JPH08315149A - Intruding body recognition device - Google Patents

Abstract

PURPOSE: To accurately recognize intruding bodies without misdetection even when they are photographed overlapping each other. CONSTITUTION: A processing circuit 7 analyzes whether or not there is a possibility that a variation area, which was judged not to be an intruding body, was judged so because of overlapping of intruding bodies as to judged information showing that the variation area is not the intruding body. When it is judged by the analysis that there is the possibility of the overlap of the intruding bodies, that is reported to a processing circuit 8. The processing circuit 8 restores the variation area wherein it is estimated that the intruding bodies overlap to several variation areas wherein it is estimated that a single intruding body is represented according to the information showing that there is the possibility of the overlap of the intruding bodies. When it is judged that one of the variation areas obtained by the restoration is the intruding body, a processing circuit 9 judges that there is the intruding body to be detected in the variation area wherein it is estimated that the intruding bodies overlap each other.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an intruding object recognizing device for recognizing the presence or absence of an intruding object which is a recognition target from image signals taken by, for example, a television camera in a state where a plurality of intruding objects partially overlap each other. is there.

[0002]

2. Description of the Related Art In recent years, an intruding object recognition system for detecting a specific intruding object from an image taken by a television camera or the like has been researched and developed.

For example, the following documents propose application examples of such a system. Reference: Proceedings of the 9th Image Sensing Technology Symposium in Industry, 1994, page 83.
~ 88, "Intrusion monitoring method on steel premises using image processing".

The detection principle of the intrusion monitoring proposed in this document is to obtain a difference in brightness between the background image and the monitor image for each corresponding pixel, and take a brightness histogram of this difference. Therefore, when there is no intrusion, a luminance peak appears near the luminance value of 0. However, when there is an intrusion, there is no correlation between the background image and the monitoring image in the area of the image corresponding to the intrusion, and therefore the histogram has a shape in which a broad distribution overlaps the peak. Therefore, with reference to the central brightness of the peak on the histogram, it is judged that there is an intrusion when another distribution is detected at a position separated by a certain threshold value or more.

[0005]

However, in the conventional technique described above, in the case of a change area in which a plurality of intruding objects appear to overlap each other, the intruding object is completely different from one changing area. For this reason, the feature amount such as the area and the circumscribed quadrangle obtained from the change region may be different from the preset reference feature amount of the particular intruding object, and the intruding object may be overlooked.

From the above, it is demanded to provide an intruding object recognition apparatus which can accurately recognize an intruding object without erroneously detecting the intruding object even when the intruding objects are photographed in an overlapping manner. ing.

[0007]

Therefore, the intruding object recognizing apparatus of the present invention recognizes the intruding object recognizing from the changing area and the extracting means for extracting the changing area from the image signal photographed in a state where a plurality of intruding objects partially overlap each other. An analysis unit that analyzes whether or not the target single intruding object is represented, and if the analysis unit determines that it does not represent the single intruding object that is the recognition target, the overlapping state of the intruding objects And a reconstructing means for reconstructing the changing regions of some intruding objects from the changing regions presumed to overlap with each other, and the presence or absence of the intruding object to be recognized from the restored changing regions. By recognizing, the above-mentioned problems are solved.

[0008]

According to the configuration of the present invention, even if a plurality of intruding objects partially overlap with each other, a change area is extracted, and it is analyzed whether or not a single intruding object that is a recognition target is represented from the change area. If it is judged that it is a recognition target, the presence or absence of an intruding object can be output immediately, and if it is judged that it does not represent a single intruding object that is a recognition target, the overlapping state of the intruding object is analyzed and By restoring the change regions of some intruding objects from the change regions that are estimated to overlap with each other, the feature amount of the restored changing regions and the reference features of the intruding object that is the recognition target obtained in advance. The presence or absence of an intruding object, which is a recognition target, can be easily recognized from the result of comparison with the amount, and thus can be recognized more accurately than in the past.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT A preferred embodiment of the present invention will be described with reference to the drawings. "First embodiment": FIG. 1 is a functional block diagram of an intruding object recognition apparatus of this embodiment. In FIG. 1, the intruder recognition apparatus includes a video camera 11, a video interface circuit 12, a background image memory circuit 1, an input image memory circuit 2, a difference processing circuit 3, a binarization processing circuit 4, Change area cutout processing circuit 5 and single intruding object determination processing circuit 6
It includes an intruding object weight judgment processing circuit 7, a single intruding object restoration processing circuit 8, a single intruding object determination processing circuit 9, and an output processing circuit 10.

