JP5153211B2 - Imaging apparatus and control method thereof - Google Patents

Description

  The present invention relates to an imaging apparatus and a control method thereof, and more particularly to an imaging apparatus capable of performing personal recognition using face detection and a control method thereof.

  2. Description of the Related Art Conventionally, an apparatus that performs personal authentication using face detection is known (see, for example, Patent Document 1). Since complicated processing is required to perform personal authentication, in Patent Document 1, an instruction to place a face in a predetermined area is issued, and processing is performed only on the face in the area, thereby reducing processing time. I am trying.

  As another technique for recognizing a specific individual, for example, there is one described in Patent Document 2. According to Patent Document 2, a feature unique to a person is extracted from a photographed image of a specific individual to be recognized and registered in advance as feature information, and the feature of the subject extracted from the photographed image is the registered feature information. And personal identification.

Japanese Patent Laid-Open No. 06-295331 JP 2004-320287 A

  However, the apparatus described in Patent Document 1 is unsuitable for application in which a person is photographed with a free composition like a digital camera in order to limit the detection area.

  On the other hand, when the detection area is not limited as in Patent Document 2, when immediacy is required as in the case of shooting with a digital camera, the time required for the recognition process becomes an issue. For example, the process of recognizing an individual is performed. It becomes necessary to simplify. However, if the recognition process is simplified, the authentication accuracy is lowered. Therefore, when the recognition process may take some time at the time of image reproduction or the like, the lack of authentication accuracy due to the simplification becomes a problem.

The present invention has been made in view of the above problems, and an object of the present invention is to enable detection of an individual stored in advance from an obtained image at a speed and accuracy according to the operation of an imaging apparatus.

To achieve the above object, an imaging apparatus of the present invention includes: a face detecting means for detecting an image signal or we face obtained by photographing a subject, a face detected by said face detecting means, advance in the database registered, by comparing the reference information for recognizing one or more individuals, a recognition means for pre SL performed for the detected face from the image signal recognizing process personal, focusing, exposure adjustment Adjusting means for adjusting photographing conditions including at least one of white balance adjustment, storage means for performing photographing processing for storing an image signal in a storage medium, the recognition means, the adjusting means, and the storing means and control means for controlling said control means, prior to the shooting process, on the display means, display is made for a specified one or more individual corresponding to the reference information to the user, prior to The recognition means, and said detected from the image signal face of the reference information, to perform a first recognition process by comparing the reference information for recognizing the designated person, the adjusting means , Adjusting the shooting condition based on the recognition result of the first recognition process, causing the adjustment unit to adjust the shooting condition, causing the storage unit to perform the shooting process, and giving an instruction from the user Without waiting, the recognition means sequentially compares the face detected from the image signal stored in the storage medium with the reference information for recognizing all individuals registered in the database in advance . Ru to perform the second recognition processing.

In addition, according to the control method of the imaging apparatus of the present invention, the display unit causes the user to designate one or more individuals among the individuals corresponding to the reference information for recognizing one or more individuals registered in the database in advance. a display step of performing display, the face detection unit, a face detection step of detecting a face from an image signal obtained by photographing a subject, the recognition unit, the face detected by the face detection process, which is the designated A first recognition step for performing a first recognition process for recognizing an individual with respect to a face detected from the image signal by comparing with reference information for recognizing the individual ; An adjustment step for adjusting photographing conditions including at least one of focus adjustment, exposure adjustment, and white balance adjustment based on the recognition result of the recognition step; and a storage means that stores the image signal after the adjustment step. Recorded in A storing step of performing imaging processing for, before Symbol recognizing means, without waiting for instructions from the user, and the storing step in detected from the image signal stored in the storage medium face, previously registered in the database and a second recognition step and the reference information for recognizing all individuals are compared sequentially performs the second recognition process.

According to the present invention, it is possible to detect an individual stored in advance from an obtained image at a speed and accuracy according to the operation of the imaging apparatus.

  The best mode for carrying out the present invention will be described below in detail with reference to the accompanying drawings.

<First Embodiment>
FIG. 1 is a block diagram showing a schematic configuration of a digital camera according to an embodiment of the present invention.

  In FIG. 1, reference numeral 1 denotes an optical system including a focus lens 2, and 3 denotes an image sensor that converts an optical image of a subject into an electrical image signal and outputs the image, and an optical image of a subject incident through the optical system 1 is obtained. Convert to image signal. 4 is a pre-processing circuit including a CDS circuit and a non-linear amplification circuit that removes output noise from the image signal output from the image sensor 3, and 5 is a digital signal of the analog image signal processed by the pre-processing circuit 4. It is an A / D converter for converting into an image signal. The image signal digitized by the A / C converter 5 is stored in the memory 7 via the memory controller 6, further converted into image data by the signal processing circuit 8, and then recorded on the recording medium 9. Reference numeral 10 denotes a display unit, and the signal processing circuit 8 also performs a process of converting the image signal stored in the memory 7 into a format suitable for display and outputting it to the display unit 10. An electronic viewfinder (EVF) function can be realized by sequentially displaying captured image data using the display unit 10.

