JP2012053313A - Imaging apparatus, automatic zooming method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an imaging apparatus that has improved stability in automatic zooming operation.SOLUTION: The imaging apparatus includes: an imager for photo-electrically converting a subject image formed by an imaging optical system including a zoom lens and outputting an image signal; subject detecting means for detecting a main subject based on the image signal output from the imager; zoom control means for performing a variable power operation by driving the zoom lens and executing automatic zooming control so that the main subject has a constant size on a screen; and control means for stopping the automatic zooming control if the amount of change of the main subject on the screen is larger than a predetermined value.

Description

  The present invention relates to an imaging apparatus having an auto zoom function.

  2. Description of the Related Art Conventionally, there has been known a camera having a so-called auto zoom function that automatically performs zooming so that an image magnification set in the camera in advance is set according to the subject distance when the camera is pointed at an arbitrary subject. . Patent Document 1 discloses a method of calculating a face image area on a display screen, comparing the face area with a reference face area, and always capturing a face image having the same size.

  Patent Document 2 discloses an imaging device that automatically detects a human face from a captured image by a technique such as shape analysis. In this imaging apparatus, when a human face is detected, the fact is displayed on the screen, a focus detection area is displayed for the detected human face, and a focus adjustment operation is performed on the focus detection area. Do. According to such an imaging apparatus, it is possible to perform a focus adjustment operation on a person's face by detecting the person's face from the entire captured image regardless of the position or range of the person on the screen.

Japanese Patent Laid-Open No. 06-153047 JP 2003-107335 A

  However, for example, in an imaging apparatus that executes an auto zoom function for a face detected using the face detection function, if the auto zoom function is enabled for a moving face, the image disappears from the shooting screen. There is a possibility that Although it is possible to place the camera in the center by the photographer's camera work, if the face moves at a high speed in the shooting screen, the frequency of disappearing from the screen immediately increases. In this way, even if the auto zoom function is enabled, if it disappears immediately from the shooting screen, the zoom operation is repeatedly executed or not executed, or the face is lost from the shooting screen However, zooming is performed. Then, when the photographer views the recorded video, there is a possibility that the user feels uncomfortable because of an uncomfortable zoom operation.

  Therefore, the present invention provides an imaging apparatus that improves the stability of auto-zoom operation.

  An image pickup apparatus according to one aspect of the present invention is based on an image pickup device that photoelectrically converts a subject image formed by an image pickup optical system including a zoom lens and outputs an image signal, and the image signal output from the image pickup device. Subject detection means for detecting the main subject, zoom control means for performing zooming operation by driving the zoom lens, and performing auto zoom control so that the main subject has a constant size on the screen, and Control means for stopping the auto-zoom control when the amount of change of the main subject on the screen is larger than a predetermined value.

  An auto zoom method and program for executing the auto zoom control also constitute another aspect of the present invention.

  Other objects and features of the present invention are illustrated in the following examples.

  According to the present invention, it is possible to provide an imaging apparatus with improved auto-zoom operation stability.

It is a block diagram which shows schematic structure of the imaging device in a present Example. It is a flowchart which shows the procedure of the auto zoom control in 1st Embodiment. It is a flowchart which shows the procedure of the movement monitoring process in 1st Embodiment. In 1st Embodiment, it is an image figure which shows the face position in the angle of view after predetermined time. It is a flowchart which shows the procedure of the other movement monitoring process in 1st Embodiment. It is a flowchart which shows the procedure of the auto zoom control in 2nd Embodiment. It is a flowchart which shows the procedure of the movement monitoring process in 2nd Embodiment. In 2nd Embodiment, it is an image figure of the presence determination of the face in the angle of view after predetermined time.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In each figure, the same members are denoted by the same reference numerals, and redundant description is omitted.

  First, an imaging apparatus according to an embodiment of the present invention will be described. FIG. 1 is a block diagram illustrating a schematic configuration of an imaging apparatus (video camera) of the present embodiment. In this embodiment, a video camera will be described as an image pickup apparatus. However, the present invention is not limited to this, and the present embodiment can also be applied to other image pickup apparatuses such as a digital still camera.

