JP2005159561A - Beautiful face camera - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a camera capable of photographing a beautiful face of distinct features with natural finish while obscuring wrinkles of skin by utilizing a focused image and a non-focused image photographed by a conventional focus technology, and capable of photographing a beautiful face simply without requiring any advanced knowledge. <P>SOLUTION: The beautiful face camera comprises a focus lens control section for controlling the position of a focus lens such that focused face image information and non-focused face image information can be acquired, a section for acquiring feature area image information from the focused face image information, a section for acquiring skin image information from the non-focused face image information, and a section for synthesizing the feature area image information and the skin image information to produce the image of face of a person. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a technique for generating an image of a human face by combining a focused image and a non-focused image.

Currently, various processes can be applied to photographs taken with the advance of digital photography. In particular, for resumes and matchmaking photos during job hunting, we try to make a good impression of job hunting and matching by reducing wrinkles and rough skin by image processing. It has become common. As a specific method of this image processing, first, there is a method of blurring the skin portion of the image by applying a soft focus (blurring filter) to the skin portion so that wrinkles and the like are not noticeable (Patent Document 1). Alternatively, secondly, there are a method of manually erasing individual wrinkles on the skin with a color and a method of sharpening each part such as eyes and nose to improve the standing of the eyes.
Japanese Patent Laid-Open No. 6-18959

  However, with the first technology above, the entire face, including the face area that you want to sharpen, such as the eyes and nose, is covered because the soft focus (blur) filter is applied to the whole, and the background etc. is blurred. There is a problem that the image becomes unnatural. In the second technology, it is also necessary to capture the captured image data on a personal computer and process individual wrinkles and parts using an application dedicated to image processing. There is a problem of requiring advanced knowledge.

  In order to solve the above problems, the present invention provides a focus lens control unit for controlling the position of a focus lens so that in-focus face image information and out-of-focus face image information can be acquired, and the in-focus face image information. A feature region image information acquisition unit that acquires feature region image information, a skin image information acquisition unit that acquires skin image information from non-focused face image information, and the feature region image information and skin image information are combined. And a facial camera having a synthesizing unit for generating an image of a human face.

  According to the present invention having the above-described configuration, the eyes and nose are sharp using the focused image and the non-focused image captured by the conventional focus technique, but the wrinkles of the skin are not clearly visible. It is possible to shoot beautiful face photos that have been blurred in a natural way. In addition, since the conventional focus technique is used, a beautiful face photo can be easily taken without a high level of knowledge if the camera incorporates the above-described configuration requirements.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. Note that the present invention is not limited to these embodiments, and can be implemented in various modes without departing from the spirit of the present invention. The first embodiment will mainly describe claim 1. In the second embodiment, claim 2 will be mainly described. In the third embodiment, claim 3 will be mainly described. In the fourth embodiment, claim 4 will be mainly described. In the fifth embodiment, claim 5 will be mainly described.

  Embodiment 1 (Concept of Embodiment 1) FIG. 1 is a diagram for explaining an example of a concept of a beauty camera of the present embodiment. (A) of this figure is an image in a state where the focus by the focus lens is in focus on the subject. As shown in this figure, in a well-focused image, the eyes and nose are clear, but wrinkles and the like are often clearly recognized. Also, (b) in the figure is an image in a state where the subject is not focused on the focus lens. As shown in this figure, in an unfocused image, there are few reflections such as wrinkles, but the standing of the eyes and nose is also ambiguous. Therefore, in the beauty camera of the present embodiment, both images are combined to generate an image in which the wrinkles and roughness are not noticeable while the eyes and nose are crisp as shown in FIG. Hereinafter, specific processing of composition will be described while explaining the configuration of the beauty camera of the present embodiment.

    (Configuration of Example 1) FIG. 2 is a diagram illustrating an example of functional blocks of the facial camera of the present example. As shown in this figure, the facial camera (0200) of the present embodiment includes a “focus lens control unit” (0201), a “feature area image information acquisition unit” (0202), and a “skin image information acquisition unit”. (0203) and “combining section” (0204).