  Reference numeral 15 denotes a face detection circuit that detects a face from an image signal obtained by imaging with the imaging element 3. Here, face information is detected and stored using an image created for EVF display on the display unit 10. This is repeated and the latest face detection result is always stored.

  As a face detection method, for example, a method using learning typified by a neutral network or a method of extracting a part having a physical shape such as an eye or nose from an image region by template matching is known. . Other methods include detecting image feature quantities such as skin color and eye shape and analyzing them using statistical methods (for example, Japanese Patent Laid-Open Nos. 10-232934 and 2000-48184). reference). Also, what is currently proposed as a product includes a method of detecting a face using wavelet transform and image feature amounts, a method of combining template matching, and the like. Many other face detection methods have been proposed, and known methods may be used.

  A control unit 13 controls the entire digital camera.

  Reference numeral 16 denotes an individual recognition circuit, which recognizes an individual using the face detection result by the face detection circuit 15. A personal database 17 stores a reference information file for recognizing an individual for each individual. Reference numeral 18 denotes an individual designation operation unit that can designate an individual to be recognized among individuals registered in the personal database 17. Then, the personal recognition circuit 16 takes out a file storing personal reference information designated by the personal designation operation unit 18 from the personal database 17 and collates it with the detected face. Thereby, it can be recognized whether or not the detected face is a designated individual. In addition, about the method of collation, what is necessary is just to use well-known methods, such as the method of patent document 2, for example. Each file stored in the personal database 17 and storing reference information for each individual is hereinafter referred to as a “personal information file”. The face detection circuit, personal recognition circuit, and personal database constitute detection means.

Reference numeral 14 denotes a focus lens driving circuit that drives the focus lens 2 based on the control of the control unit 13. The control unit 13 of the image signal obtained by photographing by the image pickup device 3, the contrast of the image signal in the distance measurement frame detects the position of the focus lens reaches a peak, drives the focus lens 2 to the position To perform focusing control.

  Reference numeral 19 denotes a shooting / playback switch for switching between a shooting mode (shooting process) and a playback mode (playback process). The switch (SW) 1 (11) is turned on by a first stroke (for example, half-pressed) of a shutter button (not shown), and instructs to start operations such as focusing control processing and automatic exposure processing. In the shooting mode, face detection and personal recognition are always performed, and when SW1 (11) is operated, a distance measurement frame is set for the personally recognized face and focus control is performed. In addition, the switch (SW) 2 (12) is turned on by a second stroke (for example, full press) of a shutter button (not shown), and instructs the main shooting and recording of the shot image. When SW2 (12) is operated, shooting and recording of an image signal obtained by shooting are performed.

[Operation in Shooting Mode] Next, the operation in the shooting mode in the first embodiment will be described in detail with reference to the flowchart of FIG.

In step S1, as well as the EVF display in the display section 10 the image of the subject, from among the individuals registered in the individual database 17, to designate individuals to be detected more individuals designation operation section 1 8. For example, when personal information files for eight people are registered in the personal database 17, as shown in FIG. 3, the registered names and face images of eight people are displayed as a list, and one or more are selected from the list. The individual to be recognized is specified. That is, the personal designation operation unit 18 and the control unit 13 constitute a control means.

  In step S2, face detection processing by the face detection circuit 15 and personal recognition processing by the personal recognition circuit 16 are performed. Here, face detection processing is performed on the image data displayed in EVF by a known method, and each detected face is compared and collated with the personal information file selected by the personal designation operation unit 18 in step S1. To determine whether the selected individual. If it is recognized that the detected face is the individual selected in step S1, the number of recognized persons, the information of the recognized individual, the face position, the size, and the like are stored. On the other hand, if it is not recognized, that fact is stored.

  In step S3, it is determined whether or not SW1 (11) has been pressed. If not, the process returns to step S1 to repeat the EVF display, face detection process, and personal recognition process described above. If SW1 (11) is pressed, the process proceeds to step S4, and the latest personal recognition result at that time is acquired. In step S5, if the acquired result indicates that the person is recognized, AF frames are set for all the faces recognized in step S6. If the person is not recognized, the screen is displayed in step S7. One AF frame is set at the center.

