JP2014035706A - Position detecting device, position detecting method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make it possible to detect the position of a hand from a variety of images.SOLUTION: A detection operation for detecting a hand area corresponding to the hand of a person is detected from the image of each frame based on the hand color, which serves as a reference, and a detection operation for detecting the face of the person from the image of each frame is started in parallel during the display of a live view image(S3). During the detection of the hand area, position information indicating the position of the detected hand area in the image is acquired, and accumulated as locus data indicating a hand locus (S4). At a point in time that the face of a person is detected (S5: YES), the skin color of the face is detected from the image (S7). The detected color in the operation of detecting the hand area is changed to the skin color of the face from the hand color serving as a reference (S10). Thereafter, the accumulation of locus data is continued (S16). As long as only the face of a person is present in an image, the position of the hand of the person can be detected.

Description

  The present invention relates to a technique for detecting the position of a person's hand from an image.

  2. Description of the Related Art In recent years, a technique that enables remote control of home appliances and the like by photographing a person with a camera and recognizing a gesture caused by the movement of the person's hand is becoming widespread. In order to recognize a gesture, it is indispensable to detect the position of a hand from an image obtained by photographing a person. As a technique for detecting the position of a hand, for example, in Patent Document 1 below, a person's face is detected from an image, the posture of the person is estimated from the position of the face in the image, and then based on the estimated posture. A technique for detecting the position of a person's hand is described.

JP 2012-989888 A

  However, in the technique described in Patent Document 1, in order to detect the position of a person's hand from an image, not only the person's face is present in the image but also the posture of the person can be estimated. There was a problem that it was necessary to include parts other than the face.

  The present invention has been made in view of such conventional problems, and provides a position detection device, a position detection method, and a program capable of detecting the position of a person's hand from images in more various states. With the goal.

  In order to solve the above problems, in the present invention, face detection means for detecting a person's face from each image constituting a moving image, and face skin color information detected by the face detection means are detected. Detected by the hand region detecting means, the hand region detecting means for detecting the hand region corresponding to the hand of the person based on the color information of the hand from each image constituting the moving image, and the hand region detecting means. Position information acquisition means for acquiring position information indicating the position of the hand area in the image and the skin color information detected by the skin color detection means for hand color information used in the detection operation of the hand area by the hand area detection means. And a detection control means for changing based on the color information.

  According to the present invention, it is possible to detect the position of a person's hand from images in more various states.

It is a block diagram of a digital camera showing an embodiment of the present invention. It is a flowchart which shows the processing content in the automatic imaging | photography mode of a digital camera.

  Hereinafter, embodiments of the present invention will be described. FIG. 1 is a block diagram showing an outline of an electrical configuration of a digital camera 1 having a function as a position detecting device of the present invention.

  As shown in FIG. 1, the digital camera 1 includes a control unit 2, a lens unit 3, an imaging unit 4, a display unit 5, an image storage unit 6, a program storage unit 7, an operation unit 8, a face detection unit 9, a hand. Each part of the area | region detection part 10 and the hand detection part 11 is provided.

  The control unit 2 includes a CPU (Central Processing Unit), its peripheral circuits, and a working internal memory such as a RAM (Random Access Memory), and controls each unit of the digital camera 1.

  The lens unit 3 includes a lens group including a focus adjustment lens and a zoom lens, a motor that drives the lens group, a diaphragm, and an actuator that opens and closes the diaphragm to adjust the opening.

  The imaging unit 4 performs predetermined analog processing on a CCD (Charge Coupled Device) or CMOS (Complementary Meta10xide Semiconductor) type imaging device and an imaging signal output from the imaging device, and then converts the analog signal into a digital signal. It consists of AFE (Analog Front End). The imaging unit 4 sequentially captures the subject at a predetermined frame rate via the lens group of the lens unit 3 while the shooting mode is set in the digital camera 1, and controls the subject image data (imaging data). To supply.