  The “focus lens control unit” (0201) has a function of controlling the position of the focus lens so that in-focus face image information and non-focus face image information can be acquired. “In-focus face image information” refers to image information focused on a human face. For example, it is acquired according to the focus lens position determined by so-called hill-climbing control (a control method in which contrast information is acquired for each focus lens position and the focus lens position at the peak is the focus lens position focused on the subject). This is image information composed of pixel color, position information, and the like of the subject. “Unfocused face image information” refers to image information of the person's face that is out of focus from the focus of the focused face image information. For example, it is image information composed of pixel color, position information, and the like of a subject acquired in accordance with a focus lens position shifted by a predetermined distance from the focus lens position from which in-focus face image information was acquired. Of course, the image based on this out-of-focus face image information is finally used as a human face image in the beauty camera of the present embodiment, and therefore it is not necessary to be a complete out-of-focus image.

  The focus lens control unit is realized by a CPU, for example, and performs control such as movement for determining a focus position of the focus lens. The “focus lens” refers to a lens that moves its position in order to focus on a subject in the camera. Of course, the focus lens is not limited to a single lens, and may be a lens group that focuses by focusing a plurality of lenses. The “focus lens position” refers to the position of the focus lens in the photographing mechanism of the facial camera, and includes information indicated by numerical values such as the number of motor pulses and the number of rotations or the actual lens movement distance. The focus lens is moved under the control of the focus lens control unit, and the in-focus face image information at the in-focus position and the out-of-focus face image information with the focus shifted therefrom can be acquired.

  The “feature area image information acquisition unit” (0202) has a function of acquiring the characteristic area image information from the focused face image information acquired based on the focus lens control of the focus lens control unit (0201). “Characteristic region image information” refers to image information of a facial feature region, and includes, for example, image information such as eyes, nose, mouth, and ears. In order to extract each feature region from the focused face image information in the feature region image information acquisition unit, for example, a technique such as region extraction using a nonlinear transducer network may be used (see Reference Document “Double Threshold Value”). Image segmentation oscillator network model for phase detection by the method and its LSI circuit configuration ”Dr. Ando, Makoto Miyake, Takashi Morie, Makoto Nagata, Satoshi Iwata (Electronic Information and Communication Society Neurocomputing Research Group NC98-126)). This is a method of extracting each feature area by connecting each pixel with an element having a non-linear characteristic and extracting an area having near luminance by phase-synchronizing the non-linear vibrators. FIG. 3 is a diagram illustrating an example of feature area image information. In this way, for example, information constituting the image such as color and position information regarding each feature region extracted using region extraction by a nonlinear vibrator network is cut out and acquired individually. Alternatively, color information other than the extracted feature region in the focused face image information may be made transparent. In this case, the composition in the composition unit described later may be performed by superimposing the image based on the feature region image information on the skin image information like a layer.

  The “skin image information acquisition unit” (0203) has a function of acquiring skin image information from non-focused face image information acquired based on the focus lens control of the focus lens control unit (0201). “Image information for skin” refers to image information of facial skin. The skin image information acquisition unit obtains the skin image information by, for example, extracting a feature region using the above-described technique of region extraction by the nonlinear vibrator network, and the extracted feature region as shown in FIG. Is obtained by converting the color information into a flesh color or deleting it according to the size of the acquired characteristic area image information. Of course, when the skin color of the color information of the feature area is performed, it is preferable to use a color that matches the color information of the surrounding skin. In addition, when the feature area image of the focused face image information is overwritten on the image of the skin image information at the time of synthesis described later, the feature areas such as eyes, nose, and mouth in the skin image are deleted as described above. It may be left as it is without being colored.

  The “synthesis unit” (0204) combines the feature region image information acquired by the feature region image information acquisition unit (0202) and the skin image information acquired by the skin image information acquisition unit (0203), thereby It has a function to make a face image. FIG. 5 is a diagram for explaining an example of synthesis by the synthesis unit. As shown in this figure, by combining the feature area image information and the skin image information, a beautiful face photograph that has sharp eyes and nose standing, but on the other hand is not clearly visible such as wrinkles on the skin, It can be done with perfect results.

  The processing at the time of combination may be superimposed on the image indicated by the skin image information like a layer if the feature region image information is transparent except for the feature region portion. In addition, if the image information is obtained by individually cutting out the feature area, it is preferable to determine and synthesize a position to be combined with the skin image from position information in the original focused face image of the feature area. This is because the in-focus face image and the out-of-focus face image have almost the same basic composition and size, and therefore the absolute position of the feature region hardly changes.

