JP2008234578A - Face detecting device, face detecting method and face detecting program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology for efficiently executing high-accuracy facial detection processing, using a simple constitution. <P>SOLUTION: A face-detecting condition storage part 16 stores partial area information for dividing an image photographed by an imaging means 11 into a plurality of partial areas, and a facial detection conditions preset with respective partial areas. A face-detecting part 13 omits the facial detection processing that does not correspond to the face-detecting conditions set in the belonging partial area of a facial detection position, when executing the face-detecting processing for detecting a face in the image. Here, various conditions, such as, the facial size condition for determining the size of the face that is set as a detecting object, a facial direction condition for determining the direction of the face that is the detecting object and illumination condition, can be selected as the face-detecting conditions. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a face detection apparatus that detects a face from an image.

  A face detection device that detects a face from an image is desired to detect the face with high accuracy.

  As a conventional technique in view of such a problem, there is an object detection device that compares an estimated distance with a face calculated from an input image and a measured distance measured by a distance sensor to determine whether it is a face or a non-face ( Patent Document 1).

  However, since this apparatus requires a distance sensor in addition to the face detection unit, the apparatus configuration becomes large and the apparatus cost also increases. Furthermore, since the information on the measured distance measured by the distance sensor is used for the face detection process, the processing load increases.

  On the other hand, there is an image processing apparatus that three-dimensionally restores images taken by two cameras (see Patent Document 2). This apparatus detects a face using the three-dimensional information of the image.

However, since this apparatus requires two cameras, the apparatus configuration becomes large and the apparatus cost increases. Furthermore, since the three-dimensional information of the image is used, the processing load increases.
JP-A-2005-78376 JP 2005-301833 A

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a technique for efficiently executing highly accurate face detection processing with a simple configuration.

  In order to achieve the above object, the present invention adopts the following configuration.

  The face detection apparatus according to the present invention stores an imaging means, partial area information for dividing an image photographed by the imaging means into a plurality of partial areas, and a face detection condition preset for each partial area. And a face detection unit that omits face detection processing that does not meet the face detection condition set in the partial region to which the face detection position belongs when executing face detection processing for detecting a face from the image And comprising.

  In the present invention, different face detection conditions can be set for each partial region. Since the face detection process of the present invention uses a face detection condition set in advance for each partial region, an improvement in accuracy can be expected compared to the face detection process in which the entire image is performed under the same condition. Furthermore, the face detection process of the present invention can eliminate the face detection process that does not correspond to the face detection condition set for each partial region, and therefore can be more efficient than the face detection process that performs the entire image under all conditions. Further, while the conventional technique requires a distance sensor and a second camera, the present invention does not require an additional device, and thus has a simple configuration.

  Here, various conditions such as a face size condition that determines the size of the face to be detected, a face direction condition that determines the orientation of the face to be detected, and an illumination condition can be selected as the face detection condition.

  By setting the face size condition for each partial area, it is possible to exclude faces with an impossible size from the beginning, so that reduction of false detection can be expected. Specifically, the face detection means has a plurality of face detection windows corresponding to different face sizes, and when performing face detection processing using the face detection window from the image, face detection is performed. It is preferable to perform face detection processing using a face detection window according to the face size condition set in the partial region to which the position belongs. As a result, the face detection process can be omitted in the face detection window other than the face detection window according to the face size condition set for each partial region, and thus face detection performed in all face detection windows for the entire image. More efficient than processing.

  By setting the face orientation condition for each partial region, an improvement in detection accuracy and a reduction in false detection can be expected. Specifically, the face detection means includes a plurality of face detectors corresponding to different face orientations, and when performing face detection processing using the face detector from the image, the face detection position is detected. The face detection process may be performed using a face detector corresponding to the face orientation condition set for the partial region to which the image belongs. Accordingly, the face detection process can be omitted by using all the face detectors for the entire image because the face detection process by a face detector other than the face detector corresponding to the face orientation condition set for each partial region can be omitted. More efficient than face detection processing.

  By setting the illumination condition for each partial region, it is possible to expect improvement in detection accuracy and reduction in false detection. Specifically, the face detection means has a plurality of face detectors corresponding to different illumination environments, and when performing face detection processing using the face detector from the image, the face detection position is detected. The face detection process may be performed using a face detector corresponding to the illumination condition set for the partial region to which the image belongs. As a result, the face detection processing can be omitted by using a face detector other than the face detector according to the illumination condition set for each partial region, and therefore the face to be performed using all the face detectors for the entire image. More efficient than detection processing.

  In the present invention, face detection processing may be performed according to any one of the face detection conditions described above, or face detection processing may be performed according to a combination of two or more face detection conditions. . If a plurality of face detection conditions are combined, the face detection process can be performed under further narrowed conditions, so that the efficiency of the face detection process is improved.

