JP2006101186A - Camera - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To make a specific subject to be tracked selectable with easy operation at photographing. <P>SOLUTION: The camera comprises an imaging part, a display part and a detection part. The imaging part photographs an optical image formed by a photographic lens and generates a video signal of a moving image. The display part has a monitor and displays the video signal on the monitor as an image. Further, the display part is a touch panel type and selects a specified portion, corresponding to a location touched on the monitor from the image displayed on the monitor. The detection part compares the specific portion with an image expressed by a video signal, generated after selecting the specific portion and detects an image part, corresponding to the specific portion from the video signal. Namely, a subject to be traced can be selected only by touching the monitor, the subject can be selected by easy operation at photographing, without requiring complicated operations. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a camera, particularly a video camera. Specifically, the present invention relates to a technique for detecting a subject that is a person by face recognition processing and tracking the subject.

  In many images taken by a video camera, a moving person is the main subject. Accordingly, it has been desired to detect and track the position of a moving person, particularly the face, and automatically adjust the exposure condition and focus to the person. In view of this, the video camera disclosed in Patent Document 1 attempts to detect a human face by detecting a hair or skin color region from an image, and to perform autofocus control following the detected face position.

Details of processing for detecting a person's face from an image are also disclosed in Patent Literature 2, Patent Literature 3, and the like.
JP-A-6-30318 JP 2001-309225 A JP 2004-128715 A

  In actual shooting, in addition to the specific person to be tracked, it is assumed that other unspecified persons are also shown. However, even though a conventional video camera can track a person, it cannot detect and track a specific person from a plurality of persons and cannot adjust the exposure condition and focus to the specific person. Therefore, there has been a demand for enabling detection and tracking of only a specific person. At that time, it is desirable that the person to be tracked can be selected by the user at the time of shooting by an easy operation.

An object of the present invention is to provide a technique for detecting and tracking only a specific person and adjusting the exposure condition or focus to that person.
Another object of the present invention is to provide a technique that enables a specific subject to be tracked to be selected at the time of photographing by an easy operation.

  The invention of claim 1 includes an imaging unit that captures a light image formed by the photographing lens and generates a video signal of a moving image, and a display unit that has a monitor and displays the video signal as an image on the monitor. Camera. The invention of this claim is characterized by the following points. 1stly, a display part is a touch panel type, and extracts the specific site | part corresponding to the location where contact operation was performed on the monitor from the display image of a monitor. Second, the image processing apparatus includes a detection unit that compares the specific part with an image represented by the video signal generated after the specific part is extracted, and detects an image part corresponding to the specific part from the video signal.

According to a second aspect of the present invention, in the camera of the first aspect, the detection unit detects an image portion corresponding to the specific part by at least one of face recognition processing, clothing pattern recognition, and voice recognition. To do.
According to a third aspect of the present invention, in the camera of the first or second aspect, the subject corresponding to the specific part is tracked within the angle of view of the photographic lens based on the detection result of the detection unit, and the exposure condition and the photographic lens It has a tracking unit for adjusting at least one of the focus to a subject to be tracked.

According to a fourth aspect of the present invention, in the camera of the third aspect, when the subject to be tracked is out of the angle of view, the display unit displays information on the position of the subject to be tracked on the monitor.
According to a fifth aspect of the present invention, the camera according to any one of the first to fourth aspects includes a storage unit for registering and storing a specific part.

  According to a sixth aspect of the present invention, the camera includes an imaging unit, a detection unit, and a tracking unit. The imaging unit captures a light image formed by the photographic lens and generates a video signal of a moving image. The detection unit compares a specific image registered in advance with an image represented by the video signal, and detects an image portion corresponding to the specific image from the video signal. The tracking unit tracks the subject corresponding to the specific image within the angle of view of the photographic lens based on the detection result of the detection unit, and matches at least one of the exposure condition and the focus of the photographic lens with the subject to be tracked.

The invention according to claim 7 is the camera according to claim 6, wherein the detection unit detects an image portion corresponding to the specific image by at least one of face recognition processing, clothing pattern recognition, and voice recognition. To do.
In the camera of any one of claims 3, 4, 6, and 7, when the subject to be tracked deviates from the angle of view, the exposure condition that fits the entire angle of view is achieved. The tracking unit changes the exposure condition step by step so as to approach.

