JP5424779B2 - Imaging apparatus, imaging method, and program - Google Patents

Description

  The present invention relates to an imaging apparatus, an imaging method, and a program, and more particularly to a technique suitable for use in photographing a person with high image quality.

  In recent years, techniques for detecting a person's face from a photographed image have been rapidly developed, and proposals have been made to improve the image quality of a face area through face detection. Japanese Patent Application Laid-Open No. 2004-133867 discloses a technique that includes a face detection unit that detects a person's face from a photographed image and lowers the quantization parameter of the detected face part to improve the image quality of the person's face part. Has been. Japanese Patent Application Laid-Open No. 2004-228561 discloses a technique for determining a face portion in an image by skin color detection and assigning a large amount of code to the skin color area to improve image quality.

JP-A-6-233292 JP-A-2005-218017

  When a person is photographed by an imaging device such as a video camera, the image quality of the body part as well as the face is regarded as important. However, in the technique described in Patent Document 1 described above, the face part can be improved in image quality by face detection, but the torso part is handled as a part other than the face, so that the torso part of a person is improved in image quality. Can not do it. In the technique described in Patent Document 2, the body part is not limited to the skin color, and therefore, the body cannot be specified only by detecting the skin color. .

  The present invention has been made in view of the above-described problems, and an object of the present invention is to improve the image quality of a torso portion together with a human face portion.

An image pickup apparatus according to the present invention detects an inclination of an image pickup apparatus body, an image pickup unit that picks up an image of a subject, generates an image signal, a face detection unit that detects a human face area from the image signal generated by the image pickup unit. An inclination detection means for detecting the person's torso area based on the detection result of the face detection means and the detection result of the inclination detection means, and the face area detected by the face detection means. have a coding means for coding the image signal and the detected body region by adjusting the quantization step size in the quantization processing so as to be high quality by the body detection means, the body detecting means, Of the inclinations detected by the inclination detection means, when the front-rear inclination with respect to the horizontal plane is within a predetermined range, the face area detected by the face detection means is used as a base point. The direction of gravity region determines a body region, when the front and rear inclination with respect to the horizontal plane exceeds a predetermined range, characterized in that as a base point of the face region, to determine the tilt direction and parallel to the region of the face and trunk areas And

  According to the present invention, the torso area can be detected appropriately, so that the image quality of the torso part as well as the human face part can be improved.

It is a block diagram which shows the structural example of the imaging device which concerns on embodiment of this invention. It is a figure which shows an example of the picked-up image containing the person who is a to-be-photographed object. It is a figure which shows an example of the state which inclined the imaging device main body. It is a figure which shows a mode that the person who is a subject is image | photographed. It is a figure which shows an example of the picked-up image when the subject person does not tilt his / her face. It is a figure which shows an example of the picked-up image when the subject person inclines the face. It is a figure which shows a mode that the sleeping face of the person who is a subject is image | photographed. It is a figure which shows the picked-up image image | photographed in the state shown in FIG. It is a flowchart which shows an example of the process sequence which determines a trunk | drum area | region. It is a figure which shows an example of the method of determining a torso area | region from the size of a face.

(First embodiment)
DESCRIPTION OF EXEMPLARY EMBODIMENTS Hereinafter, preferred embodiments of the invention will be described in detail with reference to the drawings.
FIG. 1 is a block diagram illustrating a configuration example of an imaging apparatus 100 according to the present embodiment.
In FIG. 1, an imaging signal processing unit 102 performs A / D conversion on an imaging signal obtained by an imaging unit 101 including a lens and an imaging element, and generates an image signal subjected to processing such as color matrix conversion and gamma correction. . The image signal output from the imaging signal processing unit 102 is input to the encoding unit 103 and the face detection unit 104.

  The encoding unit 103 encodes the image signal from the imaging signal processing unit 102 by a compression encoding method such as the MPEG-2 method. The encoding unit 103 divides an image into 16 × 16 pixel blocks called macroblocks, performs motion compensation prediction and orthogonal transform, further quantizes the transform coefficient, performs entropy coding, and compresses the image. Encoding is performed.

  Here, the quantization is to express the amplitude of the orthogonal transform coefficient by a certain representative value. By performing the quantization, the representative value is reduced, and the generated code amount can be reduced. However, if the quantization is performed roughly, the reproducibility of the original conversion coefficient is lowered, and the image quality of the reproduced image is deteriorated. Further, if quantization is performed finely, image quality deterioration is reduced, but the code amount is increased. The generated code amount is adjusted by controlling the quantization parameter for setting the degree of quantization.

