value indicating the degree of image quality degradation caused by the conversion is calculated. A selection section;
A presentation unit for presenting the decoded image selected by the selection unit;
An image selection device comprising: For each of the encoded data, the index value is calculated for each of a plurality of types of feature regions set in advance in each frame image,
The selection unit has a degree of deterioration indicated by an index value corresponding to a feature region designated in advance by a user among the plurality of index values from among a plurality of decoded images decoded from the plurality of encoded data. The image selection device according to claim 6, wherein the lowest decoded image is selected. The selection unit has the lowest degree of deterioration indicated by the index value from among a decoded image within a range specified in advance by a user or a decoded image within a predetermined range including a decoded image specified in advance by a user. The image selection device according to claim 6 or 7, wherein a decoded image is selected. The preset feature area includes a face area, a fixed area with a predetermined position and size, a still area with a motion vector less than a threshold, a motion area with a motion vector greater than a threshold, a divided area divided according to luminance, The image selection device according to any one of claims 6 to 8, comprising at least one of the above. The image selection device according to any one of claims 6 to 9, wherein the selection unit selects the top N decoded images (N is an integer equal to or greater than 2) in descending order of the degree of deterioration. A generating unit that generates a plurality of encoded data obtained by encoding a plurality of frame images included in a moving image;
A calculation unit that calculates an index value indicating a degree of image quality degradation caused by the encoding for at least one type of feature region set in advance in each frame image;
An encoding device comprising: The encoding device according to claim 11, wherein the calculation unit calculates the index value for each of a plurality of types of feature regions set in advance in the frame images. The preset feature area includes a face area, a fixed area with a predetermined position and size, a still area with a motion vector less than a threshold, a motion area with a motion vector greater than a threshold, a divided area divided according to luminance, The encoding device according to claim 11, comprising at least one of the encoding devices. A generating unit that generates a plurality of encoded data obtained by encoding a plurality of frame images included in a moving image, and an index value indicating a degree of image quality degradation caused by the encoding is set at least in advance in each frame image An encoding device including a calculation unit for calculating one type of feature region;
A selection unit that selects a decoded image with the lowest degree of degradation indicated by the index value from among a plurality of decoded images decoded from a plurality of encoded data generated by the encoding device, and selected by the selection unit An image selection device comprising a presentation unit for presenting the decoded image,
An image selection system. On the computer,
Encoded data generated by encoding a plurality of frame images included in a moving image, and at the time of encoding, at least one type of feature region set in advance in each frame image is encoded From among a plurality of decoded images decoded from a plurality of encoded data for which an index value indicating the degree of image quality degradation that has occurred is selected, a decoded image having the lowest degree of degradation indicated by the index value is selected,
An image selection program for executing a process of presenting the selected decoded image. On the computer,
Generating a plurality of encoded data obtained by encoding a plurality of frame images included in a moving image;
An encoding program for executing a process of calculating an index value indicating a degree of deterioration of image quality caused by the encoding for at least one feature region set in advance in each frame image. On the computer,
Generating a plurality of encoded data obtained by encoding a plurality of frame images included in a moving image;
An index value indicating the degree of image quality degradation caused by the encoding is calculated for at least one type of feature region set in advance in each frame image;
From among a plurality of decoded images decoded from the generated plurality of encoded data, select a decoded image having the lowest degree of degradation indicated by the index value,
An image encoding selection program for executing a process of presenting the selected decoded image.