  Also, for backgrounds other than face images, if you want a focused (focused) background, the background image indicated by the focused face image information is reversed. It is preferable to use the background image indicated by the dark face image information. In the above example, eye, nose, mouth, and ear image information is used as the feature area image information. However, a face outline or the like may be acquired as the feature area image information. By doing so, it is possible to take a facial photograph (a photograph taken by the facial camera of the present invention) with a clear boundary with the background.

    (Processing Flow of First Embodiment) FIG. 6 is a flowchart showing an example of the processing flow of the present embodiment. Note that the processing flow shown below can be implemented as a method, a program for causing a computer to execute, or a readable recording medium on which the program is recorded (hereinafter, the processing flow in this specification is also described). The same).

  As shown in this figure, first, the focus lens is driven to determine the focus lens position focused on the subject, and the in-focus face image information at the focus lens position is acquired (step S0601). Next, the position of the focus lens is shifted to acquire out-of-focus face image information (step S0602). Subsequently, feature area image information such as eyes, nose, mouth, and ears is acquired from the focused face image information acquired in step S0601 (step S0603). Also, skin image information is acquired from the out-of-focus face image information acquired in step S0602 (step S0604). Finally, the feature area image information acquired in step S0603 and the skin image information acquired in step S0604 are combined to form a human face (step S0605). Of course, as long as no contradiction arises, for example, the order of step S0601 and step S0602 may be switched to reverse the acquisition order of the in-focus face image information and the out-of-focus face image information.

    (Simple description of the effect of Example 1) As described above, the facial camera of the present example uses the conventional focus technology without using a dedicated application, and the eyes and nose are sharp, while the skin. Facial photos that are blurred so that no wrinkles are clearly visible can be easily generated with a natural finish.

  << Example 2 >> (Concept of Example 2) The facial camera of the present example is based on the facial camera of Example 1 and is represented by skin image information and eyes and nose indicated by the feature region image information. Create a facial photo by combining the skin part. However, the eyes and nose acquired from the in-focus face image and the skin acquired from the non-focused face image differ in terms of color tone and contrast, and the boundary between the synthesis may be unnatural. high. Therefore, the beauty camera of the present embodiment is characterized by including means for correcting the boundary so that it becomes natural.

    (Configuration of Example 2) FIG. 7 is a diagram illustrating an example of functional blocks of a beauty camera of the present example. As shown in this figure, the beauty camera (0700) of the present embodiment is based on the first embodiment, with a “focus lens control section” (0701), a “feature area image information acquisition section” (0702), “ A skin image information acquisition unit "(0703); and a" synthesizing unit "(0704). However, since the above functional blocks have already been described in the first embodiment, description thereof will be omitted. As a feature point, the synthesis unit (0704) further includes a “boundary image correction unit” (0705).

  The “boundary image correction unit” (0705) has a function of correcting an image so that a boundary of a combined boundary region between the feature region image information and the skin image information is naturally changed. Examples of the correction method of the boundary image correction means include a method of performing blurring processing on the combined boundary region and correcting the contrast and color tone so that the contrast and color tone of the combined boundary region change hierarchically. FIG. 8 is a diagram for explaining an example of correction by the boundary image correction unit. In the figure, (a) is a diagram showing the eye periphery of an image before being corrected by the boundary image correcting means. Here, when the contrast value for each pixel of this image is acquired, there is a discontinuous difference in the boundary portion between the eye portion acquired from the focused image and the skin portion acquired from the non-focused image. . Therefore, as shown by (b) in the figure, the image is corrected so that the contrast value continuously changes. Then, since the contrast of the composite boundary region seems to change continuously, the boundary seems to change naturally.

    (Processing Flow of Second Embodiment) FIG. 9 is a flowchart showing an example of the processing flow of the present embodiment. As shown in this figure, first, the focus lens is driven to determine the focus lens position focused on the subject, and the in-focus face image information at the focus lens position is acquired (step S0901). Next, the position of the focus lens is shifted to acquire unfocused face image information (step S0902). Subsequently, feature area image information such as eyes, nose, mouth, and ears is acquired from the focused face image information acquired in step S0901 (step S0903). Also, the skin image information is acquired from the out-of-focus face image information acquired in step S0902 (step S0904). Finally, image correction is performed so that the boundary of the combination boundary region between the feature region image information acquired in step S0903 and the skin image information acquired in step S0904 is changed naturally, and the image is formed as a human face image. (Step S0905).