  The image data supplied to the control unit 2 is converted into RGB data for each pixel in the control unit 2 and further converted into YUV data including a luminance (Y) component and a color difference (UV) component, and then to the display unit 5. Supplied and displayed on the display unit 5 as a live view image.

  Further, image data supplied from the imaging unit 4 to the control unit 2 at the time of shooting is converted into YUV data and then compressed by a JPEG (Joint Photographic Expert Group) method. Various attribute information is added to the compressed image data, and the image data is stored in the image storage unit 6 as a still image file compliant with the Exif (Exchangeable Image File Format) standard.

  The image storage unit 6 includes, for example, a flash memory built in the digital camera 1, various memory cards that can be attached to and detached from the digital camera 1, and a card interface that enables data input and output to the memory card.

  The image data stored as a still image file in the image storage unit 6 is read and expanded by the control unit 2 while the playback mode is set in the digital camera 1, and then supplied to the display unit 5 for screen display. Is done.

  The display unit 5 includes a color liquid crystal display panel and a display drive circuit that drives the color liquid crystal display panel in accordance with image data supplied from the control unit 2. As described above, the display unit 5 displays a subject image in the shooting mode. Live view display is performed, and in the playback mode, an existing captured image composed of image data stored as a still image file in the image storage unit 6 is displayed.

  The program storage unit 7 is configured by a nonvolatile memory in which stored data can be rewritten as needed, such as a ROM (Read Only Memory) or a flash memory, for example. The program storage unit 7 stores in advance a program for causing the control unit 2 to perform processing described later.

  The program storage unit 7 is used for a program for causing the control unit 2 to perform AE (Auto Exposure) control, AF (Auto Focus) control, AWB (Auto White Balance) control, and the like, and AE control. Various data such as data constituting a program diagram showing combinations of shutter speed, ISO sensitivity, and aperture value are also stored.

  The operation unit 8 includes a plurality of operation keys such as a power key, a shutter key, and a mode switching key used for switching between a photographing mode and a reproduction mode which are basic operation modes of the digital camera 1. The operation state of the operation key in the operation unit 8 is monitored by the control unit 2 as needed.

  Each of the face detection unit 9, the hand region detection unit 10, and the hand detection unit 11 is an image processing circuit for performing a specific image recognition process on an image captured at a predetermined frame rate by the imaging unit 4 in the shooting mode. And a memory storing various parameters, a working storage memory, and the like.

  The face detection unit 9 detects a face part of an arbitrary person in the image of each frame, acquires area information indicating an area corresponding to the detected face part (hereinafter referred to as a face area), and uses it as a control part 2 is supplied. Face detection by the face detection unit 9 is performed by searching for a specific portion in an image where eyes, noses, and mouths in a certain range of positional relationship exist using image recognition techniques such as binarization, contour extraction, and pattern matching. This is done by a known technique.

  The hand area detection unit 10 detects an area corresponding to a person's hand (hereinafter referred to as a hand area) from the image of each frame based on the color information of each pixel, and determines the position of the detected hand area in the image. The position information shown is acquired and supplied to the control unit 2.

  Specifically, the value of the H (Hue) component of the HSV color system is calculated as the color information of each pixel, and the H component value is centered on the color value of the person's hand for each frame image. A specific area composed of a group of pixels in a certain range is extracted, a difference between the specific areas is calculated between successive frames, and a portion where the difference is the maximum area is detected as a hand area.

  That is, the hand region detection unit 10 detects, as a hand region, a region corresponding to an arbitrary moving object having a color of a person's hand and a color similar thereto in the image. The color of the person's hand is set as a detection color by the control unit 2 as will be described later. In addition, the hand region detection unit 10 acquires, for example, coordinate information indicating the center position of the smallest circle or quadrangle surrounding the hand region as the position information of the detected hand region.

