JPH09271014A - Moving object recognizing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To highly precisely and quickly detect a moving object and to measure its shape and distance by using the known correspondence relation of a stereoscopic picture. SOLUTION: A picture input part 1 inputs a stereoscopic picture consisting of right and left pictures, a correspondence detection part 2 finds out correspondence relation between the right and left pictures in the stereoscopic picture inputted from the input part 1 and a correspondence storing part 3 stores the correspondence relation found out by the detection part 2. An area detection part 4 detects an area including a moving object based upon the correspondence relation stored in the storing part 3 and a stereoscopic picture inputted from the input part 1 on occasion and an object recognizing part 5 finds out the shape and distance of a moving object in the area detected by the detection part 4.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving object recognizing device, and more particularly to an moving object recognizing device for determining the shape and distance of an moving object in an image input from a television camera or the like.

[0002]

2. Description of the Related Art As a conventional device for detecting an object in an image, Japanese Patent Laid-Open No. 62-213495, "Abnormality Monitoring Device", Japanese Patent Laid-Open No. 6-217311, "Mobile Object Monitoring Device", Japanese Patent Laid-Open No. 7- Techniques such as those described in Japanese Patent Publication No. 79429 “Image Monitoring Device” are known. In these techniques, an image at a certain point of time is stored, and the difference between the images sequentially obtained is calculated. Since an image changes when a new object comes in, a new object is detected by processing this difference.

As a method of measuring the shape and distance of an object, a technique called a stereo method as described in Japanese Patent Laid-Open No. 3-81878 "Image recognition device for distance measurement using stereo image". It has been known. In this stereo method, first, two right and left images called stereo images are input, and the corresponding points of the left and right images are obtained by calculating the feature amount in the images. The corresponding points indicate where in the right image the target object at a certain position in the left image appears. For details on how to find such correspondence points,
For example, it is described in JP-A-62-107386, "Image matching method". Then, if the corresponding points of the two left and right images are obtained, the distance to the object surface can be calculated by the principle of triangulation. Therefore, it is possible to know the distance to the object and the shape of the object surface.

[0004]

In the above-described conventional apparatus for detecting an object in an image, the processing is performed based on the difference between the stored image and the sequentially obtained images, so that the illumination condition suddenly changes. However, even if a shadow such as a cloud falls on an object in the image, it is detected as a new object.

Further, the conventional method of measuring the shape and distance of an object by using the stereo method has a problem that it takes a lot of time to obtain a correspondence between images.

An object of the present invention is to provide a moving object recognizing device capable of detecting an moving object and measuring its shape and distance with high accuracy and high speed by utilizing the known correspondence relationship of stereo images. It is in.

[0007]

The moving object recognizing device of the first invention is an image input section for inputting a stereo image consisting of a left image and a right image, and a left image of the stereo image input by this image input section. And a right image, a correspondence detection unit for obtaining a correspondence relation, a correspondence storage unit for storing the correspondence relation obtained by the correspondence detection unit, a correspondence relation stored in the correspondence storage unit, and an input from the image input unit at any time A region detection unit that detects a region in which a moving object is present based on the stereo image, and an object recognition unit that determines the shape and distance of the moving object in the region detected by the region detection unit. To do.

Further, the moving object recognizing device of the second invention comprises a main image input section for inputting a main image, and an auxiliary image input section for inputting a plurality of auxiliary images from viewpoints around the viewpoint of the main image. A correspondence detection unit that obtains a correspondence relation between the main image and the plurality of auxiliary images, a correspondence storage unit that stores the correspondence relation obtained by the correspondence detection unit, and a correspondence relation stored in the correspondence storage unit, A region detection unit that detects a region in which a moving object is present based on a main image that is input from the main image input unit at any time and a plurality of auxiliary images that are input from the auxiliary image input unit at all times, and that is detected by this region detection unit. And an object recognition unit that obtains the shape and distance of the moving object in the defined region.

Furthermore, the moving object recognizing device of the third invention is
In the moving object recognition apparatus according to the first and second aspects of the invention, a change in the correspondence relationship stored in the correspondence storage unit with respect to the area outside the area detected by the area detection unit is detected, and the storage content of the correspondence storage unit is updated. It is characterized by including a corresponding update unit.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, the present invention will be described in detail with reference to the drawings.

Before going into the description of a specific embodiment, the principle of the moving object recognizing device of the present invention will be described with reference to FIGS. 9 to 15.