  In step S8, the control unit 13 performs focus adjustment control using the set image signal in the AF frame, and in step S9, the focus lens drive circuit 14 is controlled to a position focused by the focus adjustment control to focus. The lens 2 is fixed. As described above, the control unit 13 and the focus lens driving circuit 14 constitute an adjusting unit.

  In step S10, it is determined whether or not SW2 (12) has been pressed, and if it has been pressed, in step S11, the main shooting is performed for shooting the subject and recording the image signal obtained by shooting. If SW2 (12) is not pressed, it is determined in step S12 whether SW1 (11) is still pressed. If it is pressed, the process returns to step S9, and if released, the process returns to step S1. Return and repeat the above process.

  As described above, in the imaging mode, the selected specific individual is detected, and the person is focused and photographed. Since personal recognition of only a limited person is performed, the number of personal information files used for personal recognition, that is, the amount of reference information can be reduced, so that the processing time for personal recognition can be shortened.

  In the above description, the case where AF (focus adjustment) is performed by setting an AF frame using the personal recognition result has been described, but the present invention is not limited to AF. As adjustment of shooting conditions other than AF, for example, the exposure may be adjusted (exposure adjustment) or the white balance correction value (white balance adjustment) may be adjusted using image data within a set frame.

Image Classification Operation Next, the image classification operation in the first embodiment will be described in detail with reference to the flowchart of FIG. This image classification operation is performed when the reproduction mode is selected by the shooting / reproduction switch 19 and the image classification process for classifying the reproduced image using the personal recognition function is selected.

  In step S21, an image to be classified is read from the recording medium 9 and reproduced, and in step S22, the face detection circuit 15 performs face detection processing on the reproduced image. Thereafter, the personal recognition circuit 16 performs personal recognition on all detected faces in turn using all personal information files registered in the personal database 17.

  First, in step S23, one personal information file registered in the personal database 17 is read out and compared with one of the detected faces. If the individual is recognized as a result of the collation (YES in step S24), the process proceeds to step S26. On the other hand, if the individual cannot be recognized (NO in step S24), the process proceeds to step S25 to determine whether or not collation with all the personal information files registered in the personal database 17 has been completed. If not completed, the process returns to step S23 to collate with another registered personal information file. If completed, the process proceeds to step S26.

  In step S26, it is determined whether the process has been completed for all detected faces. If completed, the process proceeds to step S28. If not completed, the process proceeds to step S27, and the face to be processed is set as the next face. Set and return to step S23.

  In step S28, it is checked whether or not the individual can be recognized as a result of the personal recognition processing by the personal recognition circuit 16 described above. If it can be recognized, the process proceeds to step S29, and the control unit 13 controls the image data to be stored in a folder corresponding to the recognized individual. That is, the control unit 13 also serves as a classification unit. If there is no folder corresponding to the recognized individual, a new folder is created and image data is stored. If there are multiple faces that can be recognized individually, the image data is copied and stored in each folder, and the process proceeds to step S30. By this processing, the reproduced image can be classified for each detected individual (for each target). Note that when there are a plurality of faces that can be recognized individually, a pointer to the image data may be stored instead of copying and storing the image data itself. By doing so, it is possible to save memory capacity for storing image data.

  If it is determined in step S28 that the individual cannot be recognized, the process directly proceeds to step S30.

  In step S30, it is determined whether to classify another image. If not, the process ends. If so, the target image for classification processing is changed in step S31, and the process returns to step S21. The above process is repeated.

  As described above, according to the first embodiment, the number of personal information files used for collation, that is, the amount of reference information can be reduced by specifying in advance the individuals to be collated at the time of shooting. It is possible to shorten the time related to the process of detecting the target. On the other hand, at the time of reproduction, it is possible to recognize individuals with high accuracy by using all the recognizable individuals registered in all personal information files as targets for detection, that is, using all reference information.

  In the above-described first embodiment, the image classification process is described as the process performed in the playback mode. However, the present invention is not limited to this. For example, it recognizes a specified individual before shooting, performs AF processing according to the result, and automatically captures all personal information files registered in the personal database without waiting for an instruction from the user after shooting. The image storage location may be classified by inquiring.

<Second Embodiment>
Next, a second embodiment of the present invention will be described.

  Since the basic configuration of the digital camera is the same as that shown in FIG. 1, the description thereof is omitted here. Hereinafter, the operation in the second embodiment will be described. However, the image classification operation in the playback mode is the same as the processing described with reference to FIG. 4, and thus the description thereof will be omitted. The operation in the shooting mode will be described below. I do.

[Operation in Shooting Mode] FIG. 5 is a flowchart showing the operation in the shooting mode in the second embodiment.