  Further, the present invention may be regarded as a face detection device having at least a part of the above means, a face detection method including at least a part of the above process, a face detection program for realizing such a method, and It can also be understood as a recording medium on which the program is recorded. Each of the above means and processes can be combined with each other as much as possible to constitute the present invention.

  According to the present invention, face detection conditions can be set in advance for each partial region in an image, and no additional device is required. Therefore, highly accurate face detection processing can be efficiently executed with a simple configuration. is there.

  Exemplary embodiments of the present invention will be described in detail below with reference to the drawings.

<First Embodiment>
In the first embodiment, an example in which the present invention is applied to a monitoring device for monitoring a passerby or an intruder is given. In this type of monitoring apparatus, as shown in FIG. 1, the imaging unit 11 is fixed to a ceiling, a wall, or the like that is higher than a person's head, and in many cases, a captured image overlooking the monitoring area is obtained. Under this circumstance, the image captured by the imaging unit 11 tends to appear as a small face at the top of the image and a large face at the bottom of the image as shown in FIG. Focusing on this tendency, the face detection process in this embodiment divides the image captured by the imaging unit 11 into a plurality of partial areas, and changes the size of the face to be detected for each partial area. Note that the present invention can be preferably applied to a fixed camera such as the monitoring apparatus of the present embodiment, but the present invention is not limited to a fixed camera and can be applied to any photographing apparatus.

<Device configuration>
FIG. 3 is a block diagram showing a functional configuration of the face detection apparatus according to the embodiment of the present invention. The face detection apparatus generally includes an imaging unit 11, an image storage unit 12, a face detection unit 13, a display unit 14, a result storage unit 15, and a face detection condition storage unit 16. In the present embodiment, these functional elements are realized by an arithmetic processing unit of a computer executing software (program) and controlling hardware resources such as a camera, a memory, and a display as necessary. However, these functional elements may be configured by a dedicated chip.

  As the imaging unit 11, a digital camera including an optical system and an imaging element (CCD, CMOS sensor, etc.) can be preferably applied.

  The image storage unit 12 is a storage device that temporarily stores an image to be processed. As the storage device, any specific technique such as a volatile memory or a nonvolatile memory may be applied.

  The face detection unit 13 has a function of executing face detection processing for detecting a face from an image by image processing. Any existing technique may be applied to the face detection processing by the face detection unit 13. For example, a method for detecting a face by template matching using a reference template corresponding to the contour of the entire face, a method for detecting a face by template matching based on a facial organ (eg, eyes, nose, ears), chroma key processing Detecting vertices such as the head, detecting a face based on the vertices, detecting a region close to the skin color, detecting that region as a face, and learning with a teacher signal using a neural network And a method for detecting a face-like area as a face.

  As the display unit 14, any specific technique such as a liquid crystal display may be applied.

  The result storage unit 15 is a storage device that stores the result of the face detection process performed by the face detection unit 13. As the storage device, any specific technique such as a nonvolatile memory or a hard disk may be applied.

  The face detection condition storage unit 16 is a storage device that stores partial area information for dividing an image captured by the imaging unit 11 into a plurality of partial areas, and face detection conditions set in advance for each partial area. is there. As the storage device, any specific technique such as a nonvolatile memory or a hard disk may be applied.

An example of partial area information and face detection conditions in this embodiment is shown in FIG. As described above, the image photographed by the imaging unit 11 of the present embodiment has a tendency that the face at the top of the image appears small and the face at the bottom of the image appears large as shown in FIG. Therefore, in the example of FIG. 4, information that divides the entire image into three partial areas 21 to 23 called “upper part of image”, “middle part of image”, and “lower part of image” is set as partial area information. Here, a quadrangular partial area is set, and the partial area is defined by two coordinates of a start point coordinate and an end point coordinate of a vertex located at the diagonal of the quadrangle. The shape of the partial area is not a quadrangle, but may be a polygon or a circle. The partial area information may be in any format as long as the shape, position, and size of the partial area can be defined. Further, a face size condition that determines the size of the face to be detected is set as the face detection condition (here, “small” in the partial area 21 at the top of the image, “medium” in the partial area 22 in the middle of the image, and the lower part of the image) The face size condition of “Large” is set in the partial area 23). The face size condition may be an attribute such as “large”, “medium”, or “small”, a numerical value corresponding to the size of the face, or the size of a face detection window (described later). .

<Face detection function>
The function and processing flow of the face detection apparatus of this embodiment will be described with reference to the flowcharts of FIGS.

  The face detection apparatus according to the present embodiment executes the face detection function at regular time intervals. When the face detection function is activated, an image captured by the imaging unit 11 is captured for use in face detection processing (step S11). The input image is stored in the image storage unit 12 together with shooting time information. Note that the user can check the face detection processing image on the display unit 14 as necessary.