FIG. 9 is a diagram showing the relationship between a stereo image and a subject from above. Projection plane 1 from viewpoint 10 of the left image
A subject 14 is projected onto the image 2 to obtain a left image. Similarly, the subject 14 is projected from the viewpoint 11 of the right image on the projection surface 13 to obtain the right image.

As shown in FIG. 10, the position 15 on the subject 14 is projected to the position 16 on the projection surface 12 in the left image and to the position 17 on the projection surface 13 in the right image. Thus, if it is known that the position 16 on the left image and the position 17 on the right image are the same position 15 on the subject 14, the distance to the position 15 on the subject 14 can be measured by the principle of triangulation ( Method of measuring the distance to the subject surface by the stereo method). The correspondence relationship between the left image and the right image with respect to the subject 14 is obtained in advance. For example, there is a correspondence relationship in which the position 16 of the left image and the position 17 of the right image correspond to each other.

Thereafter, as shown in FIG. 11, it is assumed that a new object 18 has moved. According to the correspondence relationship obtained in advance, the position 16 of the left image was the same as the position 17 of the right image. However, since the image obtained from the position 19 on the new object 18 is different from the image obtained from the position 15 of the subject 14, the existence of the new object can be recognized.

Therefore, the newly obtained left image and right image are compared with each other on the basis of the correspondence previously obtained for the subject 14. Even if the image of the surface of the subject 14 changes due to a change in the illumination condition or the like, such a change is reflected in both the left image and the right image. Therefore, the left image and the right image are compared based on the correspondence relationship, and when the obtained images are different, the area in which a new object may exist can be detected. Since the process of detecting this area is a process of simply comparing the left image and the right image on the basis of the correspondence relationship, the process can be executed at extremely high speed.

However, as shown in FIG. 12, a region where the left image and the right image are different from each other based on the correspondence is, for example, from the position 20 to the position 21 on the projection plane 12 when viewed in the left image,
It is wider than the area where the object 18 actually exists.

However, as shown in FIG. 13, when a large object 22 exists, the position 23 of the left image and the position 24 of the right image may be compared to be the same image. To prevent the mistakes that occur in these cases,
The area including the object 18 as shown in FIG.
It is possible to detect the area where the new object 18 actually exists by detecting the area where 8 exists and processing the area in detail.

The detailed processing in this area is executed by obtaining the correspondence between the left image and the right image by the general stereo method. However, the area can be limited to a much narrower area than the entire image. Further, since the process of detecting the area is a process of simply comparing the left image and the right image based on the correspondence relationship, the process can be executed at an extremely high speed. Therefore, even if the increase in the number of processes for detecting the region is taken into consideration, the processing time is shorter than that for processing the entire image by the stereo method, and the processing can be performed with high accuracy.

As shown in FIG. 14, it is possible to increase the number of input images and perform more accurate processing.
In this case, similar processing is performed on the center image of the projection surface 25 and the left image of the projection surface 12, and the center image of the projection surface 25 and the right image of the projection surface 13.

Further, in FIG. 9, the subject 14 itself may move or the viewpoints 10 and 11 may move. When such a movement occurs, it is necessary to recalculate the correspondence relationship with the subject 14. However, if such movement is minute, the change in the correspondence relationship obtained in advance is also minute. Therefore, a new corresponding position is generated around the position 17 of the right image corresponding to the position 16 of the left image, based on the correspondence obtained in advance, outside the region detected as the presence of a new object. Ask. By constantly updating the correspondence relationship obtained in advance with this new correspondence relationship, the subject 14 and the viewpoints 10 and 11 can be updated.
It is possible to prevent the influence of the change in the correspondence caused by the minute movement of the.

[0021]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a block diagram showing the structure of a moving object recognition apparatus according to a first embodiment of the present invention. The moving object recognition apparatus according to the first embodiment includes an image input unit 1, a correspondence detection unit 2, a correspondence storage unit 3, a region detection unit 4, and an object recognition unit 5.
It is composed of

Referring to FIG. 2, the correspondence detecting section 2 comprises a block dividing means 201, a block selecting means 202, a correlation value calculating means 203, a correspondence detecting means 204 and a correspondence outputting means 205. .

Referring to FIG. 3, the processing of the correspondence detection unit 2 is performed by dividing the left image / right image block step A1, the left image block selecting step A2, the right image block selecting step A3, and the selected block autocorrelation value. It comprises a calculation step A4, a right image block end determination step A5, an autocorrelation value maximum right image block selection step A6, a left image block end determination step A7, and a pixel-to-pixel correspondence output step A8.