  The hand detection unit 11 detects the hand of an arbitrary person in the image and supplies position information indicating the position of the detected hand to the control unit 2. The hand detection by the hand detection unit 11 is based on shape recognition, and uses a large number of hand shape data prepared in advance to search for a human hand in an image by binarization, contour extraction, pattern matching, etc. Is done. In addition, the hand detection unit 11 acquires coordinate information indicating the center position of the smallest circle or rectangle surrounding the hand, for example, as position information indicating the position in the detected hand image.

  The digital camera 1 of this embodiment is provided with an automatic shooting mode as a lower mode of the shooting mode. The automatic shooting mode is a predetermined operation (gesture) in which the movement of a person's hand is detected from images sequentially captured during live view display in a shooting standby state, and the movement of the hand is determined in advance. In this mode, when it can be recognized, it is used as a trigger. Note that the gesture recognized by the digital camera 1 in the automatic shooting mode is a simple hand movement such as drawing a circle or an S-shape.

  Next, the operation of the digital camera 1 according to the present invention will be described with reference to FIG. Note that there are two types of operations to be described later: a gesture that the digital camera 1 recognizes in the automatic shooting mode is one that causes the gesture recognition to be shot only once, and a gesture that causes the gesture recognition to be repeatedly performed. This is the operation when.

  FIG. 2 is a flowchart showing processing executed by the control unit 2 in accordance with a program stored in the program storage unit 7 when the user sets the automatic shooting mode in the digital camera 1.

  The control unit 2 starts capturing the subject image by the imaging unit 4 together with the setting of the automatic shooting mode, and starts live view display on the display unit 5 (step S1).

  Furthermore, the control unit 2 sets a reference color that is a predetermined average hand color as a detection color that is the color of the hand region that the hand region detection unit 10 should detect (step S2). Detection of the hand region by the hand region detection unit 10 and face detection by the face detection unit 9 are started in parallel (step S3).

  Further, as the hand region detection unit 10 starts detecting the hand region, the control unit 2 uses the position information indicating the position in the image of the hand region acquired by the hand region detection unit 10 as the movement locus of the person's hand. The process of accumulating (storing) in the work internal memory as the locus data shown is started (step S4).

  Thereafter, until the face detection unit 9 can detect the face of an arbitrary person, the control unit 2 continues to accumulate trajectory data indicating the moving trajectory of the moving object with a predetermined waiting time as a limit (step S5: NO, step S12: NO).

  Eventually, when the face detection unit 9 can detect the face of an arbitrary person (step S5: YES), the control unit 2 temporarily stops the hand region detection operation by the hand region detection unit 10 at that time (step S6). .

  Subsequently, the control unit 2 detects the skin color of the face portion of an arbitrary person from the face area in the image detected by the face detection unit 9 (step S7). Specifically, the value of the H component of the HSV color system is acquired for each pixel of the face area indicated by the area information supplied from the face detection unit 9, and a color histogram is calculated. The component is acquired as color information indicating the skin color of the face portion.

  Further, the control unit 2 checks whether or not the detected color set in the process of step S2 is substantially the same color as the skin color of the face portion detected in the process of step S7 (step S8). Specifically, it is confirmed whether or not the difference between the detected color and the actual skin color in the face portion (difference in the H component value) is within a predetermined allowable range.

  Here, when the detected color is not substantially the same color as the skin color of the face portion, the control unit 2 may mean that the hand region detected by the hand region detecting unit 10 until then is not a region corresponding to a human hand. If it can be determined that the hand is high (step S8: NO), the hand trajectory data already accumulated is discarded (step S9).

  Subsequently, the control unit 2 updates the detection color to be detected by the hand region detection unit 10 to the skin color of the face acquired in step S7 (step S10), and resumes the hand region detection operation by the hand region detection unit 10. (Step S11). Then, the control unit 2 sequentially accumulates the position information of the hand region newly detected by the hand region detection unit 10 thereafter as the trajectory data indicating the movement trajectory of the hand (step S16).

