JP4333457B2 - Image processing apparatus and image processing method - Google Patents

Description

  The present invention relates to an image processing apparatus and an image processing method, and more particularly to an image processing apparatus and an image processing method for detecting an object such as a face in an image captured by an imaging unit.

When performing face orientation measurement or personal authentication, it is required to detect a face from an image including a face photographed by a camera. As a device for detecting such a face, there is a face image feature point extracting device disclosed in Japanese Patent Laid-Open No. 5-197793. This face image feature point extracting device detects a face by performing edge extraction of a face in an image and calculating a degree of coincidence with shape data after edge extraction, so-called pattern matching.
JP-A-5-197793

  However, the size of the face photographed by the camera is not constant. For this reason, in the face image feature point extraction device disclosed in Patent Document 1, in calculating the degree of coincidence with the shape data, it must have shape data of many sizes, the database becomes enormous, There has been a problem that the amount of data stored in the image processing apparatus increases. Furthermore, since the degree of coincidence is calculated for many sizes of shape data, there is a problem that it takes a long time to perform detection.

  Accordingly, an object of the present invention is to provide an image processing apparatus and an image processing method capable of reducing the amount of data to be stored and detecting a detection target at high speed.

An image processing apparatus according to the present invention that has solved the above problems is an image processing apparatus that detects an object in an image photographed by a plurality of imaging means having a predetermined positional relationship, and is an object photographed by a plurality of imaging means. A three-dimensional restoration means for restoring the three-dimensional shape, and a predetermined distance pixel extraction means for extracting a pixel that reflects a position within a predetermined range of the distance from the predetermined position in the object three-dimensionally restored by the three-dimensional restoration means; The size of the detection target object corresponding to the distance from the predetermined position is stored in advance, and the size of the detection target object and the pixels extracted by the predetermined distance pixel extraction unit occupy in the image captured by the imaging unit Comparison detection means for detecting a detection target object by comparing the size.

  An image processing method according to the present invention that solves the above-described problems is an image processing method for detecting an object in an image captured by a plurality of imaging means having a predetermined positional relationship, and is captured by a plurality of imaging means. 3D restoration process that restores a 3D shape to a captured object, and a predetermined distance pixel extraction that extracts a pixel that reflects a position within a predetermined range of a distance from a predetermined position in the object that has been three-dimensionally restored by the 3D restoration process By comparing the size of the detection target object for each distance from the predetermined position stored in advance with the size that the pixel extracted in the predetermined distance pixel extraction step occupies in the image captured by the imaging unit And a comparison detection step of detecting a detection target object.

An image processing method according to the present invention that solves the above-described problems is an image processing method for detecting an object in an image captured by a plurality of imaging means having a predetermined positional relationship, and is captured by a plurality of imaging means. 3D restoration process that restores a 3D shape to a captured object, and a predetermined distance pixel extraction that extracts a pixel that reflects a position within a predetermined range of a distance from a predetermined position in the object that has been three-dimensionally restored by the 3D restoration process The process compares the size of the detection target object according to the distance from the predetermined position stored in advance with the size occupied by the pixel extracted in the predetermined distance pixel extraction step in the image captured by the imaging unit. And a comparison detection step of detecting a detection target object.

  Here, in the present invention, since the sizes of the detection target objects are compared, the image processing apparatus stores data corresponding to the size of the detection target object, and does not need the data of the face itself. Therefore, the amount of data stored in the image processing apparatus can be reduced, and detection at a high speed can be performed.

  According to the image processing apparatus and the image processing method of the present invention, it is possible to reduce the amount of data to be stored and detect the detection target at high speed.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram of an image processing apparatus according to an embodiment of the present invention.

  As shown in FIG. 1, the image processing apparatus 1 according to this embodiment includes a three-dimensional restoration unit 11, a pixel extraction unit 12 that is a predetermined distance pixel extraction unit, and a face circle comparison unit 13 that is a comparison detection unit. Yes. The image processing apparatus 1 is connected to a first camera 2 and a second camera 3 which are a plurality of imaging means. The first camera 2 and the second camera 3 are in a predetermined positional relationship with each other. The first camera 2 and the second camera 3 output an image obtained by photographing a human face or the like to the three-dimensional restoration unit 11 in the image processing apparatus 1. In this embodiment, a human face is used as a detection target object.

  The three-dimensional restoration unit 11 three-dimensionally restores images taken by the first camera 2 and the second camera 3 and creates a three-dimensional restored image. The three-dimensional restoration unit 11 outputs the restored three-dimensional restoration image to the pixel extraction unit 12.

