JP4275171B2 - camera - Google Patents

Description

  The present invention relates to an improvement of a digital camera, and more particularly to a camera that improves the performance of a camera according to determination of a subject image and enables photography without failure.

  A technique for detecting a face desired by a user from an image is widely known. For example, in the following Patent Document 1, a proposal is made to switch the volume level by determining the distance between the subject and the device using the size of the face. Cameras that control the focus and exposure of the face position are also on the market.

Furthermore, a technique for detecting the posture of a person is also known (see, for example, Patent Document 2).
JP 2004-40491 A Japanese Patent No. 2627483

  By the way, as described above, a camera that controls focusing and exposure with respect to the position of the face is also on sale, but the technique described in Patent Document 1 analyzes a shooting scene by face, and as a result, It was not a proposal for optimal camera control.

  In addition, the technique described in Patent Document 2 is not a technique related to photographing with a camera, and further, the configuration of the apparatus is large.

  Therefore, the present invention has been made in view of the above problems, and its purpose is to determine a face from a subject, perform shooting comfortably in an optimal scene and composition without missing a photo opportunity, It is to provide a technique for obtaining shots.

  That is, the invention described in claim 1 is a face detection unit that detects a face portion included in an image signal from an image signal from the imaging unit, a position of the face in the screen, and the vicinity of the face. Determining means for determining the best image of the photographed image when the contour line connected from the screen toward the outside of the screen has a predetermined shape.

  According to the second aspect of the present invention, the face detection means for detecting the face portion included in the image signal from the image signal from the image pickup means for monitoring the image in the shooting screen is connected to the vicinity of the face portion. And determining means for determining the best image of the photographed image when a change in the contour line over time is detected as a predetermined change pattern.

  According to a third aspect of the present invention, in the first aspect of the present invention, the determination means determines that the shape is a predetermined shape when the tip of the contour line comes above the face. .

  According to a fourth aspect of the present invention, in the first aspect of the present invention, the determining means determines that the shape is a predetermined shape when the extending direction of the contour line is below the face. To do.

  According to a fifth aspect of the present invention, in the first aspect of the present invention, the determining means determines that the shape of the predetermined shape is a predetermined shape when a tip of the contour line is at a farthest distance from the face. And

  According to a sixth aspect of the present invention, in the first aspect of the present invention, the determination means is configured to perform a predetermined operation when an extension of the contour line changes the shape of an object having a shape different from the contour line. It is characterized by determining the shape.

  According to a seventh aspect of the present invention, in the second aspect of the present invention, the determination means determines that the best image of the photographed image is obtained when the movement of the contour line stops.

  According to an eighth aspect of the present invention, there is provided a face detection means for detecting a face included in a photographed image, a skin color detection means for detecting the skin color of the face from the face detection means, and the skin color detection means. And determining means for determining the position of the hand of the photographed image according to the detection result.

  The invention according to claim 9 is the invention according to claim 8, wherein the determination means determines that the hand is a good shot when the position of the hand is a predetermined position relative to the position of the face. Features.

  According to a tenth aspect of the present invention, in the invention according to the eighth aspect, the determination means determines a good shot when the position of the hand is combined.

  The invention according to claim 11 is a face detection means for detecting the position and size of the face included in the photographed image, and a skin color detection means for detecting the skin color of the face from the face detection means. And determining means for determining the position of the foot of the photographed image according to the skin color detection result.

  According to a twelfth aspect of the present invention, there is provided face detection means for detecting the position and size of a subject's face included in a photographed image, and whether the subject is a full-body image or half body according to the output result of the face detection means. And determining means for determining whether the image is an image.

  According to a thirteenth aspect of the present invention, there is provided face detection means for detecting a face portion of a subject included in a photographed image, and at least a part of the subject detected by the face detection means in the photographed image. And determining means for determining the position of.

  According to a fourteenth aspect of the present invention, in a camera having a continuous shooting function for continuously shooting a plurality of images, the imaging means for shooting an image of a subject and the image captured by the imaging means are included The face detection means for detecting the face portion of the subject, the position of the face in the screen, and the contour line connecting from the vicinity of the face toward the outside of the screen has a predetermined shape And determining means for determining the best image of the photographed image.

  According to a fifteenth aspect of the present invention, in a camera having a continuous photographing function for continuously photographing a plurality of images, the imaging means for photographing an image of a subject and the image photographed by the imaging means are included. A face detection means for detecting a face portion of the subject and a time change of a contour line connecting the vicinity of the face as a center is detected as a predetermined change pattern. Determination means for determining the best image.

  According to a sixteenth aspect of the present invention, in the invention according to the fourteenth aspect, the determination means has the tip of the contour line reaches the upper part of the face portion in the continuously photographed images. It is characterized in that a predetermined shape is sometimes determined.

  According to a seventeenth aspect of the present invention, in the invention according to the fourteenth aspect, the determining means includes an extension direction of the contour line that reaches a lower part of the face portion in the continuously photographed images. It is characterized in that it is determined as a predetermined shape when it is done.

  According to an eighteenth aspect of the present invention, in the invention according to the fourteenth aspect, the determination means sets the tip of the contour line to the farthest distance from the face portion in the continuously photographed images. It is characterized in that it is determined as a predetermined shape when it arrives.

  According to a nineteenth aspect of the present invention, in the invention according to the fourteenth aspect, the determination means has a shape in which an extended portion of the contour line is different from the contour line in the continuously photographed images. When the shape of the object is changed, the predetermined shape is determined.