According to a ninth aspect of the present invention, in the camera according to any one of the third, fourth, sixth, and seventh aspects, when the subject to be tracked is out of the field angle of the photographing lens, the tracking portion is exposed. Is switched to a condition suitable for the entire angle of view.
According to a tenth aspect of the present invention, in the camera of any one of the third, fourth, sixth, and seventh aspects, when the subject to be tracked is out of the field angle of the photographing lens, the subject is within the field angle. The tracking unit is characterized in that it does not change the exposure condition before and after being deviated from.

In one embodiment of the present invention, a video signal is displayed as an image on a monitor, and a specific part corresponding to a place where a contact operation is performed on the monitor is extracted from the display image, and then a subject corresponding to the specific part is newly generated. Detected from the video signal. Accordingly, the selection of the subject to be tracked can be executed at the time of shooting by a simple operation without complicated operation.
In another embodiment of the present invention, the person within the angle of view is not detected only by face recognition processing, but the specific image is compared with the image represented by the video signal, and the image portion corresponding to the specific image is detected as the video signal. Detect from. Therefore, if an image that can identify an individual such as a face image is registered in advance as a specific image, only a specific person can be detected separately from other persons. Since the subject corresponding to the specific image is tracked based on the detection result, the exposure condition or the focus can be adjusted to the specific person in distinction from other persons.

Hereinafter, embodiments of the present invention will be described in the order of configuration, operation, effect, and supplementary matter with reference to the drawings.
<Configuration of this embodiment>
FIG. 1 shows a block diagram of the video camera of this embodiment. As shown in the figure, the video camera 8 includes a photographing lens 12, an image sensor 20, an analog signal processing unit 24, an A / D conversion unit 28, a timing generator 32, a focus control unit 36, and a sound detection unit. 38, an operation unit 40, an MPU (Micro Processor Unit) 44, a system bus 48, an image processing unit 52, a memory 56, a registration memory 60, a recording / reading unit 64, and an exchangeable recording medium 68. And a monitor control unit 84 and a touch panel type monitor 88.

The imaging element 18 is configured by arranging a large number of pixels that selectively transmit any one of red light, green light, and blue light in, for example, a Bayer square arrangement, and is capable of moving image shooting.
The MPU 44 performs system control of the video camera 8.
The operation unit 40 includes a button group for setting a power button, a moving image pause button, a shooting mode, and the like (not shown).

The main features of this embodiment are that it has four shooting modes of a first tracking mode, a second tracking mode, a third tracking mode, and a fourth tracking mode, and that the monitor 88 is a touch panel type. One of the features of these four shooting modes is that a specific person can be detected and tracked separately from other persons.
In the first tracking mode, during the moving image display, the moving image is temporarily stopped to be displayed as a still image, the image area touched on the monitor 88 is set as an image of the subject to be tracked, and tracking is performed based on the image. In the second tracking mode, the image area touched on the monitor 88 during moving image display is used as the image of the subject to be tracked, and tracking is performed in the same manner. In the third tracking mode, the video in the recording medium 68 is reproduced on the monitor 88, and the image area touched on the monitor 88 is set as the image of the subject to be tracked. In the fourth tracking mode, any one of the specific images already registered before shooting is selected and used as the image of the subject to be tracked.

FIG. 2 shows a display example on the monitor 88. The subject to be tracked is selected by touching the face area of the subject on the display image of the monitor 88 with a finger as shown in FIG. Alternatively, it may be performed by tracing the outline of the face with a finger. If the face cannot be recognized for the area touched by the user due to the small area of the face in the image or the like, the fact that the face could not be recognized and the cause is monitored as shown in FIG. When the video camera 8 detects the subject within the angle of view after the subject to be tracked is selected, as shown in FIG. In the display image, for example, a red cross is superimposed and displayed on the face of the person to be tracked. Further, another feature of the present embodiment is that, when the subject to be tracked is out of the angle of view, the position information of the subject is displayed on the monitor 88 as shown in FIG. In this embodiment, as an example, the direction in which the subject to be tracked is considered to be present is displayed with an arrow.