  In FIG. 1, a video camera 100 has a first fixed lens 101, a variable power lens 102, a diaphragm 103, a second fixed lens 104, and a focus compensator lens 105 arranged in this order in the optical axis direction. Composed. The zoom lens 102 is a zoom lens that performs zooming by moving in the optical axis direction. A focus compensator lens (hereinafter referred to as a focus lens) 105 is a lens that has both a function of correcting the movement of the focal plane accompanying zooming and a focusing function. An image sensor 106 is disposed behind the focus lens 105. The image sensor 106 is a photoelectric conversion element configured by a CCD sensor or a CMOS sensor, and converts the formed optical image into an electrical signal and outputs it to the CDS / AGC circuit 107. Thus, the image sensor 106 photoelectrically converts the subject image formed by the imaging optical system including the variable magnification lens 102 and outputs an image signal. The CDS / AGC circuit 107 samples the output of the image sensor 106 and adjusts the gain.

  The video camera 100 includes a camera signal processing circuit 108, a monitor unit 109, a zoom drive source 110, a focusing drive source 111, an AF gate 112, an AF signal processing circuit 113, a camera / AF microcomputer 114, and a recording unit 115. The camera signal processing circuit 108 performs various kinds of image processing on the output signal from the CDS / AGC circuit 107 to generate a video signal. The monitor unit 109 is a monitor (screen) configured by an LCD or the like, and displays the video signal output from the camera signal processing circuit 108 as a captured image. The recording unit 115 records the video signal output from the camera signal processing circuit 108 on a recording medium such as a magnetic tape, an optical disk, or a semiconductor memory.

  The AF gate 112 allows the AF signal processing circuit 113 to pass only the signal in the region used for focus adjustment among the output signals of all the pixels from the CDS / AGC circuit 107. The AF signal processing circuit 113 extracts a high frequency component, a luminance difference component, and the like from the signal that has passed through the AF gate 112 to generate an AF evaluation value signal, and the generated AF evaluation value signal is converted into a camera / AF microcomputer 114 (hereinafter referred to as “AF evaluation value signal”). Simply called “computer”. The luminance difference component is, for example, a difference between the maximum value and the minimum value of the luminance level of the signal that has passed through the AF gate 112. The AF evaluation value signal represents the sharpness of an image generated based on the output signal from the image sensor 106. Since the sharpness varies depending on the focus state of the imaging optical system, the AF evaluation value signal is a signal representing the focus state of the imaging optical system.

  A computer 114 as a control unit controls the entire video camera 100. The computer 114 performs, for example, zoom control and focus control. In zoom control, the zoom drive source 110 is controlled to drive the variable power lens 102. The computer 114 and the zoom drive source 110 constitute zoom control means that performs a zoom operation by driving the zoom lens 102 and performs auto zoom control so that the main subject has a constant size on the screen.

  The focus control controls the focusing drive source 111 to drive the focus lens 105. The zoom drive source 110 and the focusing drive source 111 are configured by actuators such as a stepping motor, a DC motor, a vibration motor, and a voice coil motor. Here, for the focus control by the computer 114, control by a known TV-AF method (hereinafter simply referred to as “TV-AF”) can be used. Also, during execution of the auto zoom control operation, focusing is performed during known zoom control while controlling the zoom lens 102 and the focus lens 105 by the TV-AF method. Further, a zoom lens position detection device 120 and a focus lens position detection device 121 are provided, and detect the positions of the zoom lens 102 and the focus lens 105 in the optical axis direction, respectively. Then, zoom control is performed while the zoom lens position detection device 120 detects the zoom lens position, and focus control is performed while the focus lens position detection device 121 detects the focus lens position. Further, the zoom lens position (focal length) and zoom magnification are stored in storage means (not shown) such as a nonvolatile memory provided inside or outside the computer 114.

  Further, the video camera 100 includes a face detection processing circuit 116, a face detection mode switching unit 117, a selection instruction unit 118, and an auto zoom mode switching unit 119. A face detection processing circuit 116 as subject detection means performs a known face recognition process on the video signal output from the camera signal processing circuit 108. Then, the main subject (face area) in the shooting screen is detected, and the detection result including the subject position information is transmitted to the computer 114. The subject position information is information related to the position of the subject region in the shooting screen. In this way, the face detection processing circuit 116 detects the main subject based on the image signal output from the image sensor 106.