  Next, the face detection unit 13 reads the partial area information and the face detection condition set in advance for each partial area from the face detection condition storage unit 16 (step S12), and the face corresponding to the face detection condition for each partial area. A detection process is performed (step S13).

  FIG. 6 shows a specific example of the face detection process in step S13.

  First, the face detection unit 13 places a face detection position at an initial position (for example, (X21, Y21) in FIG. 2) in the image captured by the imaging unit 11 (step S21).

  Next, the face detection unit 13 identifies the partial area to which the face detection position belongs by comparing the coordinates of the face detection position with the partial area information (step S22). Then, the face detection unit 13 selects a face detection window to be applied to the face detection position based on the face size condition set in the specified partial region.

  The face detection window defines the size of the face to be detected (that is, only faces of the same size as the face detection window are detected, and a face that is clearly smaller or larger than the face detection window) Face is not detected.) The face detection unit 13 of the present embodiment has a plurality of types of face detection windows 24a to 24i having different sizes as shown in FIG. 2 in order to correspond to faces of various sizes. However, the face detection unit 13 does not uniformly apply all the face detection windows to the entire image, but changes the face detection window to be applied for each partial region according to the face size condition. In the example of FIG. 2, when the face detection position is in the partial area 21, the “small” face detection windows 24 a to 24 c are selected, and when the face detection position is in the partial area 22, the “medium” face is selected. When the face detection windows 24d to 24f are selected and the face detection position is in the partial area 23, the face detection windows 24g to 24i for the “large” face are selected.

  Then, the face detection unit 13 sets one of the face detection windows selected in step S22 as a face detection position (step S23), and executes face detection processing (step S24). Next, the face detection unit 13 determines whether there is a face detection window that is not used in the face detection process among the plurality of face detection windows selected in step S22 (step S25). When it is determined that there is another face detection window (step S25; YES), the face detection window is set at the face detection position (step S23), and the face detection process is executed (step S24). When face detection processing has been performed for all the face detection windows selected in step S22 (step S25; NO), the processing for the face detection position ends, and the process proceeds to step S26.

In step S26, the face detection unit 13 determines whether there is a next face detection position. If there is a next face detection position (step S26; YES), the face detection position is moved to the next face detection position (step S26). Step S27). Then, face detection processing is also performed for the next face detection position using a face detection window of an appropriate size corresponding to the partial area (steps S22 to S25). Steps S22 to S25 are repeated until there is no next face detection position (until the end of the image is reached) (step S26). Note that the movement amount of the face detection position and the movement path 25 may be programmed in advance in the face detection unit 13 (FIG. 2 shows that the face detection position moves 1 pixel at a time in the X direction and reaches 1 at the end of the line. An example of moving in the pixel Y direction is shown.).

  Next, the face detection processing result of the face detection unit 13 is stored in the result storage unit 15 (step S14 in FIG. 5). The face detection process result stored in the result storage unit 15 is preferably an image used for the face detection process, its photographing time, the number of detected faces, and information on its position coordinates. The user can confirm the face detection processing result on the display unit 14 by inputting the photographing time and the display method of the face detection processing result according to the purpose. The display method may be the total number of face detections at the input shooting time, or may display an image shot at the shooting time and mark the face detection position. In this embodiment, the face detection process result is stored in the result storage unit 15 after the face detection process (step S13). However, the face detection process result is stored in the result storage unit during the face detection process (for example, after step S24). 15 may be stored.

  As described above, in the present embodiment, only the three face detection windows 24a to 24c are used in the partial area 21 at the top of the image, and the face detection processing using the other face detection windows 24d to 24i is omitted. Similarly, in the partial area 22, the face detection processing by the face detection windows 24a to 24c and 24g to 24i is omitted, and in the partial area 23, the face detection processing by the face detection windows 24a to 24f is omitted. Therefore, compared to using all the face detection windows 24a to 24i for the entire image as shown in FIG. 7, it is possible to improve the efficiency of the face detection process (reduction in the amount of calculation). Further, by setting an appropriate face size condition for each partial area in advance, it is possible to exclude a face having an impossible size from the beginning, and therefore, it is possible to expect a reduction in false detection.

  In the face detection window switching method according to the present embodiment, the face detection window is switched for each face detection position, but the face detection process is performed after performing face detection processing for one entire partial area using one face detection window. The face detection window may be switched by moving the detection position to the initial position of the partial area.

<Second Embodiment>
In the second embodiment, a face detection apparatus will be described in a situation where the imaging unit 11 is at the intersection of the T-shaped passage and there are passages in the front and left and right directions of the imaging unit 11 (FIG. 8). . Under this circumstance, the image captured by the imaging unit 11 tends to show a larger face at the lower part of the image, as in the first embodiment. Furthermore, in the upper part of the image, people pass through the front passage, so there are many faces facing the front. In the lower part of the image, people passing through the left and right passages appear, so there is a tendency for many left and right faces to appear (FIG. 9). Focusing on such a tendency, the face detection processing in the present embodiment changes the size and orientation of the face to be detected for each partial region.