  The pixel extraction unit 12 extracts pixels that show an object within a predetermined distance from a predetermined position, for example, a position where the first camera 2 is installed. In the pixel extraction unit 12, a plurality of predetermined distance ranges from the first camera 2 are set. For each of the plurality of predetermined distances, pixels within the predetermined distance range are extracted.

  The extraction result in the pixel extraction unit 12 is output to the face circle comparison unit 13. In the face circle comparison unit 13, the shape and size of the face that is the detection target object are stored in advance as templates for each predetermined position, for example, for each distance from the first camera 2. The shape of the face can be modeled as a shape that approximates an ellipse, and the pixel extraction unit 12 stores an ellipse having a size corresponding to the distance from the first camera 2 as a template. The face circle comparison unit 13 compares the circularity from the extraction result output from the pixel extraction unit 12 and a template stored in advance. By comparing the circularity, a face that is a detection target object is detected.

  An image processing method using the image processing apparatus according to this embodiment having the above configuration will be described. FIG. 2 is a flowchart showing the procedure of the image processing method according to the present embodiment.

  As shown in FIG. 2, in the image processing method according to the present embodiment, first, an image is taken by the first camera 2 and the second camera 3 and output to the three-dimensional restoration unit 11 in the image processing apparatus 1. The three-dimensional reconstruction unit 11 performs three-dimensional reconstruction of the captured images based on the images captured by the cameras 2 and 3 and the camera parameters of the two cameras 2 and 3 (S1).

  The restoration result is calculated as a distance image when each pixel of the image viewed from the first camera 2 has the optical axis direction of the first camera 2 as the depth. The final processing result here is an image showing how far each pixel photographed by the first camera 2 is located in the depth direction of the first camera 2, and the distance is determined for each pixel. Is required.

  After three-dimensional restoration in this way, the pixel extraction unit 12 performs pixel extraction. When extracting pixels, the counter is set to 0 (S2). If the counter is set to 0, next, a pixel that reflects a position existing at a distance within a predetermined range from the first camera 2 is extracted (S3). When the counter is 0, a pixel indicating an object located within a range of 0 to 50 mm from the first camera 2 is extracted.

  As described above, when the pixels indicating the object located within the range of 0 to 50 mm from the first camera 2 are extracted, the face circle comparison unit 13 occupies the size occupied by the extracted pixels and the detection target stored in advance. Compare the size of the face that is the object. Thus, an area in which the size occupied by the extracted pixel occupies a predetermined threshold, for example, 70% or more with respect to the template stored in advance is searched (S4).

  When the face is photographed by the first camera 2, as schematically shown in FIG. 3, the nose portion of the face FA exists in the range R <b> 1 where the distance from the first camera 2 is 0 to 50 mm. At this time, as shown in FIG. 4A, the existence pixel E1 in the screen P1 occupies only a small area.

  The size of the face shown in the image varies depending on the distance from the first camera 2. The shorter the distance from the first camera 2, the larger the size of the face shown in the image. The longer the distance, the smaller the size of the face shown in the image. The face circle comparison unit 13 stores a face size template corresponding to the distance from the first camera 2.

  As shown in FIGS. 5A to 5G, the size of the template stored in the face circle comparison unit 13 varies depending on the range of distance from the first camera 2. When the distance from the first camera 2 is 0 to 50 mm, a template O1 having a size shown in FIG. 5A is stored. The face shape is generally close to an ellipse. The template O1 has an oval shape that generalizes the outline of the face, and is set assuming a plane that is orthogonal to the optical axis of the first camera 2 and crosses the front of the face FA as shown in FIG. Has been.

  When the distance from the first camera 2 is 10 to 60 mm, a template O2 having a size shown in FIG. 5B is stored. The template O1 shown in FIG. 5B is set slightly smaller than the template O1 shown in FIG.

  Thereafter, every time the range of the distance from the first camera 2 moves backward by 10 mm, the template O3 shown in FIG. 5C, the template O4 shown in FIG. 5D, the template O5 shown in FIG. The template O6 shown in (f) and the template O7 shown in FIG. These templates O2 to O6 are also set on the assumption of a plane orthogonal to the optical axis of the first camera 2 and crossing the front part of the face FA, like the template O1.

  Returning to step S4, if an area occupying an area in which the size of the extracted pixel occupies a predetermined threshold or more with respect to a template stored in advance is detected, the detected area becomes a face candidate. In addition, if a region that occupies an area in which the size of the extracted pixel occupies a predetermined threshold or more with respect to a template stored in advance is not searched, no face candidate is detected. In this case, a region that is a face candidate is not detected.