Referring to FIG. 4, the area detecting section 4 comprises a pixel reading means 401, a pixel determining means 402, a threshold setting means 403, and an area determining means 404.

Referring to FIG. 5, the processing of the area detecting section 4 is performed by a left image pixel selecting step B1, a corresponding right image pixel position reading step B2, and a pixel value comparing step B3.
A pixel value difference / threshold comparison step B4, a moving object nonexistence pixel position determination step B5, a moving object existence pixel position determination step B6, and a left image pixel end determination step B7.

FIG. 6 is an explanatory diagram showing the contents of the correspondence relationship stored in the correspondence storage unit 3. As shown in FIG. 6, it is assumed that the pixel 27 of the left image and the pixel 28 of the right image correspond to each other. Here, the pixel position of the pixel 27 is (i, j)
The pixel position of 8 is (p, q). In this case, the correspondence storage unit 3 stores (p, q) as the correspondence relationship of the pixels 27 of the left image.
Is stored. Further, (i, j) is stored as the correspondence relationship of the pixel 28 of the right image. In this way, the correspondence relationship between all the pixels of the left image and the right image is stored.

Next, the operation of the moving object recognizing apparatus according to the first embodiment having the above structure will be described.

The image input unit 1 receives analog luminance signals of a left image and a right image forming a stereo image from a stereo image input device such as a stereo camera and digitizes them as pixels to form a left image composed of a plurality of pixels. And the right image are output. For simplicity of explanation, the stereo image input here is a stereo image of the subject 14 as shown in FIG. 9, but this does not impair generality.

Next, the correspondence detecting unit 2 detects the correspondence between the left image and the right image. For example, when the position 15 of the subject 14 is projected onto the position 16 of the left image projection surface 12 and the position 17 of the right image projection surface 13 as shown in FIG. , Left image projection plane 1
It means that the position 16 of 2 corresponds to the position 17 of the projection plane 13 of the right image.

More specifically, as shown in FIG. 2 and FIG.
First, the block division unit 201 divides the left image and the right image into rectangular blocks each including, for example, m × n (both m and n are positive integers) pixels (step A1).

Next, the block selecting means 202 selects the blocks of the left image and the right image one by one (step A).
2, A3).

Then, the correlation value calculation means 203 calculates the autocorrelation value between the selected blocks (step A).
4). Here, the luminances of pixels in the blocks of the left image and the right image are L (i, j) and R (i, j) (i = 1, 1), respectively.
2, ..., _M ; j = 1, 2, ..., N), the average of the respective brightnesses is μ _L , μ _R , and the variances are σ _L , σ _R.
An autocorrelation value c such as is calculated.

[0033]

[Equation 1]

When the autocorrelation value c for all blocks of the right image is calculated for one block of the left image (step A5), the correspondence detecting means 204 determines the right image having the largest autocorrelation value c. Block is detected as a block corresponding to the block of the left image, and information indicating the combination of the block of the corresponding left image and the block of the corresponding right image is output (step A6).

After performing the above processing for all the blocks of the left image (step A7), the corresponding output means 205
Receives the information indicating the combination output by the correspondence determining unit 204, and indicates the pixel position of the right image to which each pixel of the left image corresponds, and the pixel position of the left image to which each pixel of the right image corresponds. The information is output as a correspondence relationship (step A8). However, for each pixel in the block, the pixel position is determined on the assumption that pixels at the same position correspond to each other.

The correspondence storage unit 3 shows information indicating the pixel position of the right image corresponding to each pixel of the left image detected by the correspondence detecting unit 2 and the pixel position of the left image corresponding to each pixel of the right image. The information is stored as a correspondence.

Next, the image input section 1 continues to input the stereo image, and the area detection section 4 detects the area of the stereo image where the moving object exists. As shown in FIG.
When only the subject 14 exists, the correspondence relationship between the left image and the right image does not change from the correspondence relationship stored in the correspondence storage unit 3. Therefore, the area detection unit 4 uses the correspondence storage unit 3
The left image and the right image are compared in accordance with the correspondence relationship stored in 1., and if they are the same, they are not detected as an area where a moving object exists.

As shown in FIG. 11, when an object 18 different from the subject 14 exists, the position 16 of the left image projection plane 12 and the position 17 of the right image projection plane 13 are different images. Therefore, the area detection unit 4 compares the left image and the right image according to the correspondence relationship stored in the correspondence storage unit 3,
The position when different is detected as the region where the moving object exists. However, as shown in FIG.
16 corresponds to the position 17 on the projection plane 13 of the corresponding right image.
However, the image of the object 18 is not. Thus, the detected area does not necessarily match the area on the image of the object 18 itself.