  In the present embodiment, the encoding unit 103 controls the quantization parameter similar to the Test Model (TM) 5 of MPEG-2. That is, an activity value indicating the complexity of an image is first calculated for each macroblock. Then, by calculating the activity value, the value of the quantization parameter (quantization step size) is reduced in a flat image that tends to be visually noticeable, so that no deterioration occurs. On the other hand, in the case of an image with a complicated pattern that is relatively inconspicuous, deterioration is controlled by increasing the value of the quantization parameter (quantization step size) and reducing the code amount. Also, the encoding unit 103 controls the quantization parameter so as to improve the image quality of the face area and the body area described below. The recording unit 107 records the image data compressed and encoded by the procedure as described above on the recording medium 108.

  The face detection unit 104 detects a human face area in the image signal and outputs it as face area information. For example, when a person is photographed as shown in FIG. 2, information on the position, size, and inclination of the quadrangle A surrounding the face portion is output. The output face area information is input to the body detection unit 105 and the encoding unit 103. The torso detection unit 105 determines the torso region according to the face area information output from the face detection unit 104 and the tilt information output from the tilt detection unit 106. The process flow for determining the body region will be described later. The tilt detection unit 106 is for detecting the tilt of the imaging apparatus main body, and is generally called a gravity sensor. In addition, the inclination detection unit 106 is installed inside the imaging apparatus 100 and detects the left and right inclination angles X of the imaging apparatus 100 main body with respect to the vertical plane and the front and rear inclination angles Y with respect to the horizontal plane. The relationship between the tilt of the imaging apparatus 100 main body and the detected tilt information will be described with reference to FIG.

  FIGS. 3A to 3D are diagrams illustrating an example of a state in which the main body of the imaging apparatus 100 is tilted. FIGS. 3A and 3B illustrate the main body of the imaging apparatus 100 as a lens. It is the front view seen from the side. FIG. 3A shows a state in which the main body of the imaging apparatus 100 is held vertically, and the left and right tilt angles with respect to the vertical plane are 0 degrees. FIG. 3B shows a state in which the main body of the imaging apparatus 100 is tilted to the right from the front, and the left and right tilt angles with respect to the vertical plane are X. When tilted to the left, it is expressed as a negative angle.

  3C and 3D are views of the imaging device 100 main body as viewed from the side. FIG. 3C shows a state in which the imaging angle of the main body of the imaging apparatus 100 is held horizontally, and the tilt angle before and after the horizontal direction is 0 degree. FIG. 3D shows a state in which the lens side of the main body of the imaging device 100 is tilted downward, and the tilt angle before and after the horizontal plane is Y. When the lens side is tilted upward, it is expressed as a negative angle.

Next, a processing procedure for determining the body region from the face detection result will be described. FIG. 9 is a flowchart illustrating an example of a processing procedure for determining a body region by the body detection unit 105.
In FIG. 9, when face area information is input from the face detection unit 104 and tilt information is input from the tilt detection unit 106, processing starts in step S <b> 901. In step S902, it is determined from the face area information input from the face detection unit 104 whether a face area exists. As a result of this determination, if the face area does not exist, the information is sent to the encoding unit 103 in step S905, assuming that the body area does not exist. In step S908, the body detection process is terminated. On the other hand, as a result of the determination in step S902, if a face area exists, the size of the body area is determined from the size of the face area indicated by the face area information in step S903.

FIG. 10 is a diagram illustrating an example of a method for determining the size of the body region from the size of the face region.
In FIG. 10, a rectangle A is a region (face region) set to surround the face in the face region information, and a rectangle B is a body region to be obtained. From the width Ax and the height Ay of the quadrangle A that surrounds the face, a predetermined height coefficient and a width coefficient are multiplied to calculate the width Bx and the height By of the quadrangle B in the body region.

  Next, the position of the body region is determined. In step S904, it is determined whether or not the absolute value of the tilt Y before and after the main body of the imaging apparatus 100 in the tilt information input from the tilt detection unit 106 is equal to or less than a predetermined value. That is, when the front and rear tilt angles in the tilt information is Y and the predetermined angle is Ya, it is determined whether −Ya ≦ Y ≦ Ya is satisfied. When the formula is established in this way, the absolute value of the front / rear tilt of the imaging apparatus 100 main body is within the range of the angle Ya from the horizontal plane. If the absolute value of the slope Y is equal to or smaller than the predetermined value as a result of this determination, the process proceeds to the next step S906.