  Hereinafter, the configuration and processing different from those of the first embodiment will be mainly described.

<Device configuration>
The face detection unit 13 of this embodiment includes a plurality of types of face detectors 35a to 35e as shown in FIG.

Here, the face detectors 35a to 35e are face detectors corresponding to different face orientations. The face detector 35a is a face detector for a front face, and is a face detector specialized in detecting a face facing the front (for example, the front is 0 ° and the face is ± 10 ° from side to side). . This can be realized by using only a front-facing face image as a learning teacher image or adopting a detection algorithm dedicated to the front face. Similarly, the face detector 35b is a face detector for a right oblique face specialized in detecting a right oblique face, the face detector 35c is a face detector for a left oblique face, and the face detector 35d is a right detector. The face detector for the face, the face detector 35e, is a face detector for the left face.

  Next, FIG. 11 shows an example of partial area information and face detection conditions in the present embodiment. As described above, the image photographed by the imaging unit 11 of the present embodiment tends to have a larger face as the lower part of the image as shown in FIG. Many faces are shown, and people in the middle of the image pass through the vicinity of the intersection between the front passage and the left and right passages, so there are many faces that appear diagonally left and right, and people in the lower part of the image pass through the left and right passages. , There is a tendency to see a lot of left and right faces. Therefore, in the example of FIG. 11, information that divides the entire image into three partial areas 31 to 33, “upper part of image”, “middle part of image”, and “lower part of image” is set as the partial area information (partial here) The definition of the area is the same as in the first embodiment). Further, as face detection conditions, face size conditions and face orientation conditions that determine the orientation of the face to be detected are set (here, “small” and “front” in the partial area 31 at the top of the image, and partial areas in the middle of the image The face size condition and the face orientation condition of “medium” and “diagonal” are set in 32, and “large” and “left and right” are set in the partial area 33 below the image). The face orientation condition may be an attribute such as “front”, “diagonal”, “left / right” as in the face size condition described in the first embodiment, or may be a numerical value corresponding to the face orientation. It may be an identifier of a face detector.

<Face detection function>
Since the function of the face detection apparatus of this embodiment is the same as that of the first embodiment except for the face detection process in step S13, the description thereof is omitted, and here, according to the flowchart of FIG. The face detection process (step S13) will be described in detail.

  First, the face detection unit 13 places the face detection position at the initial position in the image captured by the imaging unit 11 (step S31).

  Next, the face detection unit 13 identifies the partial area to which the face detection position belongs by comparing the coordinates of the face detection position with the partial area information (step S32). Then, the face detection unit 13 selects a face detection window and a face detector to be applied to the face detection position based on the face size condition and face orientation condition set in the specified partial region.

  As described above, the face detection unit 13 of this embodiment includes a plurality of types of face detectors 35a to 35e corresponding to different face orientations in addition to a plurality of types of face detection windows 34a to 34i having different sizes. Yes. However, the face detection unit 13 does not apply all face detection windows and face detectors uniformly to the entire image, but applies the face detection window and face detector to be applied according to face size conditions and face orientation conditions. It changes for each partial area. In the example of FIG. 9, when the face detection position is in the partial area 31, the face detection windows 34 a to 34 c for “small” face and the face detector 35 a for “front” face are selected, and the face detection position is partially If it is in the area 32, the face detection windows 34 d to 34 f for the “medium” face and the face detectors 35 b and 35 c for the “oblique” face are selected, and “large” if the face detection position is in the partial area 33. Face detection windows 34g-34i for faces and face detectors 35d and 35e for "left and right" faces are selected.

Then, the face detection unit 13 sets one of the combinations of the face detection window and the face detector selected in step S32 as a face detection position (step S33), and executes face detection processing (step S34). . Next, the face detection unit 13 determines whether there is a face detector that is not used in the face detection process among the plurality of face detectors selected in Step S32 (Step S35). If it is determined that there is another face detector (step S35; YES), the face detector is set at the face detection position (step S33), and face detection processing is executed (step S34). When face detection processing is performed for all the face detectors selected in step S32 (step S35; NO), face detection that is not used in face detection processing among the plurality of face detection windows selected in step S32 It is determined whether there is a window (step S36). When face detection processing is performed for all combinations of the face detection window and face detector selected in step S32 (step S36; NO), the processing for the face detection position is terminated, and the process proceeds to step S37.

  In step S37, the face detection unit 13 determines whether there is a next face detection position. If there is a next face detection position (step S37; YES), the face detection position is moved to the next face detection position (step S37). Step S38). Then, the face detection process is also performed for the next face detection position using a face detection window of an appropriate size corresponding to the partial area and a face detector having an appropriate face orientation (steps S32 to S36). Steps S32 to S36 are repeated until there is no next face detection position (until the end of the image is reached) (step S37).