    (Simple explanation of the effect of Example 2) As described above, it is more natural that the facial camera of this example makes it difficult to distinguish the boundary region between the feature region such as the eyes, nose, mouth and ears and the skin. You can take beautiful facial photos.

  << Example 3 >> (Concept of Example 3) FIG. 10 is a diagram for explaining an example of the concept of the beauty camera of the present example. As shown in this figure, the in-focus face image and the out-of-focus face image obtained by the beauty camera in Examples 1 and 2 are slightly different in size. This is because the position of the focus lens is deviated, and the location where the light of the bent subject is imaged changes in the imaging unit such as a CCD. Therefore, as shown in FIG. 11, the facial camera of this embodiment is characterized in that the size is corrected so that the size of the skin image matches the feature region image.

    (Configuration of Example 3) FIG. 12 is a diagram illustrating an example of functional blocks of a beauty camera according to the present example. As shown in this figure, the beauty camera (1200) of the present embodiment is based on the first or second embodiment, and includes a “focus lens control unit” (1201), a “feature area image information acquisition unit” (1202), and , “Skin image information acquisition unit” (1203) and “synthesis unit” (1204). However, since the above functional blocks have already been described in the first embodiment, description thereof will be omitted. As a feature point, the skin image information acquisition unit (1203) further includes a “magnification correction unit” (1205).

  The “magnification correction unit” (1205) has a function of correcting the image magnification based on the difference in focus between the feature area image information and the skin image information. As described above, the difference in size between the feature area image information and the skin image information is caused by the shift of the imaging location due to the shift of the position of the focus lens. Therefore, it is possible to calculate the image magnification from the value indicating the shift of the focus lens position. For example, when the shift of the focus lens position is 0.5 mm, a table may be created and stored in a memory or the like such that the image magnification is 99%. Alternatively, a predetermined calculation formula corresponding to the lens characteristics or the like may be stored in a memory or the like, and calculation processing may be performed each time. Based on the image magnification obtained in this way, the magnification correction means corrects the size of the image.

  Since the magnification correction is a correction of the relative magnification between the skin image and the feature region image, the same effect can be obtained even if the magnification correction means is provided in the “feature region image information acquisition unit”. Similarly, even when magnification correction is performed on a non-focused face image before acquiring a skin image (or a focused face image before acquiring a feature area image), the feature area image information and the skin image information are also corrected. It is possible to correct the image magnification based on the difference in focus.

  FIG. 13 is a diagram showing an example of functional blocks of another beauty camera of this embodiment having the magnification correction means as described above. As shown in this figure, the other beautiful camera of the present embodiment is based on the first or second embodiment, the “focus lens control section” (1301), the “characteristic region image information acquisition section” (1302), It has a “skin image information acquisition unit” (1303) and a “synthesis unit” (1304). As a feature point, a “magnification correction unit” (1305) for correcting the image magnification of the size of the unfocused face image is further provided.

    (Processing Flow of Embodiment 3) FIG. 14 is a flowchart showing an example of the processing flow of the present embodiment. As shown in this figure, first, the focus lens is driven to determine the focus lens position focused on the subject, and the in-focus face image information at the focus lens position is acquired (step S1401). Next, the position of the focus lens is shifted to acquire unfocused face image information (step S1402). Subsequently, feature area image information such as eyes, nose, mouth, and ears is acquired from the focused face image information acquired in step S1401 (step S1403). Also, skin image information is acquired from the out-of-focus face image information acquired in step S1402 (step S1404). Finally, the image magnification based on the difference in focus between the feature region image information acquired in step S1403 and the skin image information acquired in step S1404 is corrected, and the corrected feature region image information and skin image information are obtained. An image of a human face is synthesized (step S1405). Of course, as described above, the flow of processing for correcting the magnification of a non-focused face image or a focused face image may be used. In this case, for example, instead of the image magnification correction process in step S1405, the image magnification correction process is performed after obtaining the in-focus face image information and the out-of-focus face image information in step S1401 and step S1402.

    (Simple explanation of the effect of Example 3) As mentioned above, with the facial camera of the present Example, a subtle difference in size due to the shift of the focus lens position between the characteristic region such as the eyes, nose, mouth and ears and the skin. Can be taken, and a more natural facial photo can be taken.