  First, in step S <b> 41, the face detection circuit 15 performs face detection processing on the image data displayed on the display unit 10 by EVF. In step S42, the personal recognition circuit 16 reads one personal information file registered in the personal database 17, compares it with one detected face, and collates it. As a result of the collation, if an individual is recognized (YES in step S43), the process proceeds to step S3.

  On the other hand, if the person is not recognized (NO in step S43), the process proceeds to step S44, and it is determined whether the collation with all the personal information files registered in the personal database 17 is completed. If not completed, the process returns to step S42 to collate with another registered personal information file. If completed, the process proceeds to step S45.

  In step S45, it is determined whether or not the above process has been completed for all detected faces. If completed, the process proceeds to step S3. If not completed, the process proceeds to step S46, and the face to be processed is set as the next face. Set and return to step S42.

  Note that the processing after step S3 is the same as the processing after step S3 in FIG.

  As described above, according to the second embodiment, when one of the individuals registered in the personal database 17 is recognized at the time of shooting, the individual detection operation is exited. Therefore, at the time of shooting, the number of objects to be detected is less than or equal to the number of objects to be detected at the time of reproduction, and personal recognition can be performed in a short time, so that shooting can be performed without missing a photo opportunity. On the other hand, at the time of reproduction, it is possible to recognize individuals with high accuracy by using all the recognizable individuals registered in all personal information files as targets for detection, that is, using all reference information.

It is a block diagram which shows the structure of the digital camera which concerns on embodiment of this invention. It is a flowchart explaining the process at the time of the imaging | photography mode which concerns on the 1st Embodiment of this invention. It is a figure which shows an example of the screen for performing designation | designated of the individual who performs recognition which concerns on the 1st Embodiment of this invention. It is a flowchart explaining the image classification process at the time of the reproduction | regeneration mode which concerns on the 1st and 2nd embodiment of this invention. It is a flowchart explaining the process at the time of the imaging | photography mode which concerns on the 2nd Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Optical system 2 Focus lens 3 Image pick-up element 4 Pre-processing circuit 5 A / D converter 6 Memory controller 7 Memory 8 Signal processing circuit 9 Recording medium 10 Display part 11 Switch (SW) 1
12 Switch (SW) 2
DESCRIPTION OF SYMBOLS 13 Control part 14 Focus lens drive circuit 15 Face detection circuit 16 Personal recognition circuit 17 Personal database 18 Personal designation means 19 Shooting / reproduction | regeneration switch

Claims (3)

Face detection means for detecting a face from an image signal obtained by photographing a subject;
An individual is recognized for the face detected from the image signal by comparing the face detected by the face detection means with reference information registered in the database in advance for recognizing one or more individuals. Recognition means for performing recognition processing;
Adjusting means for adjusting shooting conditions including at least one of focus adjustment, exposure adjustment, and white balance adjustment;
Storage means for performing imaging processing for storing an image signal in a storage medium;
Control means for controlling the recognition means, the adjustment means, and the storage means,
The control means includes
Prior to the shooting process, the display means causes the user to perform display for designating one or more individuals from the individual corresponding to the reference information,
Causing the recognition means to perform a first recognition process by comparing a face detected from the image signal with reference information for recognizing the designated individual among the reference information;
Let the adjustment means adjust the shooting condition based on the recognition result of the first recognition process;
After allowing the adjustment means to adjust the shooting conditions, the storage means performs the shooting process,
Without waiting for an instruction from the user, the recognizing unit detects the face detected from the image signal stored in the storage medium, and the reference information for recognizing all individuals registered in the database in advance. The imaging device is characterized in that the second recognition process is performed by sequentially comparing the two.   The imaging apparatus according to claim 1, further comprising a classifying unit that classifies the image signal for each of the recognized individuals based on a recognition result of the second recognition process performed by the recognition unit. A display step in which the display means performs a display for allowing the user to designate one or more of the individuals corresponding to the reference information for recognizing one or more individuals registered in the database in advance;
A face detection step in which the face detection means detects a face from an image signal obtained by photographing the subject;
A recognizing unit compares the face detected by the face detection step with reference information for recognizing the designated individual, thereby recognizing the individual with respect to the face detected from the image signal . A first recognition step for performing recognition processing;
An adjusting step in which the adjusting means adjusts photographing conditions including at least one of focus adjustment, exposure adjustment, and white balance adjustment based on the recognition result of the first recognition step ;
A storing step in which a storing unit performs a photographing process of storing an image signal in a storage medium after the adjusting step ;
Before SL recognition means, without waiting for instructions from the user, to recognize said storing step with detected from the image signal stored in the storage medium face, all individuals are pre-registered in the database control method of an imaging device and having a second recognition step of performing the second recognition process by comparing the reference information in order.

Images