More specifically, as shown in FIGS. 4 and 5,
First, the pixel reading unit 401 reads the pixel position of the corresponding right image stored in the correspondence storage unit 3 for each pixel of the left image (step B1) (step B2). Then, the pixel determining unit 402 compares the luminances of the pixel positions of the corresponding left image and right image with each other (step B3), and determines whether they are the same. Here, a threshold is set in advance in the threshold setting means 403 in consideration of the influence of noise and the like, and when the difference in brightness is less than or equal to the threshold (step B4),
It is determined that there is no moving object at the pixel position of the left image (step B5), and if the difference in brightness is larger than the threshold value (step B4), it is determined that the moving object exists at the pixel position of the left image (step B4). Step B6). It should be noted that in order to determine whether the left image and the right image are the same, it is possible to use values such as color difference and hue in addition to comparing the brightness of the images.
The pixel determination unit 402 performs such determination processing on all the pixels of the left image (step B7), and outputs information indicating the pixel position determined to have a moving object. Also, the same processing is performed on the right image.

FIG. 15 is an explanatory diagram showing the areas detected by the area detector 4. As shown in FIG. 15, the area detection unit 4 detects an area 26, which is indicated by the diagonal lines in the left image projected on the projection surface 12, as an area in which a moving object exists. However, as shown in FIG.
When the object 22 is monochromatic when compared according to the correspondence relationship with 5, the position 2 on the projection plane 12 of the left image is
3 and the position 24 of the projection plane 13 of the right image may be the same image, so that a region such as a region 26 'that is not detected as a region in which a moving object exists is generated. Such an area 26 'is a hatched area 26
Area 26 ', the area detection unit 4 detects the area 26' inside the area 26 indicated by the diagonal lines as an area in which the moving object exists, and outputs a stereo image in the area.

Therefore, the area determining means 404 detects the area 26 'as shown in FIG. 15 by receiving the information output by the pixel determining means 402 and indicating the pixel position where the moving object is determined to exist. The pixel position of this area 26 'is also changed to the pixel position where the moving object exists. This area 26 '
In order to detect, the pixel position around the pixel position where the moving object does not exist is searched, and when there are pixels around which the moving object exists, it can be determined as the pixel position corresponding to the region 26 ′.

Finally, the object recognition unit 5 performs the same processing as that of the correspondence detection unit 2 on the stereo image in the area in which the moving object exists, which is output by the area detection unit 4, to obtain a left image and a right image. Seek correspondence. Then, the distance to the moving object is obtained based on the principle of triangulation. If the distance to the moving body is obtained, the shape of the moving body can be obtained by changing the distance.

As described above, the object recognition unit 5 analyzes only the stereo image in the area where the moving object exists in detail, so that the processing is executed much faster than the analysis of the entire stereo image. be able to. Further, since the processing in the area detection unit 4 only compares the luminances of the left image and the right image according to the correspondence relationship, very high-speed processing is possible. Therefore, the processing in the entire moving object recognition apparatus of the present invention,
It can run much faster. In particular, when processing moving images that are sequentially input in time series, the effect of this speedup is great.

Further, since the correspondence storage unit 3 stores the correspondence between the left image and the right image of the subject 14, even if the illumination condition changes or the shadow of another object falls on the subject 14, , Does not affect the correspondence. Therefore, even in such a case, the moving object will not be erroneously detected.

In the processing in the object recognition unit 5,
When analyzing the stereo image in the area where the moving body is present,
The correspondence between the left image and the right image is detected. The detection process may be executed by the correspondence detection unit 2.

When comparing the left image and the right image according to the correspondence, it is possible to compare not only pixel units but also small areas having a shape such as a square. By comparing the small areas with each other in this way, the area where the moving object exists can be detected more correctly.

FIG. 7 is a block diagram showing the structure of a moving object recognizing apparatus according to the second embodiment of the present invention. The moving object recognizing device according to the second embodiment includes an image input unit 1 in the moving object recognizing device according to the first embodiment shown in FIG. Auxiliary image input unit 6 for inputting auxiliary images from two viewpoints around the viewpoint of the image
It is divided into

Next, the operation of the moving object recognizing apparatus according to the second embodiment thus constructed will be described focusing on the points different from the moving object recognizing apparatus according to the first embodiment.