  Next, in step S906, the position of the body region with respect to the face region is determined according to the direction of gravity. As a procedure, first, the gravitational direction in the photographed image is obtained from the tilt information input from the tilt detection unit 106. The gravitational direction can be calculated from the front and rear inclinations and the left and right inclinations detected by the inclination detection unit 106. Then, the position is determined on the basis of the face region as the base region in the gravity direction. And it sends to the encoding part 103 as fuselage area | region information. By adopting such a procedure, it is possible to appropriately determine the position of the torso region even when the subject person tilts his face or when the photographer shoots while tilting the imaging device. Hereinafter, a specific example will be described with reference to FIGS.

  FIG. 4 is a diagram illustrating a situation where a person who is a subject is being photographed. In the case where the absolute value of the front-rear tilt of the imaging apparatus 100 body is within an angle Ya from the horizontal plane, that is, the lens of the imaging apparatus 100 is substantially horizontal. The case where it is facing is shown.

  FIG. 5A is a diagram illustrating an example of a photographed image when the subject person does not tilt his / her face and when the imaging apparatus 100 main body has no left / right tilt as illustrated in FIG. 3A. It is. FIG. 5B shows an example of a photographed image when the subject person does not tilt his / her face and the imaging apparatus 100 main body has a left / right tilt as shown in FIG. 3B. FIG. On the other hand, FIG. 6A shows an example of a photographed image when the subject person tilts his / her face and the imaging apparatus 100 main body has no left-right tilt as shown in FIG. FIG. FIG. 6B shows an example of a photographed image when the subject person tilts his / her face and the imaging apparatus 100 main body has a left / right tilt as shown in FIG. 3B. FIG.

  501 in FIG. 5A, 502 in FIG. 5B, 601 in FIG. 6A, and 602 in FIG. 6B, respectively, arrows indicating the gravitational direction obtained from the inclination information of the inclination detecting unit 106. It is. As shown in FIGS. 5 and 6, by determining the body region B in the gravitational direction with respect to the face region A, the body part can be appropriately determined as the body region.

  On the other hand, if the absolute value of the tilt Y exceeds the predetermined value as a result of the determination in step S904, the position of the body region with respect to the face region is determined according to the tilt direction of the face in step S907. And it sends to the encoding part 103 as fuselage area | region information. Thus, it is determined that the body region is in the tilt direction of the face with the face region as a base point. Hereinafter, specific examples will be described with reference to FIGS.

  FIG. 7 is a diagram illustrating a state in which a sleeping face of a person who is a subject is photographed. In the example illustrated in FIG. 7, the absolute value of the front / rear tilt of the image capturing apparatus 100 main body is an angle Ya or more from the horizontal plane, and the lens of the image capturing apparatus 100 faces obliquely below Ya degrees or more. FIG. 8 is a diagram illustrating a captured image captured in the state illustrated in FIG. As shown in FIG. 8, by determining a region parallel to the face tilt direction as the body region B with respect to the face region A, the body part can be appropriately determined as the body region.

  As described above, when the position of the body region is determined in step S906 or S907, the process proceeds to step S908, and the body region detection process is terminated. As described above, the body detection unit 105 determines the body region by the processing shown in FIG. 9 and outputs the body region information to the encoding unit 103.

  The encoding unit 103 receives the face area information output from the face detection unit 104 and the body area information output from the body detection unit 105 as described above. When the encoding unit 103 performs quantization processing of the image signal, based on the input information, the macroblock including the face area and the body area is a flat image of the above-described activity calculation result (a macroblock that is easily deteriorated). The quantization parameter is controlled so as to be handled in the same manner as in FIG. That is, the encoding unit 103 quantizes the face region and the body region in the image with a relatively small quantization step size compared to other regions. As described above, since the quantization parameter is adjusted so that the quantization is finely performed on the macroblock including the face area and the body area, the face area and the body area can be improved in image quality.

  As described above, according to the present embodiment, the body region can be appropriately determined using the face region information in various shooting states. Then, by controlling the quantization parameters for the face region and the body region, the face region and the body region can be improved in image quality.

(Other embodiments according to the present invention)
The present invention can also be realized by executing the following processing. That is, software (program) that realizes the functions of the above-described embodiments is supplied to a system or apparatus via a network or various storage media, and a computer (or CPU, MPU, or the like) of the system or apparatus reads the program. It is a process to be executed.