  According to a twentieth aspect of the present invention, in the invention according to the fifteenth aspect, the determination unit is configured to detect the captured image when the movement change of the contour portion of the continuously photographed images stops. The best image is determined.

  The invention according to claim 21 further comprises recording means for selecting and recording only an appropriate image from the best images of the continuously photographed images in the invention according to claim 14. It is characterized by doing.

  According to a twenty-second aspect of the present invention, in a camera having a continuous shooting function for continuously shooting a plurality of images, an imaging unit that captures an image of a subject, and a captured image captured by the imaging unit. A face detection unit for detecting a face part included, a skin color detection unit for detecting a skin color of the face from the face detection unit, and a position of the hand of the photographed image according to a detection result of the skin color detection unit Determining means for performing the above-described operation.

  According to a twenty-third aspect of the present invention, in the invention according to the twenty-second aspect, the determination unit is configured such that the position of the hand is predetermined with respect to the position of the face among the continuously captured images. The photographed image at the position is determined as a good shot.

  According to a twenty-fourth aspect of the present invention, in the invention according to the twenty-second aspect, the determination unit is configured to shoot when the position of the hand is combined with both hands in the continuously photographed images. The image is determined to be a good shot.

  According to a twenty-fifth aspect of the present invention, in a camera having a continuous photographing function for continuously photographing a plurality of images, an imaging means for photographing an image of a subject, and a photographed image photographed by the imaging means Face detection means for detecting the position and size of the included face, skin color detection means for detecting the skin color of the face from the face detection means, and according to the skin color detection result, Determining means for determining a position.

  According to a twenty-sixth aspect of the present invention, in a camera having a continuous photographing function for continuously photographing a plurality of images, an imaging means for photographing an image of a subject, and a photographed image photographed by the imaging means Face detection means for detecting the position and size of the face of the subject included, and determination means for determining whether the subject is a full-body image or a half-body image according to the output result of the face detection means. Features.

  According to the present invention, it is possible to provide a technique for determining a face from a subject, performing a comfortable shooting with an optimal scene and composition without missing a photo opportunity, and obtaining a good shot of a person.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(First embodiment)
FIG. 1 is a block diagram showing a basic configuration of a camera according to the first embodiment of the present invention. The camera includes a main CPU (hereinafter referred to as MPU) 11, a plurality of switches 11a, 11b, and 11c, a signal processing unit 12, a face detection unit 13, a contour detection unit 14, a flash ROM 15, and auxiliary light. Irradiation unit 16, focusing unit 18, zoom control unit 19, analog front end unit (AFE) 21, SDRAM 22, memory 23, microphone 24, speaker 26, display unit 27, and print signal output A unit 28, a wireless communication unit 29, an image sensor 31 and a photographing lens 32 are configured.

  In FIG. 1, an MPU 11 having functions of a control unit and a determination unit is constituted by a microcontroller or the like, and detects various operations by a user based on the states of switches 11a, 11b, and 11c. Further, the MPU 11 sequentially controls each block to be described later according to the detection results of the switches 11a, 11b, and 11c and the program stored in the flash ROM 15, and executes a shooting and playback sequence. The switches 11a to 11c have a function as a selection unit such as a power switch (11a) for turning on / off the power, a release switch (11b), and shooting / playback and other switches (11c) described later. The user's operation is input.

  The MPU 11 includes a signal processing unit 12 having a function such as signal compression, a face detection unit 13, a contour detection unit 14, a flash ROM 15, an auxiliary light irradiation unit 16, a focusing unit 18, and a zoom control unit 19. And are connected. The signal processing unit 12 performs processing such as color correction and signal compression. In addition, the signal processing unit 12 includes an auxiliary block 12a for changing how to perform sensitization after changing the exposure level and correction during image processing, and changing the compression rate.

  The face detection unit 13 analyzes the image signal obtained from the image sensor 31 and uses the features of the human face to detect whether there is a person in the subject and determine the position and size of the face. It is a face part detection means for doing. The contour detection unit 14 serving as a contour detection unit has a function of determining the shape of a contour obtained by sharpening an image.

  The flash ROM 15 stores various programs executed by the MPU 11 of the camera and various adjustment values related to the camera. The auxiliary light irradiating unit 16 irradiates auxiliary light toward the subject 34 to supplement exposure or capture shadows. Further, the focusing unit 18 is for controlling the photographing lens 32 so as to focus the subject 34 on the imaging surface of the image sensor 31 via the photographing lens 32 described later. The zoom control unit 19 is a control unit for switching the angle of view by the photographing lens 32.

  The signal processing unit 12 further includes an analog front end unit 21, an SDRAM 22, a memory 23, a microphone 24, a speaker 26, a display unit 27, a print signal output unit 28, and a wireless communication unit 29. It is connected. The analog front end unit 21 is connected to an image sensor 31 that is a photographing unit that forms an image of the subject 34 incident from the photographing lens 32 during photographing. The image sensor 31 is constituted by a CCD, a CMOS sensor, or the like, and photoelectrically converts the subject image into an image signal.

  The analog front end unit 18 reads the converted image signal and performs A / D conversion. The analog front end unit 18 also performs control for switching the exposure level of the input image. The digital image signal converted by the analog front end unit 18 is input to the signal processing unit 12. The signal processing unit 12 performs various image processes to generate image data. The image data is recorded in a removable memory 23 that is a recording means. Here, the image data recorded in the memory 23 is compressed and recorded in an appropriate amount of data according to the camera settings. At this time, data is appropriately stored in the SDRAM 22 in order to perform various operations. In the present invention, the moving image is also first stored here.