  Based on the received detection result, the computer 114 transmits information to the AF gate 112 in order to add an area used for focus detection to a position including a face area in the shooting screen. As a known face recognition process, for example, a skin color area is extracted from the gradation color of each pixel represented by image data corresponding to a video signal, and the face area is detected by a matching degree with a face contour plate prepared in advance. There is a way to do it. In addition, there is a method for detecting a face region by extracting feature points of a face such as eyes, nose and mouth using a well-known pattern recognition technique. The face recognition process in the present embodiment is not limited to the above-described method, and any known method may be used. Further, in order to notify the photographer of the face area recognized by the face detection processing circuit 116, the computer 114 transmits information such as the position and size of the face area to the camera signal processing circuit 108, and corresponds to the information. The displayed frame display is superimposed on the video signal and displayed on the monitor unit 109.

The face detection mode switching unit 117 has a function of switching between valid / invalid of control (face detection control) by face detection processing. When the face detection control is valid, the computer 114 performs control according to the result of the face detection process for AF control or the like. The selection instruction unit 118 has a function of instructing a photographer to select a specific face when the photographer intends to fix the specific subject in face detection control. When the photographer selects and instructs a specific face using an operation member such as a cross key or a touch panel, the computer 114 performs AF control or the like on the selected and designated face (hereinafter, “main face”). Called "selection"). The auto zoom mode switching unit 119 has a function of switching between enabling / disabling of the auto zoom function. When the auto-zoom function is valid, the computer 114 controls the zoom drive source 110 according to the result of the face detection process, thereby performing zoom control so that the detected face always has a constant size.
[First Embodiment]
Next, with reference to FIG. 2, the auto zoom control in the first embodiment of the present invention will be described. FIG. 2 is a flowchart showing a procedure of auto zoom control in the present embodiment. The face detection control process shown in the flow of FIG. 2 is executed according to a program stored in a memory in the computer 114.

  First, in step S201, the computer 114 starts auto zoom control. Note that the processing of this flow is repeatedly executed, for example, in the readout cycle of the imaging signal from the imaging device 106 for generating one field image. Next, in step S202, the face detection processing circuit 116 acquires the position and size of the face to be subjected to auto zoom control on the screen. In step S203, the current zoom position is acquired.

  In step S204, the face size acquired in step S202 is compared with the previously acquired face size to determine whether the face size has changed by a predetermined threshold size or more. If the size of the face has changed by a predetermined threshold size or more, it is determined that the face subject to auto zoom control is moving, and the process proceeds to the movement monitoring process in step S205 described later. Details of the movement monitoring process in step S205 will be described later. On the other hand, if the face size has not changed by more than the predetermined threshold size, the process proceeds to step S209.

  After the movement monitoring process in step S205, it is determined in step S206 whether the zoom operation prohibition flag is set. If the zoom operation prohibition flag is not set, the process proceeds to step 207. On the other hand, if the zoom operation prohibition flag is set, the process proceeds to step S209. If the zoom operation prohibition flag is set, the target zoom position is newly set and the zoom operation is executed, so that the face subject to auto zoom control disappears from the shooting angle of view. It is in a state where it is determined that there is a high possibility of being lost or disappearing. Accordingly, it is possible to prevent the zoom operation from being performed inadvertently when the face subject to auto zoom control is moved and disappears immediately from the shooting angle of view. In other words, the unstable operation caused by executing the auto-zoom operation despite the disappearance of the face subject to auto-zoom control from the shooting angle of view is reduced, and the control that makes the photographer feel uncomfortable Is possible.

  The transition to step S209 is that the face size acquired in step S204 is the same as the previously acquired face size, or the zoom operation prohibition flag is set in the movement monitoring process in step S205. Indicates. For this reason, a new target zoom position is not set and the target zoom position is not updated. However, if it is determined that the zoom position has not reached the previously set target zoom position, the zoom operation is executed up to the target zoom position. Therefore, in step S209, it is determined whether or not the zoom position has reached the target zoom position. If the current zoom position acquired in step S203 has not reached the target zoom position, the process proceeds to step S207. To do. On the other hand, if the current zoom position has reached the target zoom position, the process proceeds to step S210.

  In step S207, the zoom is driven to the set target zoom position. On the other hand, in step S210, the zoom drive is stopped so as not to perform the zoom operation. In step S208, information on the face position and size history subject to auto-zoom control, as well as information on the current zoom position or zoom magnification, is stored in a volatile memory (not shown). Save to the figure.