  As described above, in the present embodiment, only the three face detection windows 34a to 34c and one face detector 35a are used in the partial area 31 at the top of the image, and the other face detection windows 34d to 34i and face detection are performed. The face detection process by the devices 35b to 35e is omitted. Similarly, in the partial area 32, face detection processing by the face detection windows 34a to 34c and 34g to 34i and the face detectors 35a, 35d and 35e is omitted, and in the partial area 33, the face detection windows 34a to 34f and face detection are performed. The face detection process by the devices 35a to 35c is omitted. Therefore, compared to using all the face detection windows 34a to 34i and the face detectors 35a to 35e for the entire image, it is possible to improve the efficiency of face detection processing (reduction in the amount of calculation). In addition, by setting an appropriate face orientation condition for each partial region in advance, an improvement in detection accuracy and a reduction in false detection can be expected.

  In the partial area information of the present embodiment, a common partial area is set for the face size condition and the face orientation condition, but independent partial areas may be set. Further, the face detection window and face detector switching method switches the face detection window and face detector for each face detection position, but one set of face detection window and face detector is used for the entire partial area. After performing the face detection processing, the face detection position may be moved to the initial position of the partial area, and the face detection window may be switched.

<Third Embodiment>
In the third embodiment, a face detection apparatus will be described in a situation where the imaging unit 11 is installed so as to face indoors and outdoors, and an image captured by the imaging unit 11 includes both indoors and outdoors. (FIG. 13). Under this circumstance, the image captured by the imaging unit 11 tends to show a larger face at the lower part of the image, as in the first embodiment. Furthermore, as shown in FIG. 14, the lighting environment tends to be different between the indoor face and the outdoor face (for example, when the outdoor is brighter than the indoor, the outdoor face appears bright, and the indoor face becomes dark due to backlight. Etc.) Focusing on such a tendency, the face detection process in the present embodiment changes the size of the face to be detected and the illumination environment for each partial region.

  Hereinafter, the configuration and processing different from those of the first embodiment will be mainly described.

<Device configuration>
The face detection unit 13 of the present embodiment includes a plurality of types of face detectors 45a to 45e as shown in FIG.

  Here, the face detectors 45a to 45e are face detectors corresponding to different illumination environments. The face detector 45a is a face detector for a bright face, and is a face detector specialized for detecting a face under a bright illumination environment. This can be realized by using only a face image under a bright illumination environment as a learning teacher image or adopting a detection algorithm dedicated to a bright face. Similarly, the face detector 45b is a face detector for a slightly bright face specialized for detection of a face in a slightly bright illumination environment, and the face detector 45c is a face detector for a normal face, a face detector 45d. The face detector for a slightly dark face and the face detector 45e are face detectors for a dark face.

  Next, FIG. 16 shows an example of partial area information and face detection conditions in the present embodiment. As described above, the image captured by the imaging unit 11 of the present embodiment tends to have a larger face at the lower part of the image as shown in FIG. 14, and the illumination environment includes an indoor face and an outdoor face. Tend to be different. Therefore, in the example of FIG. 16, information for dividing the entire image into three partial areas 41 to 43 of “upper part of image”, “middle part of image”, and “lower part of image” is set as the partial area information (parts here) The definition of the area is the same as in the first embodiment). In addition, as face detection conditions, face size conditions and illumination conditions that determine the illumination environment of the face to be detected are set (here, “small” and “bright” in the partial area 41 at the top of the image, and the partial area in the middle of the image The face size condition and the illumination condition of “medium” and “dark” are set in 42, and “large” and “dark” are set in the partial area 43 below the image. The illumination condition may be an attribute such as “bright” or “dark” as in the face size condition described in the first embodiment, may be a numerical value corresponding to the illumination environment, or may be an identifier of the face detector. Etc.

<Face detection function>
Since the function of the face detection apparatus of this embodiment is the same as that of the first embodiment except for the face detection process in step S13, the description thereof is omitted, and here, according to the flowchart of FIG. The face detection process (step S13) will be described in detail.

  First, the face detection unit 13 places the face detection position at the initial position in the image captured by the imaging unit 11 (step S31).

  Next, the face detection unit 13 identifies the partial area to which the face detection position belongs by comparing the coordinates of the face detection position with the partial area information (step S32). Then, the face detection unit 13 selects a face detection window and a face detector to be applied to the face detection position based on the face size condition and the illumination condition set in the specified partial region.