<Description of operation of this embodiment>
The video camera 8 of this embodiment performs a power-on process when the power button is turned on. Thereafter, the user can set the shooting mode and the like by operating the button group of the operation unit 40. Hereinafter, the operation will be described in order for the case where one of the first to fourth tracking modes described above is set. Actually, other known shooting modes may be set, but the other shooting modes are not directly related to the present application, and thus the description thereof is omitted.

FIGS. 3, 4, 5, and 6 are flowcharts showing the operation of the video camera 8 in the first tracking mode, the second tracking mode, the third tracking mode, and the fourth tracking mode, respectively. First, the first tracking mode will be described according to the step numbers shown in FIG.
[Step S1] The user records the audio of the person to be tracked on the video camera 8. The MPU 44 stores the sound detected by the sound detection unit 38 in the registration memory 60 as the sound of the person to be tracked.

  [Step S2] The voice detection unit 38 starts voice detection. Further, the image pickup device 20 is repeatedly exposed and repeatedly accumulates and discharges charges, and continuously outputs a video signal at a predetermined frame rate (for example, 30 frames / second). This video signal is subjected to clamp processing and sensitivity correction processing by the analog signal processing unit 24 and then converted to a digital video signal by the A / D conversion unit 28. The video signal is subjected to white balance adjustment, color process processing, and the like by the image processing unit 52 and then displayed as a moving image on the monitor 88.

[Step S3] When the moving image pause button is turned on, in synchronization with it, the MPU 44 fixes the display image on the monitor 88 to the one displayed at that time. Note that the video signal generation operation by the exposure of the image sensor 20 may be continuously performed. Thereafter, the process proceeds to step S4. If the video pause button is not turned on, the MPU 44 waits.
[Step S4] The display image on the monitor 88 is stationary, and the user touches the face area of the subject person to be tracked on the monitor 88 (see FIG. 2A). When the user touches to trace a specific area, the MPU 44 extracts an image corresponding to the traced area (an image displayed in the traced area) via the monitor control unit 84. When the user touches only one point, the MPU 44 extracts an image so that the skin color portion and the black portion (corresponding to hair) are included around the touched portion.

[Step S5] The MPU 44 determines whether or not a feature portion appears in the image region extracted in step S4 to such an extent that it can be recognized as the face of a specific person. This determination is performed, for example, by performing face recognition processing and analyzing feature points when face recognition is possible.
In the face recognition processing, facial feature points such as lips, eyes, and eyebrows are searched for in the extracted image region. The position of the lips can be determined, for example, by detecting an area slightly redder than the surroundings. The position of the eyes can be determined, for example, by detecting a black round part corresponding to the pupil and white parts on both sides thereof. The position of both eyebrows can be determined, for example, by detecting two areas slightly blacker than the surroundings above the eyes. Then, an area that includes these facial feature points on the inside and can be regarded almost as a skin color is detected, and thereby the position of the face in the image and the area of the face area are determined.

  When the above-described face recognition can be performed, the combination of the coordinates of the facial features such as the ends of the eyebrows, the ends of both eyes, and the ends of the lips, the outline of the face, and the colors of the cheeks and forehead are determined. However, if the degree of left-right asymmetry is large in the obtained combination of coordinates of the characteristic portion, the determination is assumed to have failed. This may be because the face was not sufficiently facing the video camera 8. If the determination is successful, the process proceeds to step S7 on the assumption that the characteristic portion has appeared sufficiently clearly. If face recognition could not be performed, or if face recognition could be performed but feature portions could not be determined, the process proceeds to step S6.

  [Step S6] The MPU 44 determines the reason why the face recognition or the feature portion could not be determined in step S5. Causes include a small face area in the image and a face that is not sufficiently facing the video camera 8. Then, the MPU 44 displays the fact that the face recognition or the feature portion could not be determined and the cause thereof on the monitor 88 as shown in FIG. 2B (for example, “increase the area of the face area”). “Please make it more front-facing”). This prompts the user to re-execute the operations in steps S3 and S4. Thereafter, the process returns to step S3.

  [Step S7] The MPU 44 identifies the person to be tracked based on the combination of the coordinates of the facial feature determined in Step S5, the outline of the face, the colors of the cheeks and forehead, and the image area extracted in Step S4. Information is stored in the registration memory 60. Thereafter, the moving image shooting similar to step S2 and the moving image display on the monitor 88 are resumed. In addition, the user points the video camera 8 toward a specific person. Thereafter, when a user's trigger operation is performed until step S10 described later, the MPU 44 causes the recording / reading unit 64 to start recording a moving image and audio on the recording medium 68 in an interrupt process.