  Next, with reference to FIG. 3, the movement monitoring process in the auto zoom control of the present embodiment will be described. FIG. 3 is a flowchart showing the procedure of the movement monitoring process (step S205) in the present embodiment. First, when the movement monitoring process is started in step S301, a zoom position to be moved next is provisionally calculated in step S302 based on the acquired current face size and current zoom position. For example, the previous face size is compared with the current face size. If the face size has changed to 1.5 times, for example, if the current zoom magnification is 3 times, A zoom position at which the magnification is twice is temporarily calculated as a movement position. Here, by previously storing zoom magnification table data indicating the relationship between the zoom position (focal length) or the zoom magnification in the storage means, the zoom position at which the zoom magnification is 3 times or 2 times can be easily calculated. can do. In step S303, the driving speed of the variable power lens 102 when the variable power lens 102 is moved to the moving position is set. The drive speed of the zoom lens 102 may be a predetermined fixed drive speed that does not cause the photographer to feel uncomfortable when moving the zoom lens 102 when auto zoom control is being executed.

  In step S304, an angle of view after a predetermined time is calculated based on the zoom position (zoom lens position) obtained in step S302 and the driving speed (zoom speed) of the variable power lens 102 set in step S303. . The calculation of the angle of view is performed by the angle of view calculation unit of the calculation means (computer 114). Here, “after the predetermined time” may be equal to the detection cycle of the face detection processing circuit 116. For example, if the detection cycle of the face detection processing circuit 116 is one frame unit, the predetermined time is 1/30 second. In addition, as a method of calculating the angle of view after a predetermined time, for example, first, the movement amount L1 of the variable power lens 102 after a predetermined time is calculated from the predetermined time described above and the driving speed of the variable power lens 102 set in step S303. . Then, the zoom position P2 when moving by the movement amount L1 from the current zoom position is calculated, and the zoom magnification Z2 after a predetermined time is derived from the zoom magnification table data. Thereby, it is possible to calculate what angle of view the angle of view after a predetermined time becomes relative to the current angle of view. For example, when the current zoom magnification is 4 times and the zoom magnification Z2 after a predetermined time is 2 times, the angle of view after the predetermined time is about twice the angle of view with respect to the current angle of view.

  Next, in step S305, the position and size of the face subject to auto-zoom control that has been recorded in step S208, and further, the position and size of the face at the angle of view after a predetermined time calculated in step S304 from the zoom position. Is calculated. In other words, the computer 114 serving as calculation means monitors the position and size of the main subject on the screen and calculates the position and size of the main subject after a predetermined time. Here, first, the position and size of the face at the angle of view after a predetermined time using the history of the position and size of the face that is the target of auto-zoom control stored in step S208 and information about the zoom position. Is calculated. This calculation is performed by the subject position calculation unit of the calculation means. For example, the previous X position X (t), Y coordinate position Y (t), and face size F of the face that is the target of auto-zoom control that has been recorded in history and the zoom position. (T) Assume that the zoom position is a position of Z (t) times as a zoom magnification. It is also assumed that the X coordinate position X (t + 1), Y coordinate position Y (t + 1), face size F (t + 1), and zoom position of the current face are Z (t + 1) times as zoom magnification. Then, it is assumed that the angle of view after a predetermined time is a zoom position of Z (t + 2) times as a zoom magnification. The X coordinate position and the Y coordinate position are based on the center of the screen. In this case, the X coordinate position X1 (t), the Y coordinate position Y1 (t) of the previous face, and the face size F1 when the zoom position is moved to the zoom position with the zoom magnification Z (t + 2) times that is the angle of view after a predetermined time. (T) is represented as the following formulas (1-1), (1-2), and (1-3), respectively.

X1 (t) = X (t) × Z (t + 2) / Z (t) (1-1)
Y1 (t) = Y (t) × Z (t + 2) / Z (t) (1-2)
F1 (t) = F (t) × Z (t + 2) / Z (t) (1-3)
Also, the X coordinate position X1 (t + 1), the Y coordinate position Y1 (t + 1) of the previous face, and the face size F1 (when moving to a zoom position with a zoom magnification Z (t + 2) times that is the angle of view after a predetermined time. t + 1) is expressed as the following equations (2-1), (2-2), and (2-3), respectively.

X1 (t + 1) = X (t + 1) × Z (t + 2) / Z (t + 1) (2-1)
Y1 (t + 1) = Y (t + 1) × Z (t + 2) / Z (t + 1) (2-2)
F1 (t + 1) = F (t + 1) × Z (t + 2) / Z (t + 1) (2-3)
From the movement amount (displacement amount) of the X coordinate positions X1 (t) and X1 (t + 1) of the face calculated as described above, the X coordinate position X (t + 2) of the face at the angle of view after a predetermined time is expressed by the following equation: It is expressed as (3-1).