  The display unit (display unit) 21 that constitutes the classification display unit together with the MPU is configured by an LCD or the like. The composition is determined by confirming the state of the subject 34 by sequentially reproducing and displaying signals from the image sensor 31 by the display unit 21. Reproduction display is performed by reading the compressed image data in the memory 23 into the signal processing unit 12, converting it into display data, and displaying it on the display unit 27 in accordance with the user's operation of the switch 11 c. Similarly, in the audio reproduction, the compressed audio data in the memory 23 is read out to the signal processing unit 12 and converted into audio data according to the operation of the switch 11c, and is output from the speaker 27. The memory 23 may be either a built-in one or a removable one like a card.

  The captured image or the like can be transmitted to an external device (not shown) via the print signal output unit 28 and the wireless communication unit 29.

  In such a configuration, when the user operates the switches 11a to 11c, the state is detected by the MPU 11, and each block is sequentially controlled in accordance with the program stored in the flash ROM 15, and the sequence of shooting and reproduction is performed. Is executed.

  At the time of photographing, an image of the subject 34 incident from the photographing lens 32 is captured by the image sensor 31 and converted into an electrical signal, and digital image data is input to the signal processing unit 12 via the analog front end unit 21. When the signal processing unit 12 performs processing such as color correction and image compression, the resulting data is recorded in the memory 23 and the shooting is finished.

  If the camera having such a configuration to which the technology of the present invention is applied is used, it is suitable for instantaneous photographing as described below. That is, as shown in FIG. 2 (a), at the moment when the subject 41 hits the ball 43 with the club 42 in golf, the subject 45 entered the goal in the track and field competition as shown in FIG. 2 (b). The moment, as shown in FIG. 2C, is the moment when the subject 46 kicks the ball 47 in soccer.

  Since it is very difficult to capture such a subject at a desired timing, even a professional photographer has selected a good one from many shooting results (photos). Therefore, since it is more difficult for an amateur to capture such a scene, in this embodiment, a continuous shooting mode in which a plurality of images are continuously shot is used, and a desired shooting mode is selected. The method is to select a photo.

  However, if you take a lot of images, it takes time to select the best shot from them, and the data will occupy the memory recording area inside the camera. However, it is necessary to devise and leave only the necessary items. Therefore, in this embodiment, as the method for selecting the optimum image, the image is analyzed and selected in consideration of the pose of the subject.

  The images shown in FIGS. 2A to 2C are characteristic in that the face of the person who is the subject is photographed at an appropriate position, and the hands and feet are in predetermined positions. It is that the ball or sphere is deformed or skipped.

  In other words, by narrowing down images based on these conditions, an optimal scene is left as a candidate from a large number of images.

FIGS. 3A to 3C show the contours and face portions radiated from the vicinity of the face as a result of the contour detection of each scene corresponding to FIGS. 2A to 2C. FIG. In FIGS. 3A to 3C, the arrows indicate that the faces are projected radially from the face portions marked with ◯, and the tips of the radiation contours thus produced are marked with ◯ marks. That is, in the case of FIG. 3A, the radiation contour 41b extends toward the ball 49 located directly below the face 41a. Further, in the case of FIG. 3B, the radiation contours 45b ₁ and 45b ₂ toward both hands 50a ₁ and 50a ₂ are located above the face 45a, and in the case of FIG. 3C, the position of the face 46a. It can be seen that the foot extends from the position of the ball 51 to the position of the ball 51 (radiation contour 46b), which is a characteristic composition. In this way, when the limb of the subject is fully extended or occupies the entire screen, the image changes with a good appearance, and the user can easily satisfy the selection.

  In addition, as shown in FIGS. 4A to 4C, there are cases where the characteristics of the good shot are clarified due to the difference from the preceding and following frames, not the characteristics of one frame and one frame.

  That is, in the race scene as shown in FIGS. 4A to 4C, the subject 53 looks like FIGS. 4A and 4B with the arm below the shoulder until the goal. Running in. At the moment when the first person finishes, there are many discontinuous movements such as raising an arm as shown by an arrow in FIG. 4C, so this shot is regarded as a good shot (best image). To do. However, since there are cases where it is desired to photograph a person other than the first person, in this case, as shown in FIG. 4D, the subject (person) 54 to be photographed is set at the center of the screen. In this way, it may be devised not to select a shot as shown in FIG.

  Further, as shown in FIG. 5, when it is desired to capture the moment when the subject 56 kicks the ball 57, the operation is as follows. That is, in the screens of FIGS. 5A and 5B, the ball 57 exists in the screen, but in the screen shown in FIG. 5C, the ball 57 does not exist in the screen. Accordingly, the frame before the frame disappeared from the screen of the ball 57, here, the frame in FIG. 5B may be regarded as a good shot.

  In this way, since a good shot candidate is selected by comparison with the preceding and following frames, a more effective and powerful shot can be selected.

Further, in the scene shown in FIG. 6A, the radial contour from the face 46a is not connected in a straight line to the tip of the foot (ball 51) as shown in FIG. 6C. Absent. However, as shown in FIG. 6B, when the angle θ connected to the contour line l ₁ radiating from the face 46a and the next line l ₂ is equal to or greater than a predetermined angle and is close to a straight line, Try to put ingenuity that can be regarded as a series of radial contours. In this way, it is possible to select a scene that looks great with the subject 46 fully extended.

As shown in FIG. 7 (a), the scene in which the foot of the subject 58 is raised is also powerful, but with the idea shown in FIG. 6 (c), in the previous scene, If the contour l ₂ is detected and the tip is tracked and the scene located above the head is made a good shot candidate, this can also be made a candidate.