The video camera 11 is for picking up an image of an intruding object and the like, and the captured image in which the intruding object is overlapped is converted into an analog signal (for example, NTSC signal, PAL signal, SEC).
It is output as an AM signal, an RGB signal, or the like) and given to the video interface circuit 12. When the video interface circuit 12 is supplied with the captured image signal, the video interface circuit 12 converts the captured image signal into a digital signal and supplies it to the background image memory circuit 1 or the input image memory circuit 2. Particularly, the input image memory circuit 2 stores the input image signals sequentially captured by the video camera 11 as digital signals.
On the other hand, the background image memory circuit 1 stores, as a digital signal, a background image signal captured when there is no intruding object (subject).

The difference processing circuit 3 is an input image memory circuit 2
Input image signal given from the background image memory circuit 1
From the background image signal given from the difference between the pixels (for example, the absolute value of the density difference), to generate a difference image signal,
This is given to the binarization processing circuit 4. When the differential image signal is given, the binarization processing circuit 4 performs binarization with an appropriate threshold value, generates a differential binarized image signal, and gives it to the change region cutout processing circuit 5. In this differential binarized image signal, the value of the pixel whose density value has changed from the background image signal in the input image signal is set to 1, and the value of the pixel which has not changed is set to 0.

The change area cut-out processing circuit 5 expands the difference binary image several times and then contracts the same number of times in order to combine the change areas representing the intruding object into one.
Inflation means that when there is a pixel with a value of 1 around a pixel with a value of 0,
The pixel having the value 0 is converted into 1. In other words, it expands around the change area by one pixel. Contraction is the opposite, and reduces the area around the change area by one pixel. The change area estimated to represent one intruding object by these processes is extracted as one area. The change area thus obtained is surrounded by, for example, a circumscribed rectangle, and the area surrounded by the circumscribed rectangle is cut out from the differential binarized image.

The single intruding object determination processing circuit 6 detects a single intruding object having an overlap from the area of the differential binarized image cut out from the changing area cutting processing circuit 5, and the changing area is a single intruding object. Is determined.

The intruding object weight judgment processing circuit 7 continuously performs the intruding object detection processing in the case of overlapping as shown in FIG. 8 in response to the judgment information from the single intruding object judgment processing circuit 6 that the object is not the intruding object. It is a thing. That is, specifically, it is analyzed whether or not there is a possibility that the change area that has been determined to be not an intruding object by the processing up to that time may have been so determined by "overlapping of intruding objects". If it is determined by this analysis that there is no possibility that the intruding objects overlap each other, that effect is output.
When it is determined that there is a possibility that the intruding objects overlap each other, the fact is further given to the single intruding object restoration processing circuit 8.

The single intruding object restoration processing circuit 8 uses the information indicating that there is a possibility that the intruding objects are overlapped from the intruding object weight judgment processing circuit 7, and the single intruding object restoration processing circuit Are restored and given to the single intruding object determination processing circuit 9.

The single intruding object determination processing circuit 9 performs the same processing as that of the single intruding object determination processing circuit 6 from some of the restored changed areas from the single intruding object restoration processing circuit 8. That is, if it is determined that at least one of the restored changed areas is an intruding object to be detected,
Assuming that the intruding object to be detected exists in the change area in which it is estimated that the intruding objects are overlapped, the result is output to the output processing circuit 10.

The output processing circuit 10 outputs, for example, a determination result indicating that there is an intruding object in the change area from the single intruding object determination processing circuit 6 or the single intruding object determination processing circuit 9, and outputs the intruding object weight / inward determination processing circuit 7 or the single operation. The intrusion object determination processing circuit 9 outputs the determination result indicating that there is no intrusion object in the changed area.