X (t + 2) = X1 (t + 1) + (X1 (t + 1) -X1 (t)) (3-1)
Similarly, from the movement amount (displacement amount) of the face Y coordinate positions Y1 (t) and Y1 (t + 1), the face Y coordinate position Y (t + 2) at the angle of view after a predetermined time is expressed by the following equation (3- It is expressed as 2).

Y (t + 2) = Y1 (t + 1) + (Y1 (t + 1) −Y1 (t)) (3-2)
Further, from the amount of displacement of the face sizes F1 (t) and F1 (t + 1), the face size F (t + 2) at the angle of view after a predetermined time is expressed as the following equation (3-3).

F (t + 2) = F1 (t + 1) + (F1 (t + 1) −F1 (t)) (3-3)
FIG. 4 is an image diagram of the position at the angle of view after a predetermined time calculated as described above. Here, it is assumed that the subject that is the target of the auto zoom control has moved as indicated by the arrow shown in FIG. 401 is the previous and current angle of view, 402 is the previous subject, and 403 is the current subject. On the other hand, FIG. 4B shows the face position and size (subject) based on the angle of view after a predetermined time. 404 is the angle of view after a predetermined time, 405 is the previous subject based on the angle of view after the predetermined time, 406 is the current subject based on the angle of view after the predetermined time, and 407 is the angle of view after the predetermined time. This is a subject (prediction) after a predetermined time as a reference. Reference numeral 408 denotes the previous and current view angles with respect to the view angle after a predetermined time. In the image diagram shown in FIG. 4, it is assumed that the angle of view after a predetermined time is about 1.5 times the previous and current angles.

  Next, in step S306, it is determined whether or not there is a face to be subjected to auto zoom control at the angle of view after the predetermined time of the face position and size calculated by prediction in step S305. In other words, the computer 114 as the determination unit determines whether or not a predetermined ratio of the main subject after a predetermined time exists within a predetermined area of the screen based on the calculation results of the calculation unit (view angle calculation unit and subject position calculation unit). Determine whether. Specifically, in FIG. 4B, it is determined whether or not the subject at the angle of view after the predetermined time is included in the angle of view 404 after the predetermined time by at least a predetermined area ratio (predetermined ratio). For example, it is determined whether or not an area that is 3/4 times the subject area of the face position and size at a view angle after a predetermined time is included in the view angle 404 after a predetermined time. In FIG. 4B, since the subjects 402 and 403 at the angle of view 404 after a predetermined time are included in the angle of view 404, it is determined that a face exists. On the other hand, regarding the subject 407 at the angle of view 404 after a predetermined time, a predetermined area ratio (for example, 3/4 area) of the subject 407 is not included in the angle of view 404 within the angle of view 404 after the predetermined time. In this case, it is determined that no face exists.

  In step S306, if there is a face to be subjected to auto zoom control within the angle of view after a predetermined time, the process proceeds to step S307. On the other hand, if there is no face subject to auto zoom control, the process proceeds to step S308. In step S307, since there is a high possibility that a face exists, the zoom position temporarily calculated in step S302 is set as the target zoom position. On the other hand, in step S308, since it is unlikely that a face exists, the zoom operation prohibition flag is set. That is, the computer 114 serving as a stopping unit stops (inhibits) the auto zoom control when it is determined that the predetermined ratio of the main subject does not exist within the predetermined area of the screen. As described above, the computer 114 is a control unit that stops the auto zoom control when the amount of change of the main subject on the screen is larger than a predetermined value. When the zoom operation prohibition flag is set, the computer 114 does not newly set the target zoom position and does not update the target zoom position in the auto zoom control.

  In this embodiment, the discrimination angle of view is set inside the angle of view 404 after a predetermined time, and it is determined whether or not the subject 406 (or subject 407) having the face position and size exists within the discrimination angle of view. May be. The discrimination angle of view can be made variable according to the amount of change (movement amount) of the face position. FIG. 5 is a flowchart showing the procedure of the movement monitoring process (step S2051) in this case. In FIG. 5, steps S501 to S505, S507, and S508 are the same as steps S301 to S305, S307, and S308 of FIG.