  [Step S8] The MPU 44 performs the aforementioned face recognition processing on the latest frame in the generated video signal to detect a face area. If a face area is detected, an image of the face portion is extracted and then the process proceeds to step S9. If not detected, the same process is performed on the newly generated latest frame. Note that face detection need not be performed on all generated frames, and may be, for example, 1 frame / second.

[Step S <b> 9] The MPU 44 determines whether the face area extracted in Step S <b> 8 matches the identification information of the tracking target person stored in the registration memory 60. Specifically, if the combination of the coordinates of the feature portion, the face outline, and the colors of the cheeks and the forehead match at a predetermined ratio or more, it is determined that they are the same person.
When the match determination based on the face area image is delicate, the MPU 44 determines whether or not the sound detected by the sound detection unit 38 at the time of determination matches the sound of the person to be tracked registered in step S1. If the voices match, it is determined that they are the same person. The coincidence of voices may be determined, for example, based on the sound frequency. If it is determined that the person is the same person, it can be determined that the main subject is within the angle of view of the photographic lens 12, so the process proceeds to step S10. If it is not determined that the person is the same person, the process returns to step S8.

  [Step S10] The MPU 44 informs the user that the person to be tracked has been detected by performing the display as shown in FIG. 2C, and starts the tracking operation. Specifically, the MPU 44 and the focus control unit 36 focus the photographing lens 12 on the face of the person to be tracked based on the position of the face area extracted in step S9 within the angle of view. Further, the MPU 44 sets exposure conditions such as an aperture value and a shutter speed so that the face of the person to be tracked has a proper exposure. After that, the operation of detecting the face of the person to be tracked from the newly generated latest frame and updating the focus position and the exposure condition as needed is repeated until the end of shooting.

In the present embodiment, the imaging sensitivity (ISO value of the imaging device 20) in the exposure condition is fixed at 100, for example, but may be changed. In the present embodiment, as an example, when recording is not started, the recording and recording of the moving image and the sound on the recording medium 68 are started. However, the recording may not be performed unless the user performs a trigger operation.
[Step S11] A face area is detected with respect to the newly generated latest frame. When the face area is detected, it is determined whether or not the face area matches the tracking target. If they match, after adjusting the focus and exposure conditions for the extracted face area within the angle of view, the process proceeds to step S14. In step S14, the process returns to step 11 as long as there is no instruction to end photographing. Accordingly, the focus position and the exposure condition are updated at predetermined time intervals.

  This update interval, that is, after detecting the face area for the latest frame in step S11, returns to step S11 again via step S14, and the interval for detecting the face area is the frame of the image sensor 20. It does not have to be the same as the rate interval, and may be 1 second, for example. If no face area is detected in step S11, or if it does not match the tracking target even if it is detected, it is determined that the person to be tracked is out of the angle of view, and the process proceeds to step S12. Even if the person to be tracked is out of the angle of view, shooting and recording on the recording medium 68 are continued.

  [Step S12] The MPU 44 detects the face area of the person to be tracked from several frames immediately before the latest frame. The MPU 44 compares the position of the detected face area within the angle of view for each frame, and determines the direction in which the person is moving. Then, the MPU 44 displays the direction superimposed on the image on the monitor 88 with an arrow, for example (see FIG. 2D described above).

  [Step S13] The MPU 44 calculates an exposure condition in which the entire angle of view is properly exposed, and changes the exposure condition step by step from the current one so that the calculated exposure condition is obtained. This does not mean that the exposure condition calculation target area is gradually expanded from the tracking target face area to the entire screen, but each parameter of the exposure condition so that it gradually approaches the condition where the entire angle of view is properly exposed. Means to change it in several times.

  For example, if the exposure condition is {aperture 2, shutter speed 1/60 sec} during tracking and the exposure condition calculated here is {aperture 16, shutter speed 1/200 sec}, first the exposure condition is changed. For example, it is changed to {aperture 4, shutter speed 1/100 seconds}. Thereafter, after changing to {aperture 8, shutter speed 1/150 seconds}, the exposure condition is changed to the calculated exposure condition. Thereafter, the process returns to step S8.