  Accordingly, after this, after the face is detected by the face detection unit 9, only the position information of the hand region newly detected by the hand region detection unit 10 is accumulated as trajectory data indicating the hand movement trajectory. .

  In contrast to the above, when the detected color is almost the same as the skin color of the face portion, the control unit 2 is an area in which the hand area detected by the hand area detecting unit 10 until then corresponds to the human hand. When it can be determined that there is (step S8: YES), the hand region detection operation by the hand region detector 10 is immediately resumed (step S11). And the control part 2 accumulate | stores sequentially the positional information on the hand area newly detected by the hand area detection part 10 as locus | trajectory data which shows the movement locus | trajectory of a hand (step S16).

  Therefore, after this, the hand region newly detected after the face is detected by the face detector 9 following the existing trajectory data accumulated before the face is detected by the face detector 9. Are stored as trajectory data.

  On the other hand, after starting the face detection by the face detection unit 9 in the process of step S3, the control unit 2 does not detect the face of any person even after a predetermined waiting time has elapsed (step S12: YES). First, the detection of the hand area by the hand area detection unit 10 and the face detection by the face detection unit 9 are ended (step S13).

  Furthermore, after discarding the hand trajectory data that has already been accumulated (step S14), the control unit 2 starts hand detection by shape recognition in the hand detection unit 11 (step S15). Thereafter, the control unit 2 sequentially accumulates position information indicating the position of the hand in the image acquired by the hand detection unit 11 as trajectory data indicating the movement trajectory of the hand (step S16).

  Thereafter, each time the control unit 2 newly accumulates the position information acquired by the hand region detection unit 10 or the hand detection unit 11 as trajectory data indicating the movement trajectory of the hand, the movement trajectory of the hand indicated by the trajectory data. That is, it is determined whether or not the hand movement matches a predetermined movement (gesture) (step S17). That is, the control unit 2 performs gesture recognition.

  As described above, there are two types of gestures recognized here, and in the process of step S17, the control unit 2 determines that the movement locus of the hand indicated by the locus data is at least one of the two types of gestures. It is determined whether or not it matches the movement of the hand of the gesture.

  The control unit 2 continues to accumulate the trajectory data until it can be determined that the hand motion indicated by the trajectory data matches the predetermined motion (step S17: NO).

  When it is determined that the hand movement indicated by the trajectory data eventually coincides with the predetermined movement (step S17: YES), the control unit 2 performs a photographing process using this as a trigger (step S18). That is, the control unit 2 acquires an image for recording by the imaging unit 4 and stores the acquired image, that is, a captured image, in the image storage unit 6.

  Next, the control unit 2 confirms whether or not shooting should be terminated (step S19). In other words, the control unit 2 confirms whether or not the gesture that can be recognized in the process of step S <b> 17 is for performing the shooting by the gesture recognition only once.

  And the control part 2 is for making the gesture recognized by the process of step S17 perform the imaging | photography by gesture recognition only once, and should complete | finish imaging | photography (step S19: YES) ), And the processing in the automatic shooting mode is terminated at that time.

  On the other hand, if the gesture recognized in the process of step S17 is for repeatedly performing shooting by gesture recognition and the shooting should not be ended (step S19: NO), the control unit 2 has already performed. The accumulated hand trajectory data is once discarded (step S20). Then, the control unit 2 returns to the process of step S4 and starts accumulating new trajectory data, and then repeats the processes after step S5 described above.

  As a result, the user can perform a plurality of continuous shootings by performing a gesture for repeatedly performing shooting by gesture recognition within the angle of view. Also, for example, after a user has taken a picture of himself / herself as a subject, the user moves from a location within the angle of view to another location, and instead, the other user moves to a location within the angle of view and performs a predetermined gesture. Other users can also take pictures of themselves as subjects.