  In step S510 of FIG. 5, the movement amount calculation unit of the calculation unit calculates the movement amount of the face from the history of the face positions calculated by the subject position calculation unit. That is, the amount of movement of the face at the angle of view after the predetermined time is calculated based on the position of the face to be subjected to auto zoom control at the angle of view after the predetermined time calculated at step S505. In step S511, a discrimination angle of view (predetermined region) is set inside the angle of view after a predetermined time from the amount of face movement. In step S512, the determination unit determines whether or not a predetermined ratio of the face exists within the determination angle of view. For example, when the movement amount of the face is smaller than a predetermined first movement amount, the determination angle of view is made equal to the angle of view after a predetermined time. On the other hand, when the movement amount of the face is larger than the predetermined first movement amount, an area smaller than the angle of view after a predetermined time is set as the determination angle of view. As described above, the determination unit changes the size of the predetermined area so that the predetermined area on the screen becomes smaller as the movement amount of the main subject calculated by the movement amount calculation unit is larger.

  If the face position exists at the end of the angle of view after a predetermined time, an effective operation may be difficult in the next auto zoom control. In addition, when the amount of change in the face position is large, there is a possibility that it will disappear from the screen in the next auto zoom control. For this reason, a discrimination field angle is set inside the angle of view 504 after a predetermined time in accordance with the position of the face and the change of the face position, and it is determined whether or not the angle of view exists. As a result, it is possible to determine whether or not to continue the execution of the auto zoom control before determining at the next control cycle, and it is possible to more reliably determine whether or not the face subject to the auto zoom control exists. Is possible. For this reason, an inadvertent zoom operation can be reduced, and a photographer's uncomfortable feeling can be reduced.

In an imaging device having an auto zoom function that can always shoot with the face size kept constant, if the face that is subject to auto zoom control is a fast-moving subject or a moving subject There is a possibility of disappearing from the angle of view. At this time, if the zoom operation is performed carelessly, the zoom operation is performed even when there is no face that is subject to auto zoom control from the shooting screen, which may cause the photographer to feel uncomfortable or uncomfortable. In addition, when the target face is arranged in the screen and the zoom operation is performed, the target face disappears repeatedly, so that stable control cannot be performed and the quality of the captured image may deteriorate. For this reason, in this embodiment, when the face size is changed using the history of the face to be subjected to auto zoom control, whether or not the target face exists within the shooting angle of view that is changed by the auto zoom operation. Determine whether. If the face is present, the auto zoom operation is continued. If the face is not present, the auto zoom operation is prohibited. As a result, the stability of auto zoom control is improved, and it is possible to reduce the sense of discomfort and discomfort of the photographer.
[Second Embodiment]
Next, auto zoom control (tracking control) in the second embodiment of the present invention will be described with reference to FIG. FIG. 6 is a flowchart showing a procedure of auto zoom control in the present embodiment. The face detection control process shown in the flow of FIG. 6 is executed according to a program stored in a memory in the computer 114, as in the first embodiment. In FIG. 6, steps S601 and S603 to S610 are the same as steps S201 and S203 to S210 of FIG.

  This embodiment is a video camera that performs tracking control when face detection cannot be performed by the face detection processing circuit 116 during the automatic zoom control of the first embodiment described with reference to FIGS. The present invention relates to (imaging device). The tracking control is executed by the computer 114, and a known process is applied. For example, there is a template matching method, and a partial image obtained by cutting out an image area to be tracked (in this embodiment, a face to be subjected to auto-zoom control and the face detected last) is used as a reference image (template image). sign up. Then, the luminance or color information in the reference image is used as a feature amount, an area having the highest correlation with the feature amount of the reference image in the image is estimated, and a specific subject (a face to be subjected to auto zoom control in this embodiment) is estimated. ).

  First, in step S620 in FIG. 6, it is determined whether or not the face to be subjected to auto zoom control is detected by the face detection processing circuit 116. If a face is detected in step S620, the process proceeds to step S621, and the position and size of the face on the screen to be subjected to auto zoom control are acquired from the face detection processing circuit 116. In step S621, the first subject detection circuit included in the face detection processing circuit 116 detects a subject (face) from a specific screen. On the other hand, if the face cannot be detected, the process proceeds to step S622. In step S622, it is determined whether or not the tracking control is being executed with the face to be subjected to auto zoom control as the tracking target. When the tracking control is not executed, the auto zoom control is not executed. On the other hand, when the tracking control is being executed, the process proceeds to step S623, and the position and size of the face in the tracking target area are acquired. In step S623, the second subject detection circuit included in the face detection processing circuit 116 detects a subject (face) from a plurality of time-series continuous screens. After step S603, the same processing as in the first embodiment described with reference to FIG. 2 is performed.