101 imaging unit, 103 encoding unit, 104 face detection unit, 105 torso detection unit, 106 tilt detection unit

Claims (8)

Imaging means for imaging a subject and generating an image signal;
Face detection means for detecting a human face area from the image signal generated by the imaging means;
Tilt detecting means for detecting the tilt of the imaging apparatus main body;
A torso detecting means for detecting a torso area of the person based on a detection result by the face detecting means and a detection result by the inclination detecting means;
Coding means for encoding the image signal by adjusting a quantization step size in a quantization process so that the face area detected by the face detection means and the body area detected by the body detection means have high image quality. It has a door,
The body detecting means is an area in the gravity direction with the face area detected by the face detecting means as a base point when the inclination detected by the inclination detecting means is within a predetermined range with respect to the horizontal plane. Is determined as the body region, and when the front-rear inclination with respect to the horizontal plane exceeds a predetermined range, the region parallel to the tilt direction of the face is determined as the body region with the face region as a base point. .   The encoding means reduces the quantization step size relative to the face area detected by the face detection means and the trunk area detected by the trunk detection means relative to other areas. The imaging apparatus according to claim 1, wherein the imaging apparatus is characterized. The body detection means, the imaging apparatus according to claim 1 or 2, characterized in that to determine the size of the body region according to the size of the face region detected by the face detecting means. An imaging step of imaging an object and generating an image signal;
A face detection step of detecting a human face region from the image signal generated in the imaging step;
A tilt detection step for detecting the tilt of the imaging device body;
Based on the detection result in the face detection step and the detection result in the tilt detection step, a torso detection step for detecting the torso region of the person,
An encoding step of encoding the image signal by adjusting a quantization step size in a quantization process so that the face area detected in the face detection step and the body region detected in the body detection step have high image quality. And
In the torso detecting step, of the inclinations detected in the inclination detecting step, when the front-rear inclination with respect to the horizontal plane is within a predetermined range, the face region detected in the face detecting step is used as a base point in the gravitational direction. The region is determined as a body region, and when the front-rear inclination with respect to the horizontal plane exceeds a predetermined range, the region parallel to the tilt direction of the face is determined as the body region with the face region as a base point. Method. An imaging step of imaging an object and generating an image signal;
A face detection step of detecting a human face region from the image signal generated in the imaging step;
A tilt detection step for detecting the tilt of the imaging device body;
Based on the detection result in the face detection step and the detection result in the tilt detection step, a torso detection step for detecting the torso region of the person,
An encoding step of encoding the image signal by adjusting a quantization step size in a quantization process so that the face area detected in the face detection step and the body region detected in the body detection step have high image quality. And let the computer run
In the torso detecting step, of the inclinations detected in the inclination detecting step, when the front-rear inclination with respect to the horizontal plane is within a predetermined range, the face region detected in the face detecting step is used as a base point in the gravitational direction. determine the area and the body area, if the front and rear inclination with respect to the horizontal plane exceeds a predetermined range, as a base point to the face area, characterized that you determine the area parallel with the inclination direction of the face and trunk areas program. Imaging means for imaging a subject and generating an image signal;
Face detection means for detecting a human face area from the image signal generated by the imaging means;
A gravity sensor for detecting the direction of gravity relative to the imaging device body;
A torso detecting means for detecting a torso region of the person based on a detection result by the face detecting means and a detection result by the gravity sensor;
Coding means for encoding the image signal by adjusting a quantization step size in a quantization process so that the face area detected by the face detection means and the body area detected by the body detection means have high image quality. It has a door,
The torso detecting means determines the inclination of the imaging device main body based on the detection result by the gravity sensor, and when the front-rear inclination with respect to a horizontal plane is within a predetermined range, the face area detected by the face detecting means The area in the gravitational direction is determined as the body area from the base point, and when the front-rear tilt with respect to the horizontal plane exceeds a predetermined range, the area parallel to the tilt direction of the face is determined as the body area from the face area. An imaging apparatus characterized by that. The encoding means reduces the quantization step size relative to the face area detected by the face detection means and the trunk area detected by the trunk detection means relative to other areas. The imaging apparatus according to claim 6 . The imaging apparatus according to claim 6 or 7 , wherein the body detection unit determines the size of the body region according to the size of the face region detected by the face detection unit.

Images