  Next, based on such a concept, the photographing operation of the camera in the present embodiment will be described with reference to the flowcharts of FIGS.

  FIG. 8 shows processing when the shooting mode is selected by the switch 11c. When this sequence is started, first, the state of the release switch 11b is determined. This is determined by whether or not the release switch 11b is pressed at the timing when the photographer wants to shoot. As a result, when the release switch 11b is pressed, this is detected by the MPU 11, and the photographing step after step S2 is entered. On the other hand, since the release switch 11b is not pressed for a scene shot before that, the process proceeds to step S27 described later.

  In FIG. 9, in step S27, whether or not the playback mode is set is determined based on the state of the switch 11c. If the playback mode is not set, the process proceeds to step S28 and the signal output from the image sensor 31 is displayed (reproduced) on the display unit 27 in real time. The user determines the push timing of the release switch 11b while viewing the image displayed on the display unit 27.

  At this time, in steps S29 to S31, a face portion may be detected before photographing. The detection of the face portion is performed by using a known technique, for example, a technique such as pattern matching, which analyzes and detects the arrangement of the subject's eyes, nose muscles, and mouth from the image data (signal from the image sensor). Or you may detect so that the reflected light pattern at the time of projection from the auxiliary light irradiation part 16 may be determined with sensors, such as the image pick-up element 31. FIG. Alternatively, these methods may be used in combination.

  As described above, by detecting the face portion before actually shooting, the processing after the release switch is pressed can be speeded up.

  When a face is detected in step S29, a face frame (not shown) is displayed at a position on the display unit 27 corresponding to that part in the subsequent step S30, and is correctly detected by the user. The fact is announced. In step S31, the coordinate position (the position of the face being displayed) is monitored. In this way, the face position can be determined immediately even during shooting.

  In such a state, when the user performs an operation to start shooting, the process proceeds to step S2 in FIG. 8 to determine whether or not a face has been detected in advance. If no face is detected, the process proceeds to step S3, and average shooting is performed to improve the screen average without respecting the depiction of the face. Thereafter, the process proceeds to step S23. On the other hand, if a face portion is detected in step S2, the process proceeds to step S4, and proper face photographing is performed so that the exposure and focus of the portion are appropriate.

  Next, in steps S5 and S6, the result of determining the face coordinates is recorded for each frame. Then, in step S7, it is determined whether or not the face size or position is inappropriate. If the result is inappropriate, it is not considered a good shot, and the process proceeds to step S23 described later. On the other hand, if the position and size of the face are appropriate, the process proceeds to step S8, and the contour of the image is enhanced by sharpness processing or the like in the image processing by the signal processing unit 12.

In step S9, a radially extending contour is detected and selected from the vicinity of the face coordinates recorded in step S6. Next, in steps S10 and S11, the coordinates of the tip (see the sphere in FIG. 3A, the hands 50a ₁ and 50a ₂ in FIG. 3B, and the ball 51 in FIG. 3C) are determined. Recorded. Through such a process, the coordinates of the face of the subject and the tip of the limb are recorded. As a result, it is possible to determine by tracking changes in these coordinate positions, and it is also possible to determine a scene as shown in FIG.

  Next, in step S12, a change in the contour determined in step S11 is determined as a radiation contour. Here, for example, as shown in FIG. 4B to FIG. 4C, when the change of the contour portion is large, the process proceeds to step S14, but when the change is small, the process proceeds to step S13 and the change flag is set to “0”. Set to "". Here, the position of the radial contour tip coordinate in step S10 and the amount of change in the radial contour in step S11 may be recorded in advance in the flash ROM 15, and the determination may be made based on this data.

  When the change of the contour portion is large, the change flag is set to “1” in step S16 described later. This shooting sequence is repeated until the end operation for instructing release OFF is performed by the user in step S23, which will be described later. In this continuous determination, the change flag is changed from “0” to “1”. If it changes, the presence or absence of this change is determined in step S14. If the change flag changes, the subsequent image is set as a candidate image in step S15. On the other hand, if the flag does not change in step S14, step S15 is skipped. Thereafter, in step S16, the change flag is set to “1”.

  Furthermore, a shot with an appropriate subject pose is also determined in steps after step S17. That is, in step S17, for example, as in the scene shown in FIGS. 2B and 7A, the radial contour tip coordinate position (the position of the tip of the hand or foot) is recorded in step S6. It is determined whether or not it exists above the face position. As a result, when the radial contour tip coordinate position exists above the face position, the process proceeds to step S18 to be a good shot candidate image, but when it does not exist, step S18 is skipped. .

  In step S19, for example, as in the scene shown in FIG. 2A, it is determined whether or not the radial contour tip coordinate position exists immediately below the face coordinates. As a result, when the radial contour tip coordinate position exists immediately below the face coordinates, the process proceeds to step S20 to be a good shot candidate image, but when it does not exist, step S20 is skipped. .

  In step S21, it is determined whether or not the shape of the adjacent portion of the radial contour (the shape of the ball 57 or the like in FIGS. 5A and 5B) has changed. As a result of the determination, if the shape of the adjacent portion of the radial contour has changed, the process proceeds to step S22 to be a good shot candidate image, but if it has not changed, step S22 is skipped.

  In this way, the gorgeous scene where the hands and feet are high, the scene where the hand is stretched down and tightened, the image of the moment when the ball is hit, etc., as a good shot by the processing steps of steps S17 to S22. Can leave. Each determination in steps S17, S19, and S21 is performed by, for example, pre-recording template data having each coordinate position or each region information in the flash ROM 15 and comparing it with this. A technique that can make the same determination may be used.