  After that, the control unit 2 performs the shooting process several times by the gesture recognition (step S18), and then the gesture recognized by the process of step S17 is to perform the shooting by the gesture recognition only once. If the shooting should be terminated (step S19: YES), the processing in the automatic shooting mode is terminated at that time.

  Although omitted in FIG. 2, the control unit 2 performs the photographing process in step S <b> 18 after a certain time (for example, 0.5 seconds) elapses after it is determined that the hand movement matches the predetermined movement. Thus, the user in a state after the predetermined gesture ends is photographed.

  As described above, in the digital camera 1 according to the present embodiment, when recognizing a gesture caused by hand movement in the operation in the automatic shooting mode, the hand region corresponding to the hand of the person in the image is determined based on the hand color information. The position of a person's hand is acquired by detecting. Therefore, as long as there is a person's face in the image, the position of the person's hand is reliably detected even if the image does not contain any part other than the face that can estimate the posture of the person. be able to. Therefore, the position of a person's hand can be detected from images in more various states.

  Moreover, in the present embodiment, the position of the person's hand is acquired before the person's face can be detected, and after the person's face has been detected, the detected area is updated to update the hand region from the image of each frame. The detection operation based on the skin color of the face of an actual person is changed from the detection operation based on the reference color which is a predetermined average hand color, that is, the reference hand color information. change. Therefore, after the person's face can be detected, a more accurate position can be acquired as the position of the person's hand than before the person's face can be detected.

  As a result, in this embodiment, it is possible to recognize a gesture caused by the movement of a person's hand at higher speed and more accurately.

  In this embodiment, when detecting a hand region from an image of each frame, a moving object having a reference color as a detection color and a color that approximates the detection color is set before the human face can be detected. An area corresponding to a part is detected as an area, and after a human face is detected, an area corresponding to a moving body part having a skin color of an actual person's face and a color similar to the skin color is detected as a hand area. I tried to do it.

  Therefore, the hand region corresponding to the person's hand can be detected more accurately from the image of each frame, and as a result, the gesture caused by the movement of the person's hand can be recognized more accurately.

  In the second and subsequent shootings, the detected color before the human face can be detected is the skin color of the face detected during the previous shooting.

  In the present embodiment, the actual color acquired when the detected face used for detecting the hand area from the image of each frame before the human face can be detected is detected when the human face can be detected. If the color is not the same as or similar to the skin color of the face, the position information of the hand area stored before the human face can be detected is discarded once.

  In other words, if the detected color before the human face can be detected is not the same or similar to the skin color of the actual human face, the position of the hand region acquired and stored after the human face is detected Based on only information, it was determined whether or not the hand movement matches the default movement (gesture).

  Therefore, if there is an error in the position information of a series of hand regions acquired before the person's face can be detected due to the fact that the set detection color is significantly different from the skin color of the actual person's face, an error in the gesture will occur. Recognition can be prevented in advance.

  Further, in this embodiment, when a human face cannot be detected from the image of each frame within a predetermined waiting time, hand detection by shape recognition is performed on the image of each frame, thereby Trajectory data (hand position information) indicating the movement trajectory was acquired. Therefore, even in a situation where a person's face cannot be detected, it is possible to reliably recognize a gesture caused by a person's hand movement.

  In this embodiment, a reference color, which is a predetermined average hand color, is set as a detection color when detecting a hand area from an image of each frame before a human face can be detected. However, a color after correcting the reference color to a color closer to the skin color of the user may be set as the detected color. When correcting the reference color, for example, it may be used to infer the user's skin color such as the language used in the digital camera 1 and the user's personal information (name, email address, etc.). Any information that can be used can be used.

  Furthermore, a plurality of types of skin colors may be prepared in advance as reference colors, and the user may select a skin color to be used as a reference color when the automatic shooting mode is set or at an earlier stage. . In that case, a sample face photo obtained by processing each face (skin) color into a different reference color may be prepared, and the user may select an arbitrary face, that is, a reference color.