  Next, the movement monitoring process (step S605) in the present embodiment will be described in detail with reference to FIG. FIG. 7 is a flowchart showing the procedure of the movement monitoring process in the present embodiment. In FIG. 7, steps S701 to S705, S707, and S708 are the same as steps S301 to S305, S307, and S308 of FIG.

  When the movement monitoring process is started, the same processes (steps S702 to S705) as those in the first embodiment described with reference to FIG. 3 are performed. In step S731, it is determined whether or not the tracking control is being executed with the face to be subjected to auto zoom control as the tracking target. When the tracking control is not executed, the process proceeds to step S732, and an area smaller than the angle of view after the predetermined time by a predetermined first ratio is determined inside the angle of view after the predetermined time. Set as. On the other hand, when the tracking control is being executed, the process proceeds to step S733, and an area smaller by a predetermined second ratio than the angle of view after the predetermined time is determined inside the angle of view after the predetermined time. Area). Next, in step S734, based on the position and size of the face after the predetermined time calculated in step S705, the object of auto zoom control is determined for the above-described discrimination angle of view (predetermined region) at the angle of view after the predetermined time. It is determined whether or not a face is present. Then, processing similar to that in the first embodiment (steps S707 and S708) is performed according to the result of determining whether or not there is a face to be subjected to auto zoom control in the discrimination angle of view.

  Here, the region smaller by the second ratio is smaller than the region smaller by the first ratio. That is, the determination means is configured so that when the face is detected by the second subject detection circuit, the predetermined area (determination angle of view) of the screen is smaller than when the face is detected by the first subject detection circuit. The size of the predetermined area is changed. The area that is smaller by the first ratio may be an area of the entire angle of view after a predetermined time. FIG. 8 is an image diagram of face presence determination at an angle of view after a predetermined time. FIG. 8A shows a case where an area that is smaller by a first ratio than the angle of view after the predetermined time inside the angle of view after the predetermined time is set as the first discrimination angle of view 804 (for example, an area that is 15/16 times larger). Indicates. Further, in FIG. 8B, an area smaller by a second ratio than the angle of view after the predetermined time inside the angle of view after the predetermined time is set as the second discrimination angle of view 806 (for example, an area of 3/4 times). Show the case. Here, 802 and 807 have the same position and the same size, and 803 and 808 have the same position and the same size.

  In the case of FIG. 8A, the face position and the predicted size position 802 at the angle of view 801 after a predetermined time are included in the first discrimination angle of view 804, so it is determined that a face exists. On the other hand, the predicted position 803 of the face position and the size at the angle of view 801 after a predetermined time includes a predetermined area ratio (for example, an area of 3/4) in the first determination angle of view 804 in the first determination angle of view 804. It is determined that no face exists. On the other hand, in the case of FIG. 8B, the face position and the predicted size position 807 at the angle of view 805 after a predetermined time are included in the second discrimination angle of view 806, so it is determined that a face exists. On the other hand, the predicted position 808 of the face position and the size at the angle of view 805 after a predetermined time includes a predetermined area ratio (for example, 3/4 area) in the second determination angle of view 806 in the second determination angle of view 806. It is determined that no face exists.

  As described above, in the present embodiment, the face to be subjected to auto zoom control is subject to auto zoom control depending on whether the face is information obtained from the result of face detection or the information obtained from the execution result of tracking control. The discrimination angle of view for determining whether or not the face is within the angle of view after a predetermined time is changed. When tracking control is executed, the reliability of information on the position and size of a face to be subjected to auto zoom control is lower than the result of face detection. That is, in the tracking control, a position having a high matching correlation in the screen is determined as the tracking target region, so that there is a possibility that the background is erroneously tracked when an object similar to the background exists. In addition, when the face moves at various angles, the correlation of matching cannot be accurately grasped and is not always equal to the size of the face. For this reason, auto-zoom control cannot be executed stably, and the quality of the captured video image may deteriorate. Therefore, in the present embodiment, when tracking control is executed, the discrimination angle of view is set narrower than when face information is obtained by face detection. As a result, the execution of the auto zoom control is limited, the stability of the auto zoom control is improved, and it is possible to reduce the sense of discomfort and discomfort of the photographer.