  Thus, when the determination of the pose of the subject is completed, it is determined in step S23 whether or not an operation for ending the shooting has been performed. Here, if not finished yet, the process proceeds to step S4, and the subsequent processing is repeated. On the other hand, when an operation for ending photography is performed, the process proceeds to step S24.

  For scenes that cannot be determined for each frame, processing steps of steps S24 and S25 are provided in order to compare and select the previous and subsequent frames after shooting.

  That is, in step S24, the coordinates of the tip of the radial contour and the coordinates of the face are compared, and the scene with the farthest distance is set as the candidate image. This is a good shot candidate for a scene in which the limbs are stretched across the screen, and is suitable for photography that shows the intensity of movement. In step S25, for example, as shown in FIGS. 5B and 5C, the change in the adjacent portion of the radial contour (in this case, the ball 57) is caused by a scene change in which the ball 57 disappears from the screen. Good shot candidates are selected from the size. By selecting the preceding scene (for example, FIG. 5B in FIGS. 5A to 5C), the moment with the most impact is selected. Note that when the process of step S3 described above is performed, no face is detected, and therefore, steps S24 and S25 are substantially not processed.

  In step S26, the captured image is recorded.

  As described above, when good shot detection is performed during and after shooting, and the user switches the mode switch (such as 11c in FIG. 1) to set the playback mode, the process proceeds from step S27 to step S32.

  In step S32, the presence / absence of a candidate image is determined. As a result, if there is a candidate image, the process proceeds to step S33 and the image is preferentially displayed. On the other hand, if there is no candidate image, the process proceeds to step S34, and four frames divided at equal intervals from the continuous shot are displayed as candidate images. Here, whether or not it is a candidate image records attribute information for each captured image. Alternatively, only the information linking each captured image and the presence / absence of a candidate image may be recorded separately as a control file.

  Next, in step S35, whether or not to erase other than the display image is displayed on, for example, the display unit 27 to notify the user. Here, if the user performs an operation to delete other than the display image, the process proceeds to step S36, where the operation is executed, and a useless image is deleted from the memory 23 so that the next shooting is not hindered. To be. In step S35, step S36 is skipped if the display image is not erased. In this way, by always making a margin in the memory 23, the problem that the photographing cannot be performed due to a lack of the capacity of the memory 23 is avoided while leaving only a good shot.

  Further, in step S37, it is determined whether or not an operation for switching the reproduction display image has been performed. Here, when the user performs an operation of switching the reproduction display image, the process proceeds to step S38 and is switched to the reproduction display image. On the other hand, if the reproduction display image is not switched in step S37, step S38 is skipped.

  Thus, this sequence is completed.

  Thus, according to the first embodiment, it is possible to provide a camera that can accurately capture a person's quick movement.

  In other words, continuous shooting was performed at high speed, and only images suitable for good shots were automatically selected from a series of images and recorded preferentially. In this way, it is possible to analyze a subject's movement and shape with a simple configuration, determine a pose and a moment that are most suitable for recording, and provide a camera that enables even beginners to perform the best shooting.

  Note that only candidate images may be recorded in step S26. In this case, since a candidate image can be selected during shooting, more good shots can be shot at high speed without waste by not storing images other than candidate images.

  In the first embodiment, even if a candidate image is obtained, the presence / absence of a candidate image is determined and determined based on other conditions. However, the present invention is not limited to this. May not be executed. In this case, the number of candidate images is limited, but the processing speed can be further increased.

(Second Embodiment)
Next, a second embodiment of the present invention will be described.

  In the second embodiment described below, the basic configuration of the camera is the same as that of the first embodiment described above. A reference number is attached | subjected, the illustration and description are abbreviate | omitted, and only a different part is demonstrated.

  FIG. 10 is a flowchart for explaining the photographing operation of the camera according to the second embodiment of the present invention.

  This second embodiment is based on the idea of determining the skin color of the subject in accordance with the face detection result at the time of photographing (or prior to that), and based on the idea of determining the skin color of the subject, and is almost half of the face size. A skin-colored part of a certain size is determined as a palm (or back of hand) part. The skin color is determined from the color information of the image data, for example, the ratio of R / G, B / G, which is the ratio of R (red information), G (green information), and B (blue information) from the captured image data. The determination is made based on the distribution in a predetermined area in the screen. This skin color determination is performed by a program executed in the MPU. In this sense, the MPU can be referred to as a skin color detection unit. Thus, when a pose of a subject to be photographed as shown in FIGS. 11A to 11C is detected, this is used as a good shot candidate.

  When this sequence is started, face detection is first performed in step S41 to detect the position and size in the screen. Next, in step S42, skin color determination is performed from the color of the face portion detected in step S41. Further, in step S43, a subroutine “whole body image determination” is executed. The detailed operation of this subroutine “whole body image determination” will be described later.

  Next, in step S44, it is determined from the position and size of the face of the subject whether the shooting is for half body or whole body. If it is determined that half-body shooting has been performed, the process proceeds to step S45, and a skin color portion in the vicinity of the face is detected based on the skin color determined in the face portion. If the detected skin color portion is approximately half the size of the face portion as described above, it corresponds to the hand size in step S46. It is determined whether or not.

  If the skin color portion corresponds to the size of the hand, the process proceeds to the hand position determination process in step S47 and subsequent steps. That is, in step S47, for example, as shown in FIG. 11A, it is determined whether or not the detected position of the hand 60b exists below the face 60a portion of the subject 60. Here, if it is determined that the detected hand position exists below the face portion, the process proceeds to step S48, and a good shot candidate image is obtained. On the other hand, if it is determined in step S47 that the detected hand position does not exist under the face part, step S48 is skipped.