  In the present embodiment, the digital camera 1 including the face detection unit 9, the hand region detection unit 10, and the hand detection unit 11 has been described. However, when the present invention is implemented, for example, the face detection unit 9, the hand detection unit 9, Each of the image recognition processes described above by the respective units of the area detection unit 10 and the hand detection unit 11 may be performed by the control unit 2 in accordance with a program stored in advance in the program storage unit 7.

  Further, the present invention can be applied not only to a digital camera but also to an arbitrary device such as another home appliance. Furthermore, the present invention is also applicable to, for example, supplying position information indicating the position of a person's hand in an image of each frame to an external device in real time by wire or wireless, and causing the external device to perform gesture recognition. Can do.

As mentioned above, although several embodiment of this invention and its modification were demonstrated, if these are in the range in which the effect of this invention is acquired, it can change suitably, and embodiment after change is also a claim. It is included in the scope of the invention described in the scope of the invention and the invention equivalent to the invention.
The invention described in the scope of the claims of the present application will be appended below.
[Claim 1]
Face detecting means for detecting the face of a person from each image constituting the moving image, skin color detecting means for detecting color information of the skin of the face detected by the face detecting means, and each constituting the moving image From the image, a hand region detecting means for detecting a hand region corresponding to a person's hand based on hand color information, and position information indicating the position of the hand region detected by the hand region detecting means in the image are acquired. And a detection control unit that changes hand color information used in the hand region detection operation by the hand region detection unit based on the skin color information detected by the skin color detection unit. A position detecting device characterized by that.
[Claim 2]
2. The hand region detection unit, when no face is detected by the face detection unit, detects a hand region corresponding to a human hand based on reference hand color information. The position detection device described.
[Claim 3]
The detection control means changes the color information of the hand used in the detection operation of the hand area by the hand area detection means every time the skin color information detected by the skin color detection means changes. Item 3. The position detection device according to Item 1 or 2.
[Claim 4]
Storage means for sequentially storing position information indicating the position of the hand region in each of the images acquired by the position acquisition means, and movement of a person's hand based on a series of position information stored in order in the storage means The position detection apparatus according to claim 1, further comprising: a recognition unit that recognizes a predetermined gesture by.
[Claim 5]
The color indicated by the hand color information used in the hand region detection operation by the hand region detection means is the same color as or similar to the color indicated by the skin color information detected by the skin color detection means. A judgment means for judging whether or not there is a color, and the recognition means uses the judgment means to determine whether the color indicated by the color information of the hand used in the hand region detection operation by the hand region detection means is the color; When it is determined that the color is not the same as or similar to the color indicated by the skin color information detected by the skin color detection means, the face acquisition means acquires the face after the face is detected by the face detection means, and 5. The position detection apparatus according to claim 4, wherein a predetermined gesture caused by a movement of a person's hand is recognized based only on the position information stored in the storage means.
[Claim 6]
The hand region detecting means detects, as the hand region, a region corresponding to a moving body part having a color indicated by the color information of the hand and a color similar to the color from each of the images. The position detection device according to claim 1.
[Claim 7]
Hand detection means for detecting a person's hand from each of the images using a shape recognition technique when the face detection means cannot detect a person's face within a predetermined time after the start of the detection operation; The position acquisition means is position information indicating the position of the hand area detected by the hand area detection means when the face detection means cannot detect a human face within a predetermined time after the start of the detection operation. 7. The position detection device according to claim 1, wherein position information indicating a position in each image of a person's hand detected by the hand detection unit is acquired instead of the position information.
[Claim 8]
A step of detecting a person's face from each image constituting the moving image, a step of detecting a skin color of the detected face, and a person based on hand color information from each image constituting the moving image Detecting a hand region corresponding to the hand of the user, obtaining a positional information indicating a position of the detected hand region in the image, and color information of the hand used for detecting the hand region of the facial skin. And a step of changing based on the color information.
[Claim 9]
A computer having a position detection device that detects the position of a person's hand from each image constituting a moving image, a process for detecting a person's face from each image constituting the moving image, and a skin of the detected face A process for detecting color information, a process for detecting a hand region corresponding to the hand of a person based on the color information of the hand from each image constituting the moving image, and a position of the detected hand region in the image A program for executing a process for acquiring position information indicating the position and a process for changing color information of a hand used for detection of the hand region based on the color information of the skin of the face.