[Step S14] If an instruction to end shooting is not given, the process returns to step S11 to continue tracking. When an instruction to end shooting is given, shooting and recording are ended. The above is the description of the operation in the first tracking mode.
Next, the second tracking mode will be described according to the step numbers shown in FIG.
[Step S31] As in step S2, voice detection, moving image shooting, and moving image display on the monitor 88 are started. In the second tracking mode, these processes are continuously performed until the end of shooting.

  [Step S32] The user touches the face area of the subject person to be tracked on the monitor 88 on which the moving image is displayed. The MPU 44 extracts an image corresponding to the touched or traced area as in step S4. In addition, the MPU 44 stores, in the registration memory 60, the sound detected by the sound detection unit 38, for example, for about 5 seconds before and after the touched point, as the sound of the person to be tracked.

  [Step S33] The face recognition process similar to that in step S5 is performed on the image region extracted in step S32. If the face recognition is completed, the feature portion is determined. If the determination of the combination of the coordinates of the feature portion, the face outline, and the color such as the cheek or forehead is successful, the process proceeds to step S35. If face recognition could not be performed, or if face recognition could be performed but feature portions could not be determined, the process proceeds to step S34.

[Step S34] Similar to step S6, the face recognition or the feature portion cannot be determined and the cause thereof are displayed on the monitor 88. Further, the MPU 44 deletes the voice stored in the registration memory 60 as the voice of the person to be tracked in step S32. Thereafter, the process returns to step S32.
[Step S35] The MPU 44 identifies the person to be tracked based on the combination of the coordinates of the facial feature determined in Step S33, the face outline, the color of the cheeks and the forehead, and the image area extracted in Step S32. Information is stored in the registration memory 60. If the user performs a trigger operation thereafter, recording of moving images and audio on the recording medium 68 is started in an interrupt process. The subsequent processes in steps S36 to S42 are processes for detecting a tracking target person and adjusting the focus and exposure conditions thereto, and are the same as steps S8 to S14 in the first tracking mode except for the difference in step numbers. is there. The above is the description of the operation in the second tracking mode.

Next, the third tracking mode will be described according to the step numbers shown in FIG.
[Step S61] The MPU 44 causes the recording / reading unit 64 to read the video signal recorded on the recording medium 68, displays the moving image on the monitor 88, and reproduces the sound from a sound output unit (not shown).
[Step S62] When an image of a person to be tracked is displayed, the user selects the face area on the monitor 88. The MPU 44 extracts an image in the touched or traced area in the same manner as in step S32, and registers, for example, a voice of about 5 seconds before and after the touched point as the voice of the person to be tracked. Store in the memory 60. Since it is not a moving image display while shooting, but playback, it is possible to select a still image display by pressing a pause button (not shown) or the like.

[Step S63] The face recognition process similar to that in Step S5 is performed on the image region extracted in Step S62, and when face recognition is completed, the feature portion is determined. If this determination is successful, the process proceeds to step S65. If face recognition could not be performed, or if face recognition could be performed but feature portions could not be determined, the process proceeds to step S64.
[Step S64] Similar to step S34, the fact that face recognition or feature portion determination has failed and the cause thereof are displayed on the monitor 88, and the voice of the person to be tracked is recorded in the registration memory 60 in step 62. Erase the stored voice. Thereafter, the process returns to step S62.

[Step S65] The MPU 44 identifies the person to be tracked by the combination of the coordinates of the facial feature determined in Step S64, the face outline, the color of the cheeks and the forehead, and the image area extracted in Step S62. Information is stored in the registration memory 60.
[Step S66] In the same manner as in step S2, exposure of the image sensor 20 is started, and moving image display on the monitor 88 is started. In addition, the sound detection unit 38 starts detecting sound. Thereafter, when the user performs a trigger operation, recording of a moving image and audio on the recording medium 68 is started in an interrupt process. Subsequent steps S67 to S73 are the same as steps S8 to S14 in the first tracking mode except for the difference in the step numbers. The above is the description of the operation in the third tracking mode.