  Similarly, in step S49, for example, as shown in FIG. 11B, it is determined whether or not the detected position of the hand 60b exists below the face 60a portion of the subject 60. Here, when it is determined that the detected hand position is present beside the face portion, the process proceeds to step S50 to be a good shot candidate image. On the other hand, if it is determined in step S49 that the detected hand position does not exist next to the face portion, step S50 is skipped.

  Then, in step S51, for example, as shown in FIG. 11C, it is determined whether or not the hand 60b portion of the subject 60 is assembled. If it is determined that the hand portion of the subject is assembled, the process proceeds to step S52 to be a good shot candidate image. As shown in FIG. 12, the palm portion 62 having the same color as the face portion 61 is detected to determine whether or not both hands of the subject are assembled, and sharpness enhancement processing is performed as in the first embodiment described above. It is possible to determine by detecting whether or not two contour lines detected by, for example, extend radially from the palm portion.

  On the other hand, if it is determined in step S44 that the photograph is not an upper body photograph, it is determined that the whole body has been photographed, and the process proceeds to step S53. In this step S53, it is determined whether or not the skin color portion determined in step S42 exists in the lower part of the screen. That is, when the portion having the contour extending downward is used as the foot of the subject, and there is no color other than the skin color between them, the process proceeds to step S54, and both the feet of the subject as shown in FIG. The intersecting pose is used as a candidate screen.

  Note that as a determination that both hands are assembled, a template indicating a coordinate position and a region as a determination condition is stored in the flash ROM 15 in advance as in the first embodiment described above, and compared with this. You may make it do. Furthermore, it is of course possible to determine a scene using data that tracks changes in coordinates and areas.

  Thus, if only candidate images satisfying a predetermined condition can be extracted from a large number of captured images, a good shot image can be easily obtained.

  Furthermore, the subroutine “whole body image determination” in step S43 in the flowchart of FIG. 10 is performed based on the concept shown in FIG.

That is, when the coordinates of the face 64a of the person 64 in the screen are obtained as x _K and y _{K based} on the lower left reference and the face size is obtained as Δy, the distance from the bottom of the screen to the top of the head is about y _K. + Δy / 2.

In such a state, when the relationship between the face and the height is set to 1: 6 and 6 × Δy is subtracted, the coordinates of the tip of the foot are obtained as y _A. Therefore, if this y _A is larger than y = 0 of the origin, it can be determined as a whole body image, and if it is small, it can be determined that the tip of the foot is not captured. Although there are individual differences and differences between children and adults, here, the average value is taken and the calculation method is as shown in the following equation (1).
y _A = y _K + (Δy / 2) −6Δy (1)
FIG. 14 is a flowchart for explaining the operation of the subroutine “whole body image determination” for determining whether the whole body image or the half body image based on such a concept.

That is, when entering this subroutine, the face coordinates x _K and y _K are first detected in step S61, and then the face size Δy is detected in step S62. This is detected by a method such as pattern matching with shadows of eyes and nose and mouth used in personal authentication and security cameras.

In step S63, y _A (foot position) is obtained by calculation according to the above equation (1). Next, in step S64, it is determined whether y _A (foot position) obtained in step S63 is outside the screen or inside the screen. Here, if y _A is equal to or greater than 0, the process proceeds to step S65, and processing as a whole body image is performed. On the other hand, if y _A is smaller than 0 in step S64, the process proceeds to step S66 and a half-body image is processed.

  In addition, if such a technique is used, it can be effectively used when it is desired to leave only an image that captures the scene with the best timing from a plurality of images obtained by repeatedly shooting at high speed.

As shown in the timing chart of FIG. 15, it takes a time of Δt ₂ from imaging (Δt ₁ ) to displaying and determining a good shot. If it is determined to start shooting at the best timing based on the determination result of the image taken at this time, the shooting timing is affected by a time delay of Δt ₂ . Therefore, in the present embodiment, all the captured images are once recorded, and the result of performing the good shot determination from each image is added to the image file as a priority flag.

  There is no influence of the time lag between the image recorded in this way and the time required for determination, and a good shot image can be preferentially displayed after shooting.

Further, in order to effectively display the image determined by the good shot, as shown in FIG. 16A, each captured image (for example, the image t _n determined so that the image t _n determined by the good shot is at the top). , T _n−2 , t _n−1 , t _{n + 1 to} t _{n + 3} ) are superposed and displayed on the display unit 27. Alternatively, as shown in FIG. 16 (b), by displaying the largest image t _n which is good shot determined in the center of the screen, an image (e.g., taken before and _{after, t n-4 ~t n-} 1 , T _{n + 1 to} t _{n + 5} ) are displayed as small thumbnail images around them. At this time, the user determines necessity other than the image that has been determined to be a good shot, and if not necessary, performs an operation of instructing deletion to delete all images other than the good shot.

  FIG. 17 is a flowchart for explaining the above processing operation.

  The image data captured in step S71 is subjected to image processing in the subsequent step S72. Thereafter, the image data processed in the recording data format is recorded in the memory 23 in step S73. Thereafter, the live view image is displayed as a through image in step S74.

  Next, in step S75, a good shot is determined from the image data recorded in step S74. As this determination method, the method described above is used. If it is determined in step S75 that the image is a good shot image, the process proceeds to step S76, and a flag representing the good shot image is recorded as incidental information in the image file. On the other hand, if it is determined in step S75 that the image is not a good shot image, step S76 is skipped.