  << Embodiment 4 >> (Concept of Embodiment 4) The facial camera of the present embodiment can receive in-focus face image information and out-of-focus face image information by inputting one image information acquisition signal such as pressing a shutter once. Both are acquired. This means that the in-focus face image information and the out-of-focus face image information are acquired by one shooting operation of the user. Therefore, the user can obtain the in-focus face image information and the in-focus face image information by two or more shooting operations. The time lag (time difference) between the acquisition of both of them is less than when the non-focused face image information is acquired separately. In other words, the in-focus face image information and the out-of-focus face image information that are acquired when the subject moves during this time lag or the position of the facial camera is shifted due to camera shake or the like becomes image information that is significantly different. The possibility can be suppressed.

    (Configuration of Example 4) FIG. 15 is a diagram illustrating an example of functional blocks of a beauty camera of the present example. As shown in this figure, the facial camera (1500) of the present embodiment is based on any one of the first to third embodiments, and includes a “focus lens control unit” (1501) and a “feature area image information acquisition unit” ( 1502), “skin image information acquisition unit” (1503), and “synthesis unit” (1504). However, since the above functional blocks have already been described in the first embodiment, description thereof will be omitted. As a feature point, an “image information acquisition signal input unit” (1505) is further provided.

  The “image information acquisition signal input unit” (1505) has a function of inputting an image information acquisition signal for acquiring in-focus face image information and non-focused face image information. Etc. Then, based on one image information acquisition signal input to the image information acquisition signal input unit, the focus lens control unit (1501) controls the position of the focus lens and acquires the in-focus face image information and the non-focusing image information. Both of the facial image information are acquired.

    (Processing Flow of Embodiment 4) FIG. 16 is a flowchart showing an example of the processing flow of the present embodiment. As shown in this figure, first, one image information acquisition signal is input (step S1601). Then, the focus lens is driven to determine the position of the focus lens focused on the subject, and the in-focus face image information at the focus lens position is acquired (step S1602). In-focus face image information is acquired (step S1603). Subsequently, feature area image information such as eyes, nose, mouth, and ears is acquired from the focused face image information acquired in step S1602 (step S1604). Also, skin image information is acquired from the out-of-focus face image information acquired in step S1603 (step S1605). Finally, the feature area image information acquired in step S1604 and the skin image information acquired in step S1605 are combined to form a human face (step S1606).

    (Simple description of the effect of Example 4) As mentioned above, a subject moves between the time lags of acquisition of focused face image information and non-focused face image information by the facial camera of the present example, and the facial camera Therefore, it is possible to suppress the possibility that the in-focus face image information and the out-of-focus face image information acquired by shifting the position of the camera due to camera shake or the like are greatly different from each other. Therefore, it is possible to suppress a situation in which a beautiful face photo that is completed with image information that has almost no difference in size and position between the feature area image information and the skin image information to be combined differs greatly from the original image.

  << Example 5 >> (Concept of Example 5) FIG. 17 is a diagram for explaining an example of the concept of the beauty camera of this example. As shown in this figure, the autofocus mechanism of the camera acquires image information at various focus lens positions in order to determine the focus lens position focused on the subject, and compares the contrast data. As described above, the camera acquires the out-of-focus face image information before determining the focus lens position for acquiring the in-focus face image information. Therefore, by storing and using the non-focused face image information acquired when determining the focus lens position in a memory or the like, it is possible to save the trouble of acquiring the non-focused face image information again and shorten the processing time. it can.

    (Configuration of Example 5) FIG. 18 is a diagram illustrating an example of a functional block of the facial camera of the present example. As shown in this figure, the facial camera (1800) of the present embodiment is based on any one of the first to fourth embodiments, and includes a “focus lens control unit” (1801) and a “feature area image information acquisition unit”. (1802), “skin image information acquisition unit” (1803), and “synthesis unit” (1804). However, since the above functional blocks have already been described in the first embodiment, description thereof will be omitted. As a feature point, the focus lens control unit further includes an “auto focus lens position acquisition unit” (1805).

  The “auto focus lens position acquisition means” (1805) has a function for automatically acquiring the position of the focus lens. For example, means for collecting out-of-focus face image information for each focus lens position, It consists of means for acquiring contrast data (luminance values, etc.) of a dark face image, and arithmetic means for processing and comparing the contrast data. Of the non-focused face image information collected here, for example, the non-focused face image information acquired at the closest focus lens position to the focus lens position from which the focused face image information was acquired is the skin image. Used to obtain information.