  As described above, the face detection unit 13 of this embodiment includes a plurality of types of face detectors 45a to 45e corresponding to different illumination environments in addition to a plurality of types of face detection windows 44a to 44i having different sizes. Yes. However, the face detection unit 13 does not apply all face detection windows and face detectors uniformly to the entire image, but partially applies the face detection windows and face detectors to be applied according to face size conditions and illumination conditions. It changes from region to region. In the example of FIG. 14, when the face detection position is in the partial area 41, the face detection windows 44a to 44c for the “small” face and the face detectors 45a and 45b for the “bright” face are selected, and the face detection position is selected. Is in the partial area 42, the face detection windows 44d to 44f for the “medium” face and the face detectors 45d and 45e for the “dark” face are selected, and when the face detection position is in the partial area 43, “ Face detection windows 44g-44i for the “large” face and face detectors 45d and 45e for the “dark” face are selected.

Then, the face detection unit 13 sets one of the combinations of the face detection window and the face detector selected in step S32 as a face detection position (step S33), and executes face detection processing (step S34). . Next, the face detection unit 13 determines whether there is a face detector that is not used in the face detection process among the plurality of face detectors selected in Step S32 (Step S35). If it is determined that there is another face detector (step S35; YES), the face detector is set at the face detection position (step S33), and face detection processing is executed (step S34). When face detection processing is performed for all the face detectors selected in step S32 (step S35; NO), face detection that is not used in face detection processing among the plurality of face detection windows selected in step S32 It is determined whether there is a window (step S36). When face detection processing is performed for all combinations of the face detection window and face detector selected in step S32 (step S36; NO), the processing for the face detection position is terminated, and the process proceeds to step S37.

  In step S37, the face detection unit 13 determines whether there is a next face detection position. If there is a next face detection position (step S37; YES), the face detection position is moved to the next face detection position (step S37). Step S38). Then, the face detection process is also performed for the next face detection position using a face detection window of an appropriate size corresponding to the partial area and a face detector having an appropriate face orientation (steps S32 to S36). Steps S32 to S36 are repeated until there is no next face detection position (until the end of the image is reached) (step S37).

  As described above, in the present embodiment, only the three face detection windows 44a to 44c and the two face detectors 45a and 45b are used in the partial area 41 at the top of the image, and the other face detection windows 44d to 44i are used. The face detection process by the face detectors 45c to 45e is omitted. Similarly, in the partial area 42, face detection processing by the face detection windows 44a to 44c and 44g to 44i and the face detectors 45a to 45c is omitted, and in the partial area 43, the face detection windows 44a to 44f and the face detector 45a are omitted. The face detection process by ~ 45c is omitted. Therefore, compared to using all the face detection windows 44a to 44i and the face detectors 45a to 45e for the entire image, it is possible to improve the efficiency of face detection processing (reduction in the amount of calculation). In addition, by setting an appropriate illumination condition for each partial region in advance, an improvement in detection accuracy and a reduction in false detection can be expected.

  In the partial area information of this embodiment, a common partial area is set for the face size condition and the illumination condition, but independent partial areas may be set. Further, the face detection window and face detector switching method switches the face detection window and face detector for each face detection position, but one set of face detection window and face detector is used for the entire partial area. After performing the face detection processing, the face detection position may be moved to the initial position of the partial area, and the face detection window may be switched.

<Fourth Embodiment>
In the fourth embodiment, as shown in FIG. 17, a face detection device will be described in a situation where there are walls on the front and left and right within the range of an image captured by the imaging unit 11 (the front wall needs to be a dead end). There may be a passage on the left and right like a T-shaped passage). Under this circumstance, the image captured by the imaging unit 11 tends to show a larger face at the lower part of the image, as in the first embodiment. Furthermore, as shown in FIG. 18, there is a tendency that a part of an image photographed by the imaging unit 11 (for example, the left and right walls and the front wall described above) cannot include a face. Focusing on such a tendency, the face detection process in the present embodiment changes the size of the face to be detected and whether or not to perform the face detection process for each partial region.

  Hereinafter, the configuration and processing different from those of the first embodiment will be mainly described.

An example of partial area information and face detection conditions in this embodiment is shown in FIG. As described above, the image photographed by the imaging unit 11 of the present embodiment tends to show a larger face at the lower part of the image as shown in FIG. 18, and the image of the face appears in a part of the image photographed by the imaging unit 11. There is a tendency to have areas that cannot be obtained. Therefore, in the example of FIG. 19, as the partial area information, the entire image is divided into five partial areas 51, which are “upper left part of image”, “upper right part of image”, “upper part of image”, “middle part of image”, and “lower part of image”.
Information to be divided into .about.55 is set (the definition of the partial area here is the same as in the first embodiment). As face detection conditions, face size conditions and presence / absence conditions for executing face detection processing for determining whether or not to perform face detection processing are set (here, the upper left portion of the image, the upper right portion of the image, and the partial regions 51 to 51 at the upper portion of the image). 53 is “not set” and “no face detection processing is executed”, “middle” and “face detection processing is executed” in the partial area 54 in the middle of the image, “large” and “face detection processing is executed in the partial area 55 in the lower part of the image A face size condition “Yes” and a condition for whether or not to perform face detection processing are set.). Note that the presence / absence condition of the face detection process execution may be an attribute such as “no face detection process execution” or “face detection process execution” as in the face size condition described in the first embodiment. A numerical value corresponding to whether or not to perform processing may be used.