  Thereafter, a release OFF instruction is determined in step S77. Here, when it is determined that an OFF instruction has been issued, the imaging is terminated, and the flagged image is preferentially displayed among the images captured on the display unit 27 in step S78. Thereafter, in step S79, a priority image is selected.

  On the other hand, when it is determined in step S77 that shooting is continued, the process proceeds to step S71, and the above-described shooting process is continued.

  FIG. 18 is a flowchart for explaining detailed processing operations of the subroutine “priority image selection” in step S79 of FIG. 17 described above.

  That is, in step S81, it is determined whether to delete a recorded image other than the priority image. This instruction is determined based on the operation result of the user based on the image or notification displayed on the display unit 27. If it is determined to delete other than the priority image, the process proceeds to step S82, and all the captured and recorded image data other than the priority image are deleted (erased) from the memory 23. On the other hand, if it is determined in step S81 that no image other than the priority image is deleted, step S82 is skipped. Thereafter, the subroutine is exited.

  As another application of priority image selection, as shown in the flowchart of FIG. 19, when a good shot determined image is not desired, a good shot image may be manually selected from other images. Good.

  At that time, as shown in FIG. 16C, a good shot image is displayed on the entire screen of the display unit 27, and it can be confirmed that only this image has been recorded. As a result, shooting can be performed more reliably at the best timing, and memory is not wasted.

  FIG. 19 is a flowchart for explaining another processing operation of the subroutine “priority image selection” in step S79 of FIG. 17 described above.

In step S91, it is determined whether to delete a recorded image other than the priority image. Here, if it is determined that the images other than the priority image are to be deleted, the process proceeds to step S92, and all the captured recording image data other than the priority image are deleted (erased) from the memory 23. On the other hand, if it is determined in step S91 that no image other than the priority image is deleted, the process proceeds to step S93, and the deleted image is manually selected. In this selection, as shown in FIG. 16B, the image to be deleted from the thumbnail-displayed images (for example, t _{n−4 to} t _n−1 , t _{n + 1 to} t _{n + 5} ) It is selected by the key operation corresponding to 11c etc. and is confirmed by the confirmation key.

  If this determination is made, in step S94, the display image is updated so that the image selected as the saved image is displayed. Thereafter, the image data selected as the deleted image is deleted from the memory 23 in step S95.

  The images selected as the stored images are stored with frame numbers in the order of shooting. In addition, although the case of manual selection has been described in the above-described example, if the result of manual selection is stored and there is a correlation between the good shot determination result and the manual selection result, for example, the deviation in the shooting time is biased If it can be seen, the amount of deviation may be judged and fed back to good shot judgment to be learned by a known method. In this case, it is possible to select an image at a timing more suitable for the user's preference.

  Further, in the present embodiment, the example in which image processing, recording, display, and good shot determination are processed in time series has been described. However, the present invention is not limited to this, and for example, a good shot determination process using a system having a plurality of MPUs. Other processes may be processed in parallel. In this way, higher speed processing can be performed.

  Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the present invention.

  Further, the above-described embodiments include inventions at various stages, and various inventions can be extracted by appropriately combining a plurality of disclosed constituent elements. For example, even if some constituent requirements are deleted from all the constituent requirements shown in the embodiment, the problem described in the column of the problem to be solved by the invention can be solved, and the effect described in the column of the effect of the invention Can be extracted as an invention.

It is a block diagram which shows the basic composition of the camera which concerns on one Embodiment of this invention. It is the figure which showed the example of the photograph used as the best shut at the time of image | photographing the to-be-moved subject. FIG. 3 is a diagram in which contours and face portions emitted radially from the vicinity of a face are extracted and displayed as a result of contour detection of each scene corresponding to FIGS. (A)-(c) is a diagram showing an example in which the characteristics of a good shot are clear due to the difference from the preceding and following related frames, and (d) is a diagram showing an example in which a desired subject is photographed at the center. It is. It is a figure for demonstrating the moment when a to-be-photographed object kicks a ball | bowl. It is a figure for demonstrating the example for obtaining the good shot of the moment where a to-be-photographed object kicks a ball | bowl. It is a figure for demonstrating the other example for obtaining the good shot of the moment when a to-be-photographed object kicks a ball | bowl. 5 is a flowchart for explaining a photographing operation of the camera in the first embodiment of the present invention. 5 is a flowchart for explaining a photographing operation of the camera in the first embodiment of the present invention. It is a flowchart for demonstrating imaging | photography operation | movement of the camera in the 2nd Embodiment of this invention. It is a figure for demonstrating the face part detection in 2nd Embodiment. It is a figure for demonstrating determination of whether the to-be-photographed object's both hands are assembled. It is a figure for demonstrating the idea of the subroutine "whole body image determination" of step S43 in the flowchart of FIG. It is a flowchart for demonstrating the detailed operation | movement of the subroutine "whole body image determination" of step S43 in the flowchart of FIG. It is a timing chart which shows the operation timing after performing an image display, performing display and determining a good shot. It is a figure for demonstrating the example which displays effectively the image by which the good shot determination was carried out. It is a flowchart for demonstrating the processing operation | movement for obtaining the good shot image of FIG. 18 is a flowchart for explaining detailed processing operations of a subroutine “priority image selection” in step S79 of FIG. 17; 18 is a flowchart for explaining another processing operation of the subroutine “priority image selection” in step S79 of FIG.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 11 ... Main CPU (MPU), 11a, 11b, 11c ... Switch, 12 ... Signal processing part, 13 ... Face detection part, 14 ... Contour detection part, 15 ... Flash ROM, 16 ... Auxiliary light irradiation part, 18 ... Focusing , 19 ... zoom control unit, 21 ... analog front end unit (AFE), 22 ... SDRAM, 23 ... memory, 24 ... microphone, 26 ... speaker, 27 ... display unit, 28 ... print signal output unit, 29 ... wireless communication Part 31, imaging element 32, photographing lens 34, subject.