  Note that the first embodiment and the second embodiment may be combined. At this time, in order to determine whether or not a face exists after a predetermined time from the movement amount of the face position in the first embodiment, the first determination angle of view in the second embodiment may be used. Furthermore, the second discrimination angle of view may be used depending on whether face detection has been performed or whether tracking control has been performed.

  As mentioned above, although preferable embodiment of this invention was described, this invention is not limited to these embodiment, A various deformation | transformation and change are possible within the range of the summary.

  For example, a software program for realizing the functions of the above-described embodiments is supplied from a recording medium to a system or apparatus (imaging apparatus) having a computer that can execute the program directly or using wired / wireless communication. Is also included in the present invention. Therefore, the program code itself supplied to and installed in this computer in order to realize the functional processing of the present invention by the computer, that is, the computer program itself for realizing the functional processing of the present invention is also included in the present invention. Note that as long as it has a function of a program, the form of the program is not limited, such as an object code, a program executed by an interpreter, and script data supplied to the OS.

  As the recording medium for supplying the program, for example, a magnetic recording medium such as a hard disk or a magnetic tape, an optical / magneto-optical storage medium, or a nonvolatile semiconductor memory is used. As a program supply method, a computer program that forms the present invention is stored in a server on a computer network, and a connected client computer downloads and installs the computer program.

102 Variable magnification lens 106 Image sensor 110 Zoom drive source 114 Camera / AF microcomputer 116 Face detection processing circuit

Claims (7)

An image sensor that photoelectrically converts a subject image formed by an imaging optical system including a zoom lens and outputs an image signal; and
Subject detection means for detecting a main subject based on the image signal output from the image sensor;
Zoom control means for performing zooming operation by driving the zoom lens and performing auto zoom control so that the main subject has a constant size on the screen;
An image pickup apparatus comprising: control means for stopping the auto zoom control when a change amount of the main subject on the screen is larger than a predetermined value. The control means includes
Calculating means for monitoring the position and size of the main subject on the screen and calculating the position and size of the main subject after a predetermined time;
Determination means for determining whether a predetermined ratio of the main subject after the predetermined time exists within a predetermined area of the screen based on a calculation result of the calculating means;
The imaging apparatus according to claim 1, further comprising: a stopping unit that stops the automatic zoom control when it is determined that the predetermined ratio of the main subject does not exist within a predetermined area of the screen. A storage unit for storing a history of the position and size of the main subject detected by the subject detection unit and information on the position of the zoom lens;
The calculating means includes
An angle-of-view calculating unit that calculates an angle of view after the predetermined time based on the position of the zoom lens and the zoom speed;
A subject position calculation unit that calculates the position and size of the main subject at the angle of view after the predetermined time calculated by the angle of view calculation unit from the information stored in the storage unit;
The determination means determines whether or not a predetermined ratio of the main subject exists within a predetermined area of the screen based on calculation results of the angle of view calculation unit and the subject position calculation unit. The imaging device according to claim 2. The calculation means includes a movement amount calculation unit that calculates a movement amount of the main subject from a history of the position of the main subject calculated by the subject position calculation unit;
The determination means changes the size of the predetermined area so that the predetermined area of the screen becomes smaller as the movement amount of the main subject calculated by the movement amount calculation unit is larger. 3. The imaging device according to 3. The subject detection means includes
First subject detection means for detecting the subject from a specific screen; and second subject detection means for detecting the subject from a plurality of time-series continuous screens;
The determination unit is configured so that when the subject is detected by the second subject detection unit, the predetermined area of the screen is smaller than when the subject is detected by the first subject detection unit. The imaging apparatus according to claim 2, wherein the size of the predetermined area is changed. Photoelectrically convert the subject image formed by the imaging optical system including the zoom lens and output an image signal;
A main subject is detected based on the image signal,
The zoom lens is driven to perform a zooming operation, and auto zoom control is performed so that the main subject has a certain size on the screen,
An auto-zoom method, wherein the auto-zoom control is stopped when a change amount of the main subject on the screen is larger than a predetermined value. A program that is executed by an imaging apparatus that detects a main subject based on an image signal output by photoelectrically converting a subject image and performs auto zoom control so that the main subject has a certain size on the screen. ,
A program for stopping the auto-zoom control when a change amount of the main subject on the screen is larger than a predetermined value.

Images