(Detailed Configuration of Single Intruding Object Determination Processing Circuit 6): FIG. 2 is a specific functional configuration diagram of the single intruding object determination processing circuit 6. In FIG. 2, the single intruding object determination processing circuit 6 is a single change area feature amount calculation processing circuit 6
1 and a single intruding object determination processing circuit 62. The single change region feature amount calculation processing circuit 61 calculates change region feature amounts such as the area of the change region and the aspect ratio of the circumscribed quadrangle from the region of the difference binarized image cut out from the change region cutout processing circuit 5. , The single intruding object determination processing circuit 62. The single intruding object determination processing circuit 62 compares the characteristic amount from the single changing area feature amount calculation processing circuit 61 with a preset reference characteristic value of the changing area of a specific intruding object (such as a person) to be detected. Then, it is determined whether or not the change area in the input image is a specific intruding object to be detected, the result is output, and if the extracted change area is determined to be the intruding object, this is output processing circuit 10 If it is determined that the object is not an intruding object, the effect is given to the intruding object weight / weight determination processing circuit 7.

(Detailed Structure of Intruding Object Weight Determining Processing Circuit 7): A change area determined to be not an intruding object by the processing up to the single intruding object determining processing circuit 6 is determined as such by overlapping intruding objects. In order to analyze whether or not there is a possibility that the change area is surrounded, the change area is surrounded by a circumscribed quadrangle as shown in FIGS. 3A and 3B, and the circumscribed quadrangle is vertically scanned from above to change the change area. The values of the coordinates of the points that first collide with are calculated for all on the x-axis in the circumscribed quadrangle, and the silhouette function f (x) as shown in FIGS. After that, the silhouette function f (x) is smoothed in order to remove the influence of noise.

Further, as can be seen from FIG. 3 (c), when the intruding objects are overlapped with each other, many irregularities are generated in the silhouette function f (x), and many mountains can be formed.
When the intruding objects do not overlap with each other as shown in FIG. 3D, only one mountain can be formed in the silhouette function f (x).

Therefore, the maximum points are calculated from the silhouette function f (x), and if there are two or more maximum points, it is determined that the intruding objects may overlap, and the single intruding object restoration processing circuit 8 is executed. The processing is continued. Further, if the number of maximum points is one, it is determined that there is no possibility that the intruding objects overlap each other, and the processing ends.

(Detailed Structure of Single Intruding Object Restoration Processing Circuit 8): In the single intruding object restoration processing circuit 8, a change estimated to represent a single intruding object from a change region in which the intruding objects are estimated to overlap. The procedure for restoring the area will be described with reference to FIG. Therefore, first, in FIG.
As shown in (a), the values of the x-coordinates of the maximum points at the extreme ends of the silhouette function f (x) determined by the intruder weight / weight determination processing circuit 7 are determined as xl and xr.

Next, the change region is divided in the vertical direction by the x-coordinates xl and xr. In this case, the area can be divided into three areas, which are A, B, and C from the left. Further, since the inside change area B is complicatedly overlapped by the intruding object, it is not easy to cut out a single intruding object from this portion. Therefore, it can be said that the changing regions of A and C at both ends represent the right half body or the left half body of the intruding objects at both ends, since there is little overlap between the intruding objects.

Here, for example, "if the intruding object to be detected is a person who is walking (the walking object may be in the lateral direction or in the front), the change region is substantially left-right symmetric". Therefore, by utilizing such a property, FIG.
As shown in FIG. 4B, the change regions A and C at both ends are x = x
They are folded inward and symmetrically moved inward with respect to the axes of l and x = xr.

That is, as shown by A'and C'in FIG. 4C, the change area A on the left side of x = xl is on the right side with respect to x = xl, and the change area on the right side of x = xr. C is x = xr
It moves symmetrically to the left with respect to. Furthermore, when the change areas A ′ and C ′ that have moved symmetrically and the original change areas A and C that have not been moved symmetrically are respectively combined, the single intruding objects existing at both ends are restored.