Claims (26)

Face detection means for detecting a face portion included in the image signal from the image signal from the imaging means;
Determination means for determining the best image of the photographed image when the position of the face in the screen and the contour line connected from the vicinity of the face toward the outside of the screen have a predetermined shape;
A camera comprising: A face detection means for detecting a face portion included in the image signal from an image signal from an imaging means for monitoring an image in the shooting screen;
Determination means for determining the best image of the photographed image when a time change of the contour line connecting the vicinity of the face center is detected as a predetermined change pattern;
A camera comprising:   The camera according to claim 1, wherein the determination unit determines the predetermined shape when a tip of the contour line comes above the face.   The camera according to claim 1, wherein the determination unit determines the predetermined shape when the extending direction of the contour line comes below the face part.   2. The camera according to claim 1, wherein the determination unit determines a predetermined shape when a tip of the contour line is at a farthest distance from the face portion. 3.   The camera according to claim 1, wherein the determination unit determines that the extension portion of the contour line has a predetermined shape when the shape of an object having a shape different from the contour line is changed.   The camera according to claim 2, wherein the determination unit determines that the captured image is the best image when the change in movement of the contour line stops. Face detection means for detecting a face included in the captured image;
Skin color detection means for detecting the skin color of the face from the face detection means;
Determining means for determining the position of the hand of the photographed image according to the detection result of the skin color detecting means;
A camera comprising:   9. The camera according to claim 8, wherein the determination unit determines a good shot when the position of the hand is a predetermined position relative to the position of the face.   The camera according to claim 8, wherein the determination unit determines that the position of the hand is a good shot when both hands are assembled. Face detection means for detecting the position and size of the face included in the captured image;
From the face detection means, skin color detection means for detecting the skin color of the face;
Determining means for determining the position of the foot of the photographed image according to the skin color detection result;
A camera comprising: Face detection means for detecting the position and size of the face of the subject included in the captured image;
Determining means for determining whether the subject is a full-body image or a half-body image according to the output result of the face detection means;
A camera comprising: Face detection means for detecting the face of the subject included in the captured image;
Determination means for determining the position of at least a part of the subject detected by the face detection means in the photographed image;
A camera comprising: In a camera having a continuous shooting function for continuously shooting a plurality of images,
Imaging means for capturing an image of a subject;
Face detection means for detecting a face portion of the subject included in the image taken by the imaging means;
Determining means for determining the best image of the continuously captured images when the position of the face in the screen and the contour line connected from the vicinity of the face toward the outside of the screen have a predetermined shape; ,
A camera comprising: In a camera having a continuous shooting function for continuously shooting a plurality of images,
Imaging means for capturing an image of a subject;
Face detection means for detecting a face portion of the subject included in the image taken by the imaging means;
A determination unit that determines the best image of continuously captured images when a change in time of a contour line centering on the vicinity of the face is detected as a predetermined change pattern;
A camera comprising:   The camera according to claim 14, wherein the determination unit determines that the shape is a predetermined shape when a tip of the contour line reaches above the face portion in the continuously captured images.   The camera according to claim 14, wherein the determination unit determines the predetermined shape when the extension direction of the contour line reaches below the face portion in the continuously captured images.   The camera according to claim 14, wherein the determination unit determines the predetermined shape when a tip of the contour line reaches a distance farthest from the face portion in the continuously captured images. .   The determination unit determines that the predetermined shape is obtained when an extension of the contour line in the continuously captured images changes the shape of an object having a shape different from the contour line. The camera according to claim 14.   The camera according to claim 15, wherein the determination unit determines the best image of the photographed image when the movement of the contour portion stops among the continuously photographed images.   15. The camera according to claim 14, further comprising recording means for selecting and recording only appropriate images from the best images of the continuously photographed images. In a camera having a continuous shooting function for continuously shooting a plurality of images,
Imaging means for capturing an image of a subject;
Face detection means for detecting a face included in a captured image taken by the imaging means;
Skin color detection means for detecting the skin color of the face from the face detection means;
Determining means for determining the position of the hand of the photographed image according to the detection result of the skin color detecting means;
A camera comprising:   23. The determination unit according to claim 22, wherein the determination unit determines that a captured image in which the position of the hand is a predetermined position relative to the position of the face among the continuously captured images is a good shot. The listed camera.   23. The camera according to claim 22, wherein the determination unit determines that the captured image is a captured image when both hands are combined among the continuously captured images. In a camera having a continuous shooting function for continuously shooting a plurality of images,
Imaging means for capturing an image of a subject;
Face detection means for detecting the position and size of the face included in the captured image taken by the imaging means;
From the face detection means, skin color detection means for detecting the skin color of the face;
Determining means for determining the position of the foot of the photographed image according to the skin color detection result;
A camera comprising: In a camera having a continuous shooting function for continuously shooting a plurality of images,
Imaging means for capturing an image of a subject;
Face detection means for detecting the position and size of the face of the subject included in the captured image taken by the imaging means;
Determining means for determining whether the subject is a full-body image or a half-body image according to the output result of the face detection means;
A camera comprising:

Images