The main image input section 1'inputs a main image.
The auxiliary image input unit 6 also inputs a left image and a right image arranged on the left and right of the main image.

Next, the correspondence detection unit 2, the correspondence storage unit 3, the area detection unit 4, and the object recognition unit 5 are the same as those in the first embodiment for the combination of the main image and the left image and the combination of the main image and the right image. The same processing as that of the moving object recognizing device is performed to obtain the shape of the moving object and the distance to the moving object. Then, the object recognition unit 5
Is the distance to the moving object obtained by the combination of the main image and the left image, and the moving object obtained by the combination of the main image and the right image, for each pixel in the area where the moving object of the main image exists. And the distance to are averaged and output as the final recognition result.

In the second embodiment, the left image and the right image are described as being input from the auxiliary image input section 6, but
Arrangement of auxiliary images is arbitrary, and three or more auxiliary images can be used. Although the processing amount increases by increasing the number of auxiliary images, the accuracy of the recognition result can be improved.

FIG. 8 is a block diagram showing the structure of a moving object recognizing apparatus according to the third embodiment of the present invention. The moving object recognizing device according to the third embodiment is different from the moving object recognizing device according to the first embodiment shown in FIG. 1 in that the correspondence storage unit 3 stores the area outside the area detected by the area detecting unit 4. The correspondence update unit 7 that detects a change in the created correspondence relationship and updates the stored contents of the correspondence storage unit 3 is added.

Next, the operation of the moving object recognizing apparatus according to the third embodiment constructed as described above will be described focusing on the points different from the moving object recognizing apparatus according to the first embodiment.

Correspondence updating section 7 receives the information indicating the area where the moving object exists, which is output from area detecting section 4, and the correspondence relationship stored in correspondence storing section 3.

Here, the area other than the area where the moving object exists is a stereo image of the subject 14. Therefore, the correspondence relationship of this area is stored in the correspondence storage unit 3. However, there is a possibility that the correspondence relationship may change due to a slight change in the relative position between the subject 14 and the stereo image input device, a slight deformation of the subject 14, or the like.

Therefore, the correspondence updating unit 7 newly obtains the correspondence relation of the areas other than the area where the moving body exists. This process is almost the same as the process in the correspondence detection unit 2, except that the correspondence relation is not obtained for the pixels included in the region where the moving object exists.

Further, as shown in FIG. 3, the correspondence detecting unit 2 obtains the autocorrelation value of one block of the left image with all the blocks of the right image, whereas the correspondence updating unit 7 Another difference is that the autocorrelation value with some blocks in the right image is obtained.

That is, in the area other than the area in which the moving object exists, the correspondence relationship stored in the correspondence storage unit 3 hardly changes. Therefore, the size of the search range is set in advance in the correspondence updating unit 7. As shown in FIG.
It is assumed that the pixel 27 of the left image and the pixel 28 of the right image correspond to each other. In this case, when obtaining the autocorrelation value between the block including the pixel 27 of the left image and the block of the right image, the pixel 2 indicated by the size of the preset search range is calculated.
Only blocks included in the search area 29 around 8 are to be processed. As a result, the amount of processing is much smaller than that of obtaining all the blocks and the autocorrelation value of the right image.

Finally, the correspondence updating unit 7 stores the newly obtained correspondence relationship in the correspondence storage unit 3. As a result, even if a slight change in the relative position between the subject 14 and the stereo image input device or a slight deformation of the subject 14 itself occurs, the correspondence relation that is always stored in the correspondence storage unit 3 is the latest correspondence relation. Can be updated to.

If the variation or deformation is not small,
The area detection unit 4 detects the area where the moving object exists.
Therefore, even if a large change occurs in the subject 14, no malfunction occurs.

[0061]

As described above, the moving object recognition apparatus of the present invention has an image input section for inputting a stereo image consisting of a left image and a right image, and a left image of the stereo image input by the image input section. Correspondence detection section for finding the correspondence relationship with the right image, correspondence storage section for storing the correspondence relationship found by the correspondence detection section, correspondence relationship stored in the correspondence storage section, and stereo image input from the image input section at any time An area detection unit that detects an area in which a moving object is present based on, and an object recognition unit that obtains the shape and distance of the moving object in the area detected by the area detection unit, thereby providing a highly accurate and high-speed animal. There is an effect that the body can be detected and its shape and distance can be measured.