    (Processing Flow of Embodiment 5) FIG. 19 is a flowchart showing an example of the processing flow of the present embodiment. As shown in this figure, first, image information is collected in order to automatically acquire the focus lens position focused on the subject (step S1901). Next, in-focus face image information at the focus lens position determined based on the collection result in step S1901 is acquired (step S1902). Subsequently, out-of-focus face image information is acquired from the image information collected in step S1901 (step S1903). Further, feature area image information such as eyes, nose, mouth, and ears is acquired from the focused face image information acquired in step S1902 (step S1904). In addition, skin image information is acquired from the out-of-focus face image information acquired in step S1903 (step S1905). Finally, the feature area image information acquired in step S1904 and the skin image information acquired in step S1905 are combined to form a human face (step S1906).

    (Simple explanation of the effect of Example 5) As mentioned above, the face camera of a present Example uses the image information which an autofocus mechanism collects in order to determine a focus lens position, and it becomes a non-focusing again. The processing time can be shortened by saving the trouble of acquiring the face image information.

  The beauty camera of the present invention can be used not only for a normal camera to be carried but also for an instant ID photo photographic device installed on a street, for example. As described in the prior art, it is preferable to use a facial camera to make a good impression in job hunting activities, etc. However, it is convenient if the ID photo can be taken with such an instant ID photo shooting device on the street.

The figure for demonstrating an example of the concept of the beauty camera of Example 1. FIG. The figure showing an example of the functional block of the beauty camera of Example 1 The figure showing an example of the characteristic area image information of Example 1 The figure showing an example of the image information for skin of Example 1 The figure for demonstrating an example of a synthesis | combination in the synthetic | combination part of the beauty camera of Example 1. FIG. The flowchart showing an example of the flow of a process of Example 1. The figure showing an example of the functional block of the beauty camera of Example 2. The figure for demonstrating an example of the correction | amendment of the boundary image correction | amendment means of the beauty camera of Example 2. FIG. The flowchart showing an example of the flow of processing of Example 2. The figure for demonstrating an example of the concept of the beauty camera of Example 3. The figure for demonstrating an example of the concept of the beauty camera of Example 3. The figure showing an example of the functional block of the beauty camera of Example 3 The figure showing an example of the functional block of the other beauty camera of Example 3 The flowchart showing an example of the flow of a process of Example 3. The figure showing an example of the functional block of the beauty camera of Example 4 The flowchart showing an example of the flow of a process of Example 4. The figure for demonstrating an example of the concept of the beauty camera of Example 5. The figure showing an example of the functional block of the beauty camera of Example 5 Flowchart showing an example of the flow of processing of the fifth embodiment

Explanation of symbols

0200 Beauty camera 0201 Focus lens control unit 0202 Feature region image information acquisition unit 0203 Skin image information acquisition unit 0204 Compositing unit

Claims (5)

In-focus face image information, which is image information focused on a person's face, and unfocused face image information, which is image information of the person's face that is out of focus from the focus of the in-focus face image information. A focus lens control unit for controlling the position of the focus lens as possible,
A feature region image information acquisition unit that acquires feature region image information that is image information of a feature region of a face from in-focus face image information acquired based on control of the focus lens of the focus lens control unit;
A skin image information acquisition unit that acquires skin image information that is image information of the skin of the face from unfocused face image information acquired based on control of the focus lens of the focus lens control unit;
A synthesis unit that combines the feature region image information acquired by the feature region image information acquisition unit with the skin image information acquired by the skin image information acquisition unit to form an image of a human face;
A beautiful face camera.   The facial camera according to claim 1, wherein the synthesizing unit includes boundary image correction means for correcting an image so that a boundary of a synthesis boundary region between the feature region image information and the image information for skin changes naturally.   The facial camera according to claim 1 or 2, wherein the synthesis in the synthesis unit is performed by correcting an image magnification based on a difference in focus between the feature region image information and the skin image information. It has an image information acquisition signal input unit,
Both focused face image information and non-focused face image information acquired by controlling the position of the focus lens by the focus lens control unit are one image information acquisition signal input to the image information acquisition signal input unit. The face camera according to any one of claims 1 to 3, which is acquired together on the basis of. The focus lens control unit has automatic focus lens position acquisition means for automatically acquiring the position of the focus lens,
5. The facial camera according to claim 1, wherein the out-of-focus face image information is image information collected by the automatic focus lens position acquisition unit to automatically acquire a focus position of a focus lens.

Images