DESCRIPTION OF SYMBOLS 1 Digital camera 2 Control part 3 Lens part 4 Imaging part 5 Display part 6 Image memory | storage part 7 Program memory | storage part 8 Operation part 9 Face detection part 10 Hand area detection part 11 Hand detection part

Claims (9)

Face detecting means for detecting a human face from each image constituting the moving image;
Skin color detection means for detecting color information of the skin of the face detected by the face detection means;
Hand region detecting means for detecting a hand region corresponding to a person's hand based on hand color information from each image constituting the moving image;
Position acquisition means for acquiring position information indicating the position in the image of the hand area detected by the hand area detection means;
And a detection control unit configured to change hand color information used in the hand region detection operation by the hand region detection unit based on the skin color information detected by the skin color detection unit. Detection device.   2. The hand region detection unit, when no face is detected by the face detection unit, detects a hand region corresponding to a human hand based on reference hand color information. The position detection device described.   The detection control means changes the color information of the hand used in the detection operation of the hand area by the hand area detection means every time the skin color information detected by the skin color detection means changes. Item 3. The position detection device according to Item 1 or 2. Storage means for sequentially storing position information indicating the position of the hand region in each of the images acquired by the position acquisition means;
The position detecting device according to claim 1, further comprising: a recognizing unit that recognizes a predetermined gesture caused by movement of a person's hand based on a series of position information sequentially stored in the storage unit. . The color indicated by the hand color information used in the hand region detection operation by the hand region detection means is the same color as or similar to the color indicated by the skin color information detected by the skin color detection means. A judgment means for judging whether or not there is,
The recognizing unit uses the skin color information detected by the skin color detecting unit to detect the color indicated by the hand color information used in the hand region detecting operation by the hand region detecting unit. When it is determined that the color is not the same as or similar to the displayed color, the person is based only on the position information acquired by the position acquisition unit and stored in the storage unit after the face is detected by the face detection unit. The position detection device according to claim 4, wherein a predetermined gesture caused by the movement of the hand is recognized.   The hand region detecting means detects, as the hand region, a region corresponding to a moving body part having a color indicated by the color information of the hand and a color similar to the color from each of the images. The position detection device according to claim 1. When the face detection means cannot detect a human face within a predetermined time after the start of the detection operation, the face detection means further comprises a hand detection means for detecting a person's hand from each image using a shape recognition technique,
A position indicating a position in the image of the hand region detected by the hand region detecting unit when the face detecting unit cannot detect a human face within a predetermined time after the start of the detection operation; The position detection device according to any one of claims 1 to 6, wherein, instead of information, position information indicating a position in each image of a person's hand detected by the hand detection unit is acquired. Detecting a person's face from each image constituting the moving image;
Detecting the color of the detected skin of the face;
Detecting a hand region corresponding to a person's hand based on hand color information from each image constituting the moving image;
Obtaining position information indicating the position of the detected hand region in the image;
Changing the hand color information used for detecting the hand region based on the color information of the skin of the face. In a computer having a position detection device that detects the position of a person's hand from each image constituting a moving image,
A process of detecting a human face from each image constituting the moving image;
A process for detecting the color information of the detected facial skin;
A process of detecting a hand region corresponding to a person's hand based on hand color information from each image constituting the moving image;
Processing for obtaining position information indicating the position of the detected hand region in the image;
A program for executing a process of changing color information of a hand used for detection of the hand region based on color information of the skin of the face.

Images