Further, the moving object recognition apparatus of the present invention includes a main image input section for inputting a main image, an auxiliary image input section for inputting a plurality of auxiliary images from viewpoints around the viewpoint of the main image, and the main image. Correspondence detection unit for obtaining the correspondence relation between the correspondence image and a plurality of auxiliary images, a correspondence storage unit for storing the correspondence relation obtained by the correspondence detection unit, the correspondence relation stored in the correspondence storage unit, and input from the main image input unit at any time An area detection unit that detects an area in which a moving object exists based on a plurality of auxiliary images that are input from the main image and the auxiliary image input unit, and the shape of the moving object in the area detected by the area detection unit and By providing the object recognition unit that calculates the distance, there is an effect that the moving object can be detected with high accuracy and high speed, and the shape and the distance thereof can be measured.

Furthermore, the moving object recognition apparatus of the present invention detects a change in the correspondence stored in the correspondence storage unit with respect to the outside of the area detected by the area detection unit, and updates the storage content of the correspondence storage unit. By providing the part, there is an effect that the moving object can be detected with high accuracy and high speed, and the shape and the distance thereof can be measured.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a moving object recognition apparatus according to a first embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of a correspondence detection unit in FIG.

FIG. 3 is a flowchart showing a process of a correspondence detection unit in FIG.

FIG. 4 is a block diagram showing a configuration of an area detection unit in FIG.

FIG. 5 is a flowchart showing a process of an area detection unit in FIG.

FIG. 6 is a diagram illustrating a correspondence relationship stored in a correspondence storage unit in FIG.

FIG. 7 is a block diagram showing a configuration of a moving object recognition apparatus according to a second embodiment of the present invention.

FIG. 8 is a block diagram showing a configuration of a moving object recognition apparatus according to a third embodiment of the present invention.

FIG. 9 is a diagram showing the relationship between a stereo image and a subject from above.

FIG. 10 is a diagram illustrating a method of measuring a distance to a subject surface by a stereo method.

FIG. 11 is an explanatory diagram when a new object moves.

FIG. 12 is a diagram illustrating that a different region between the left image and the right image is wider than a region where an object actually exists.

FIG. 13 is a diagram illustrating a case where a large object exists.

FIG. 14 is an explanatory diagram in the case of increasing the number of input images and performing more accurate processing.

FIG. 15 is a diagram illustrating an area detected by an area detection unit in FIG.

[Explanation of symbols]

 1 Image Input Section 1'Main Image Input Section 2 Correspondence Detection Section 3 Correspondence Storage Section 4 Area Detection Section 5 Object Recognition Section 6 Auxiliary Image Input Section 7 Correspondence Update Section 10, 11 Viewpoints 12, 13, 25 Projection Surface 14 Subject 15- 17, 19 to 21, 23, 24 Position 18, 22 Object 26, 26 'Area 27, 28 Pixel 29 Search area 201 Block dividing means 202 Block selecting means 203 Correlation value calculating means 204 Correspondence detecting means 205 Corresponding output means 401 Pixel reading Means 402 Pixel judging means 403 Threshold setting means 404 Region determining means

Claims (3)

[Claims] 1. An image input section for inputting a stereo image consisting of a left image and a right image, and a correspondence detecting section for obtaining a correspondence relationship between the left image and the right image of the stereo image input by the image input section, A correspondence storage unit that stores the correspondence relationship obtained by the correspondence detection unit, and detects a region where a moving object exists based on the correspondence relationship stored in the correspondence storage unit and the stereo image input from the image input unit as needed. A moving object recognizing device, comprising: an area detecting unit for detecting a moving object shape; 2. A main image input unit for inputting a main image, an auxiliary image input unit for inputting a plurality of auxiliary images from viewpoints around the viewpoint of the main image, the main image and the plurality of auxiliary images. The correspondence detection unit that obtains the correspondence relation between the correspondence detection unit, the correspondence storage unit that stores the correspondence relation obtained by the correspondence detection unit, the correspondence relation stored in the correspondence storage unit, and the main input that is input from the main image input unit at any time. A region detection unit that detects a region in which a moving object exists based on an image and a plurality of auxiliary images that are input from the auxiliary image input unit at any time, and the shape and distance of the moving object in the region detected by this region detection unit. An object recognizing device, comprising: 3. The correspondence update unit that updates a stored content of the correspondence storage unit by detecting a change in the correspondence relation stored in the correspondence storage unit with respect to a region outside the region detected by the region detection unit. Alternatively, the moving object recognizing device according to item 2.