  Thus, when the detection at i = 0 is completed, the counter is incremented by 1 (S5), and i = n (= 8) is determined to determine whether or not to end (S6). As a result, if it is determined that the process is not ended, the process returns to step S3, and pixels indicating an object located in the range R2 whose distance from the first camera 2 is 10 to 60 mm are extracted.

  Thereafter, the face circle comparison unit 13 compares the size occupied by the extracted pixels with the size of the face that is the detection target object stored in advance. Thus, an area that occupies an area in which the size of the extracted pixel occupies a predetermined threshold or more with respect to the template stored in advance is searched (S4).

  Thereafter, in the same manner, the processing from step S2 to step S4 is repeated until i = n is reached, the range R3 in which the distance from the first camera 2 is 20 to 70 mm, and the distance from the first camera 2 is 30 to 80 mm. The range R4 and the range R5 having a distance from the first camera 2 of 40 to 90 mm are searched for a region that occupies a region equal to or greater than a predetermined threshold with respect to the template. Further, a template is used for a range R6 in which the distance from the first camera 2 is 50 to 100 mm, a range R7 in which the distance from the first camera 2 is 60 to 110 mm, and a range R8 in which the distance from the first camera 2 is 70 to 120 mm. Are searched for an area occupying an area equal to or greater than a predetermined threshold.

  In this manner, pixels existing within a range of 50 mm intervals are extracted at intervals of 10 mm. A face candidate is determined when the ratio of the area in the template is equal to or greater than the threshold value by comparison with the template at each step. FIGS. 4B to 4G show the relationship between the existing ranges E2 to E7 displayed in the images P2 to P7 and the templates O2 to O7. In the example shown in FIG. 4, the cases shown in (d) and (e) exceed the threshold value and can be detected as a face.

  As described above, in the image processing method according to the present embodiment, the face can be reliably detected by comparing the face size with the size occupied by the pixels. Further, the image processing apparatus 1 stores data corresponding to the size of the face and does not need data for the face itself. For this reason, the data amount in the image processing apparatus 1 can be reduced, and detection at high speed can be achieved.

  The preferred embodiment of the present invention has been described above, but the present invention is not limited to the above embodiment. For example, in the above embodiment, a human face is illustrated as a moving object, but a moving object such as an automobile can be used as the moving object. In the above embodiment, the face template is set to be oval. However, in consideration of the unevenness of the face, for example, a shape excluding the nose portion can be set as the template. As described above, the shape of the template can be appropriately adjusted according to the detection target, whereby detection with higher accuracy can be performed.

1 is a block configuration diagram of an image processing apparatus according to the present invention. It is a flowchart which shows the procedure of the image processing method which concerns on this invention. It is a schematic diagram which shows the distance relationship between a 1st camera and a face. (A)-(g) is a figure which shows the relationship between the pixel in which the object reflected on the image exists, and a template. (A)-(g) is a figure which shows the magnitude | size of a template. It is a figure explaining the positional relationship at the time of comparing an image and a template.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Image processing apparatus, 2 ... 1st camera, 3 ... 2nd camera, 11 ... Three-dimensional decompression | restoration part, 12 ... Pixel extraction part, 13 ... Face circular comparison part, FA ... Face, O1-O7 ... Template, P1- P7 ... screen, E1-E7 ... existing range.

Claims (2)

An image processing apparatus for detecting an object in an image captured by a plurality of imaging means having a predetermined positional relationship,
Three-dimensional restoration means for restoring an object photographed by the plurality of imaging means to a three-dimensional shape;
Predetermined distance pixel extracting means for extracting pixels that reflect a position within a predetermined range of a distance from a predetermined position in the object three-dimensionally restored by the three-dimensional restoration means;
The size of the detection target object corresponding to the distance from the predetermined position is stored in advance, and the size of the detection target object and the pixel extracted by the predetermined distance pixel extraction unit are captured by the imaging unit Comparison detecting means for detecting a detection target object by comparing the size occupied in the medium,
An image processing apparatus comprising: An image processing method for detecting an object in an image photographed by a plurality of imaging means having a predetermined positional relationship,
A three-dimensional restoration step of restoring an object photographed by the plurality of imaging means into a three-dimensional shape;
A predetermined distance pixel extracting step of extracting pixels that reflect a position within a predetermined range of the distance from the predetermined position in the object three-dimensionally restored by the three-dimensional restoration step;
The size of the detection target object corresponding to the distance from the predetermined position stored in advance is compared with the size occupied by the pixel extracted in the predetermined distance pixel extraction step in the image captured by the imaging unit. A comparison detection step of detecting a detection target object,
An image processing method comprising:

Images