  Next, the fourth tracking mode will be described. In the fourth tracking mode, the registration memory 60 includes a plurality of person identification information (a face image, a combination of coordinates of facial features, a color of a face outline, a cheek, a forehead, etc.), and a voice ID number. Thus, each person is distinguished and registered in advance. For example, as in the second tracking mode, the registration method is such that a specific person is photographed and displayed on the monitor 88, face recognition and feature part determination are performed on the image part touched on the monitor 88, and the sound together with the determination result May be stored in the registration memory 60. Alternatively, the video in the recording medium 68 may be reproduced on the monitor 88 and registered in the same manner as in the third tracking mode.

Hereinafter, the process flow of the fourth tracking mode will be described according to the step numbers shown in FIG.
[Step S91] The MPU 44 displays on the monitor 88 face images of a plurality of persons registered in the registration memory 60.
[Step S92] The user touches on the monitor 88 the face image of the person to be tracked. The MPU 44 sets a person corresponding to the touched face image as a tracking target.

  [Step S93] As in step S2, exposure of the image sensor 20 is started, and moving image display on the monitor 88 is started. In addition, the sound detection unit 38 starts detecting sound. If the user performs a trigger operation thereafter, recording of moving images and audio on the recording medium 68 is started in an interrupt process. The subsequent steps S94 to S100 are the same as steps S8 to S14 in the first tracking mode except for the difference in the step numbers. The above is the description of the operation in the fourth tracking mode.

<Effect of this embodiment>
In the first tracking mode, when the moving image pause button is turned on during moving image shooting before tracking and recording, the display image on the monitor 88 is fixed to the one displayed at that time. The user can select a person as a tracking target simply by touching the face portion of the person to be tracked in the still image on the monitor 88. Therefore, since there is no possibility of missing timing or complicated operations at the time of selection, a person to be tracked can be selected at the time of shooting by an easy operation.

  Based on the face image of the person to be tracked, the combination of the coordinates of the facial features, the face outline, and the color of the cheek and forehead, etc. are obtained, and whether or not these identification information corresponds to the person to be tracked It is used for the match judgment. Since these pieces of identification information usually differ from person to person, even if a plurality of persons are present in the angle of view, only a specific person as a tracking target can be detected from them. Therefore, only a specific person can be tracked and distinguished from other persons, and the exposure condition and focus can be adjusted to that person.

When the match determination based on the identification information based on the face image is delicate, the determination is made by comparing the voice of the person to be tracked registered in advance with the voice detected at the match determination. Therefore, for example, even when the person to be tracked leaves the video camera 8 and the face image becomes small and the match determination becomes delicate, the determination can be made reliably.
When the tracking target person deviates from the angle of view, motion detection is performed from a plurality of frame images immediately before deviating, and the moving direction of the person is determined. Based on this direction, the direction of the current position of the person to be tracked is displayed on the monitor 88. Therefore, the user can place the person to be tracked within the angle of view again simply by pointing the video camera 8 in the displayed direction.

In addition, when the tracking target person deviates from the angle of view, the exposure condition is changed step by step so as to approach the exposure condition suitable for the entire angle of view. Therefore, there is no possibility that the brightness of the screen changes rapidly in the recorded video, and the video can be recorded as a good video.
In the second pairing mode, the operation is the same as in the first tracking mode, except that the moving image display is continued without switching to the still image display when the person to be tracked is selected. When the movement of the person to be tracked is relatively slow, the face image can be touched on the monitor 88 without missing the timing even in the moving image display state. Accordingly, the same effect as that of the first tracking mode described above can be obtained in the second tail mode.

  In the third pairing mode, the same operation as in the first tracking mode is performed except that the video in the recording medium 68 is reproduced on the monitor 88 and a person to be tracked is selected from the video. Therefore, the same effect as that of the first tracking mode described above can be obtained in the third tail mode. In this mode, it is not necessary to photograph the person when the person to be tracked is selected. Therefore, for example, it is effective when it is desired to detect a specific person in a security camera but it is difficult to photograph the person in advance.

  In the fourth tracking mode, the same operation as in the first tracking mode is performed except that a tracking target person is selected from face images of a plurality of persons registered in advance. Thus, the same effect as in the first tracking mode is obtained. It is done. Also, in the fourth tracking mode, it is not necessary to photograph the person when the tracking target person is selected, as in the third pairing mode. Therefore, the same effect as the third tracking mode can be obtained.