<Face detection function>
Since the function of the face detection apparatus of this embodiment is the same as that of the first embodiment except for the face detection process in step S13, the description thereof is omitted, and here, the function of this embodiment is described along the flowchart of FIG. The face detection process (step S13) will be described in detail.

  First, the face detection unit 13 places the face detection position at the initial position in the image captured by the imaging unit 11 (step S51).

  Next, the face detection unit 13 identifies the partial area to which the face detection position belongs by comparing the coordinates of the face detection position with the partial area information (step S52). Then, the face detection unit 13 performs a face detection window and a face detection process to be applied to the face detection position based on the face size condition set in the specified partial region and the presence / absence condition of the face detection process. Select whether or not.

  The face detection unit 13 of the present embodiment has a plurality of types of face detection windows 56a to 56f having different sizes. However, the face detection unit 13 does not uniformly apply all the face detection windows to the entire image, but applies the face detection window and the face detection process to be applied according to the face size condition and the presence / absence condition of the face detection process. Whether or not to do this is changed for each partial area. In the example of FIG. 18, when the face detection position is in the partial areas 51 to 53, the face detection window is not selected, and the face detection process at the face detection position is not performed. When the face detection position is in the partial region 54, face detection windows 56a to 56c for the “medium” face are selected, and face detection processing at the face detection position is performed. When the face detection position is in the partial area 55, the face detection windows 56d to 56f for the “large” face are selected, and the face detection process at the face detection position is performed.

  Then, the face detection unit 13 determines whether or not to perform face detection processing in step S53. When it is determined that “face detection processing is performed” (step S53; YES), the process proceeds to step S54, and processing at the face detection position is continued, and when “face detection processing is not performed” is determined (step S53; NO). Then, the process for the face detection position ends, and the process proceeds to step S57.

  Next, the face detection unit 13 sets one of the face detection windows selected in step S52 as a face detection position (step S54), and executes face detection processing (step S55). Next, the face detection unit 13 determines whether there is a face detection window that is not used in the face detection process among the plurality of face detection windows selected in step S52 (step S56). If it is determined that there is another face detection window (step S56; YES), the face detection window is set at the face detection position (step S54), and face detection processing is executed (step S55). When face detection processing has been performed for all the face detection windows selected in step S52 (step S56; NO), the processing for the face detection position ends, and the process proceeds to step S57.

In step S57, the face detection unit 13 determines whether or not there is a next face detection position. If there is a next face detection position (step S57; YES), the face detection position is moved to the next face detection position (step S57). Step S58). Then, it is determined whether or not to perform face detection processing for the next face detection position, and when it is determined that “face detection processing is performed”, face detection processing is performed using a face detection window of an appropriate size for the partial area. Is performed (steps S52 to S56). Steps S52 to S56 are repeated until there is no next face detection position (until the end of the image is reached) (step S57).

  As described above, in this embodiment, the face detection process is omitted in the upper left part of the image, the upper right part of the image, and the partial areas 51 to 53 in the upper part of the image. Only the three face detection windows 56a to 56c are used in the partial area 54 in the middle of the image, and face detection processing using the other face detection windows 56d to 56f is omitted. Similarly, in the partial area 55, the face detection process by the face detection windows 56a to 56c is omitted. Therefore, compared to using all the face detection windows 56a to 56f for the entire image, it is possible to improve the efficiency of the face detection process (reduction in the amount of calculation). Further, by setting appropriate presence / absence conditions for executing face detection processing for each partial region, it is possible to exclude faces at positions that cannot be found from the beginning, so that reduction of false detection can be expected.

  The above four embodiments show an example of face detection processing according to different face detection conditions, but the present invention is a face according to at least one of the four face detection conditions in the above embodiments. Perform detection processing.

  In addition, it is preferable that the information memorize | stored in the face detection condition memory | storage part of the said embodiment can be set so that it may change according to time. For example, when indoors and outdoors are captured in an image as in the third embodiment, the indoor and outdoor lighting environments are different during the day and at night. In this case, “daytime: bright”, “nighttime: dark”, etc. may be set as the lighting conditions for the outdoor portion, and “daytime: dark”, “nighttime: bright”, etc., as the lighting conditions for the indoor portion. It should be set.

  In the first to fourth embodiments, the size of the face detected in advance is determined for each partial region of the image, and the direction of the face detected in advance is determined in advance. The present invention can be preferably applied when the face illumination environment is determined, and when the face detection area is determined in advance. For example, in the third embodiment, an example in which a difference between indoor and outdoor illumination environments is assumed is described, but the present invention can be preferably applied to a case where a part of an image is illuminated with illumination. In the fourth embodiment, “no face detection process” is set in “area where face cannot be captured”. However, even if there is a person, “there is a part of a large number of people” It can be preferably applied to the case where “a person does not count”.

FIG. 1 is a diagram illustrating an installation mode of an imaging unit according to the first embodiment. FIG. 2 is a diagram illustrating an image photographed by the imaging unit and face detection processing of the image in the first embodiment. FIG. 3 is a block diagram illustrating a functional configuration of the face detection apparatus. FIG. 4 is a diagram illustrating an example of partial region information and face detection conditions (face size conditions) in the first embodiment. FIG. 5 is a flowchart showing the flow of processing of the face detection apparatus. FIG. 6 is a flowchart showing the flow of face detection processing in the first embodiment. FIG. 7 is a diagram illustrating face detection processing using all face detection windows for the entire image. FIG. 8 is a diagram illustrating an installation mode of the imaging unit in the second embodiment. FIG. 9 is a diagram illustrating an image photographed by the imaging unit and face detection processing of the image in the second embodiment. FIG. 10 is a block diagram illustrating a face detection unit and a face detector in the second embodiment. FIG. 11 is a diagram illustrating an example of partial region information and face detection conditions (face size condition and face orientation condition) in the second embodiment. FIG. 12 is a flowchart showing the flow of face detection processing in the second and third embodiments. FIG. 13 is a diagram illustrating an installation state of the imaging unit in the third embodiment. FIG. 14 is a diagram illustrating an image photographed by the imaging unit and face detection processing of the image in the third embodiment. FIG. 15 is a block diagram showing a face detector and a face detector in the third embodiment. FIG. 16 is a diagram illustrating an example of partial region information and face detection conditions (face size conditions and illumination conditions) in the third embodiment. FIG. 17 is a diagram illustrating an installation state of the imaging unit according to the fourth embodiment. FIG. 18 is a diagram illustrating an image photographed by the imaging unit and face detection processing of the image in the fourth embodiment. FIG. 19 is a diagram illustrating an example of partial region information and face detection conditions (face size condition and presence / absence condition for execution of face detection processing) in the fourth embodiment. FIG. 20 is a flowchart showing the flow of face detection processing in the fourth embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Image pick-up part 12 Image memory | storage part 13 Face detection part 14 Display part 15 Result memory | storage part 16 Face detection condition memory | storage part 21-23, 31-33, 41-43, 51-55 Partial region 24a-24i, 34a-34i, 44a ˜44i, 56a to 56f Face detection window 25 Movement path 35,45 Face detector

Claims (9)

Imaging means;
Face detection condition storage means for storing partial area information for dividing an image captured by the imaging means into a plurality of partial areas and face detection conditions set in advance for each of the partial areas;
A face detection unit that omits a face detection process that does not meet the face detection condition set in the partial region to which the face detection position belongs when performing a face detection process for detecting a face from the image;
A face detection apparatus comprising: The face detection apparatus according to claim 1, wherein the face detection condition includes a face size condition that determines a size of a face to be detected. The face detection means has a plurality of face detection windows corresponding to different face sizes, and a partial area to which a face detection position belongs when performing face detection processing using the face detection window from the image The face detection device according to claim 2, wherein the face detection process is performed using a face detection window corresponding to the face size condition set to, and face detection processing using other face detection windows is omitted. The face detection apparatus according to claim 1, wherein the face detection condition includes a face orientation condition that determines a face orientation to be detected. The face detection means has a plurality of face detectors corresponding to different face orientations, and is set to a partial region to which a face detection position belongs when performing face detection processing using the face detector from the image The face detection apparatus according to claim 4, wherein face detection processing is performed using a face detector corresponding to the face orientation condition, and face detection processing by other face detectors is omitted. The face detection apparatus according to claim 1, wherein the face detection condition includes an illumination condition. The face detection means has a plurality of face detectors corresponding to different illumination environments, and is set to a partial region to which a face detection position belongs when performing face detection processing using the face detector from the image The face detection apparatus according to claim 6, wherein face detection processing is performed using a face detector corresponding to the illumination condition, and face detection processing by other face detectors is omitted. A computer that performs face detection processing for detecting a face from an image captured by an imaging unit,
Storing partial area information for dividing an image photographed by the imaging means into a plurality of partial areas and face detection conditions set in advance for each partial area;
A face detection method that omits a face detection process that does not satisfy a face detection condition set in a partial region to which a face detection position belongs when performing a face detection process for detecting a face from the image. To a computer that performs face detection processing for detecting a face from an image taken by an imaging means,
Storing partial area information for dividing an image captured by the imaging unit into a plurality of partial areas and face detection conditions set in advance for each partial area;
Omitting the face detection process that does not meet the face detection condition set in the partial region to which the face detection position belongs when executing the face detection process for detecting a face from the image;
Face detection program for running.

Images