<Supplementary items of this embodiment>
In the first to fourth tracking modes, the example in which the exposure condition is changed stepwise when the tracking target person deviates from the angle of view has been described. The present invention is not limited to such an embodiment. The exposure condition may not be changed when the tracking target person falls outside the angle of view. This is because if there is a high possibility that the tracking target person will soon return to the angle of view, it is better to leave the exposure conditions according to the tracking target person. Alternatively, when the person to be tracked is out of the angle of view, the exposure setting may be switched at once so that the exposure setting matches the entire angle of view.

  An example has been described in which the recording on the recording medium 68 is not interrupted when the tracking target person falls outside the angle of view. The present invention is not limited to such an embodiment. Recording may be paused at the moment when the tracking target person falls outside the angle of view, and the recording may be resumed when the tracking target person enters the angle of view again. That is, only the frame in which the person to be tracked is shown may be recorded. By doing so, it is possible to save the recording time and to efficiently perform the work when extracting the image from the frame image in which the person to be tracked is shown.

  An example in which only one registered subject is detected and tracked has been described. The present invention is not limited to such an embodiment. For example, a plurality of persons may be registered in advance with a priority order, and the plurality of persons may be set as tracking targets. In that case, when only one person is detected within the angle of view, only that person is tracked, and when a plurality of persons are detected, the person with the highest priority may be tracked.

In the second tracking mode, an example has been described in which voice is also registered at the same time when a face image of a person to be tracked is selected (step S32). The present invention is not limited to such an embodiment. The voice registration may be performed separately from the selection of the face image, for example, before step S31.
In the first to fourth tracking modes, the example in which both the exposure condition and the focus are adjusted for the person to be tracked has been described. However, only one of the exposure condition and the focus may be adjusted.

As an example of informing the user that the tracking target person has been detected within the angle of view, an example in which superimposed display as shown in FIG. 2C is performed has been described. The present invention is not limited to such an embodiment. For example, a method in which a lamp is provided as one of the operation units 40 and the lamp blinks may be used. Since recording has started, it is not desirable to notify by sound.
Although the example of applying the technique for tracking a specific person according to the present invention and the technique for selecting a person to be tracked by a touch panel at the time of shooting to a video camera has been described, the present invention is also applied to other cameras such as a digital still camera. Is possible. Further, the tracking target can be carried out even if it is another animal that is not a person.

The example in which the image area corresponding to the part directly touched by the user on the monitor 88 is extracted has been described. The present invention is not limited to such an embodiment. For example, a touch pen or the like may be used to extract an image area corresponding to a portion where the tip of the pen contacts on the monitor surface.
An example has been described in which a specific person is detected from frames of continuously generated moving images. The present invention is not limited to such an embodiment. A person to be tracked may have a wireless transmission remote controller, and the person may be detected based on radio waves transmitted from the remote controller.

In the first to fourth tracking modes, an example has been described in which the determination is made by voice when the matching determination between the detected face image and the person to be tracked is delicate. The present invention is not limited to such an embodiment.
Audio may not be used as a determination criterion, but may be determined only from identification information based on a face image.
Alternatively, the face image and identification information of the person to be tracked may not be registered, only the sound may be registered, and the person to be tracked may be detected only from the sound. Specifically, for example, the direction of the sound source is determined from the detected sound, and when a face image is detected in that direction, it may be determined as a person to be tracked.

Alternatively, face images and voices are not registered, images of the clothes of the person to be tracked are extracted, the colors and patterns of the clothes are recognized, and the match determination of the person to be tracked is performed only with clothes. Good.
Alternatively, the face image, the identification information based on the face image, the clothing image, and the pattern recognition result for the clothing image may be registered without registering the voice. In this case, for example, when the match determination is ambiguous only with the identification information based on the face image, the determination may be made based on clothes.

Alternatively, the identification information based on the face image may not be used, and the matching determination of the person to be tracked may be performed based on the clothes pattern recognition result and the voice.
Alternatively, the identification information based on the face image, the pattern recognition result of the clothes, and the voice may be used for the match determination.
The correspondence between the claims and the present embodiment is as follows, for example.

The image pickup unit described in the claims includes the photographing lens 12, the image pickup device 20, the analog signal processing unit 24, the A / D conversion unit 28, the timing generator 32, the focus control unit 36, and the sound detection unit 38, and generates a video signal therein. This corresponds to the function of the MPU 44 to be executed. The details of the operation for generating the video signal are well known, and the description thereof is omitted.
The display unit described in the claims corresponds to the monitor 88 and the functions of the monitor control unit 84 and the MPU 44 that cause the monitor 88 to display an image.

The detection unit recited in the claims corresponds to the function of the MPU 44 that detects a subject corresponding to the face image of a person set as a tracking target within the angle of view.
The tracking unit described in the claims is a function of the MPU 44 that controls each unit of the video camera 8 to track the subject corresponding to the face image of the person set as the tracking target within the angle of view and to adjust the exposure condition and the focus. Correspond.

The specific portion described in the claims corresponds to the image region extracted in steps S4, S32, and S62 in the first to third tracking modes.
The specific image described in the claims corresponds to the face image registered in the registration memory 60 in the fourth tracking mode.
The storage unit described in the claims corresponds to the registration memory 60 and the function of the MPU 44 that stores the identification information of the tracking target person in the registration memory 60.

  As described in detail above, the present invention can be used greatly in the field of video cameras.

It is a block diagram of the video camera of this embodiment. It is explanatory drawing which shows the example of a display of a monitor. It is a flowchart which shows the operation | movement in the 1st tracking mode in the video camera of this embodiment. It is a flowchart which shows the operation | movement in the 2nd tracking mode in the video camera of this embodiment. It is a flowchart which shows the operation | movement in the 3rd tracking mode in the video camera of this embodiment. It is a flowchart which shows the operation | movement in the 4th tracking mode in the video camera of this embodiment.

Explanation of symbols

8 Video Camera 12 Shooting Lens 20 Image Sensor 24 Analog Signal Processing Unit 28 A / D Conversion Unit 32 Timing Generator 36 Focus Control Unit 38 Audio Detection Unit 40 Operation Unit 44 MPU
48 system bus 52 image processing unit 56 memory 60 registration memory 64 recording / reading unit 68 recording medium 84 monitor control unit 88 monitor

Claims (10)

An imaging unit that captures a light image formed by the photographing lens and generates a video signal of a moving image;
A camera having a monitor and a display unit for displaying the video signal as an image on the monitor,
The display unit is a touch panel type, and extracts from the display image of the monitor a specific part corresponding to the place where the contact operation is performed on the monitor,
A camera having a detection unit that compares the specific part and an image represented by the video signal generated after the specific part is extracted, and detects an image part corresponding to the specific part from the video signal. . The camera of claim 1,
The detection unit detects an image portion corresponding to the specific part by at least one of face recognition processing, clothing pattern recognition, and voice recognition. The camera according to claim 1 or 2,
Based on the detection result of the detection unit, the subject corresponding to the specific part is tracked within the angle of view of the photographing lens, and at least one of the exposure condition and the focus of the photographing lens is matched to the subject to be tracked A camera characterized by having a part. The camera according to claim 3.
When the subject to be tracked is out of the angle of view, the display unit displays information on the position of the subject to be tracked on the monitor. The camera according to any one of claims 1 to 4,
A camera having a storage unit for registering and storing the specific part. An imaging unit that captures a light image formed by the photographing lens and generates a video signal of a moving image;
A detection unit that compares a specific image registered in advance with an image represented by the video signal and detects an image portion corresponding to the specific image from the video signal;
Based on the detection result of the detection unit, the subject corresponding to the specific image is tracked within the angle of view of the photographing lens, and at least one of the exposure condition and the focus of the photographing lens is matched to the subject to be tracked And a camera. The camera according to claim 6, wherein
The detection unit detects an image portion corresponding to the specific image by at least one of face recognition processing, clothes pattern recognition, and voice recognition. The camera according to any one of claims 3, 4, 6, and 7,
When the subject to be tracked is out of the angle of view, the tracking unit changes the exposure condition step by step so as to approach the exposure condition that fits the entire angle of view. The camera according to any one of claims 3, 4, 6, and 7,
When the subject to be tracked deviates from within the angle of view of the photographing lens, the tracking unit switches the exposure condition to an appropriate condition within the entire angle of view. The camera according to any one of claims 3, 4, 6, and 7,
When the subject to be tracked is out of the field angle of the photographing lens, the tracking unit does not change the exposure condition before and after the subject is out of the field angle.

Images