FIGS. 5 and 6 show the case where similar processing is performed on different overlapping human image change regions. Also,
As can be seen from FIGS. 4 (c) and 6 (c), it can be seen that the intruding objects at both ends are roughly restored by the above-described processing. The next single intruding object determination processing circuit 9 recognizes the intruding object from these two changed regions, and if at least one of them is determined to be an intruding object, it is estimated that the intruding object overlaps. Assuming that there is an intruding object to be detected in the changed area, the result is output.

(Operation of Device): An analog image signal from the video camera 11 is given to the video interface circuit 12 and converted into a digital image signal here. The sequentially captured input image signals supplied from the video interface circuit 12 are stored in the input image memory circuit 2, and the background image signal is stored in the background image memory circuit 1.

The difference between the input image signal and the background image signal provided from the background image memory circuit 1 is obtained by the difference processing circuit 3 to generate a difference image signal. This difference image signal is given to the binarization processing circuit 4, binarization is performed with an appropriate threshold value, a difference binarization image signal is generated, and given to the change region cutout processing circuit 5.

The change area cut-out processing circuit 5 extracts the change area of the overlapping intruding object from the differential binary image and supplies it to the single intruding object determination processing circuit 6. In the cut-out differential binary image area, a single intruding object determination processing circuit 6 detects an overlapping single intruding object, and determines whether or not the changed area represents the single intruding object. If the changed region extracted here is determined to be an intruding object to be recognized, the fact is given to the output processing circuit 10. If it is determined that the object is not an intruding object, the fact is given to the intruding object weight / weight determination processing circuit 7.

Next, the change area determined to be not an intruding object by the processing up to the single intruding object determination processing circuit 6 is
The intruder weight / weight determination processing circuit 7 analyzes whether or not there is a possibility of such determination due to the overlap of the intruding objects. The maximum point is calculated from the silhouette function f (x) in this analysis, and if the number of maximum points is one, it is considered that there is no possibility that the intruding objects overlap, and the output processing circuit 10
To notify. If the number of maximum points is two or more, it is determined that the intruding objects may overlap, and the single intruding object restoration processing circuit 8 performs the processing.

The single intruding object restoration processing circuit 8 restores the changed area estimated to represent the single intruding object from the changing area estimated to overlap the intruding objects, and supplies it to the single intruding object determination processing circuit 9. . If the single intruding object determination processing circuit 9 determines from the restored changed regions that at least one of the restored changed regions is an intruding object to be detected, a change in which it is estimated that the intruding objects overlap Assuming that an intruding object to be detected exists in the area, the result is output to the output processing circuit 10. Output processing circuit 1
In response to the notification from the single intruding object determination processing circuit 6 or the single intruding object determination processing circuit 0, the determination result indicating that there is an intruding object in the change area is displayed and output, and the intruder weight / weight determination processing circuit 7 or the single intruding object determination processing In response to the notification from the circuit 9, the determination result indicating that there is no intruding object in the changed area is displayed and output.

(Hardware Configuration of Device): FIG. 7 is a hardware configuration diagram of an example for realizing the intruding object recognition device of the embodiment of the present invention. In FIG. 7, the hardware configuration of the intruder recognition apparatus is the video camera 1.
1, a workstation body 22, a CRT display 26, a keyboard 25, a mouse 24, and an external storage device 27. The workstation main body 22 is composed of a CPU 20, a main memory 21, and a video board 28.

The image signal taken by the video camera 9 is
An analog signal (RGB signal or NTSC signal) is provided to the video board 28. This video board 28
Converts the input image signal from an analog signal into a digital signal and stores the background image signal and the sequentially captured input image signal in the main memory 21. The main memory 21 stores the image signal from the video board 28, and also stores various program data for performing the above-described image recognition of an intruding object, various necessary work data, and the like. . The CPU 20 processes these data in the main memory 21 to perform image recognition processing of an intruding object.

The external storage device 27 is used as a backup of the program data stored in the main memory 21, and also stores various data necessary for image recognition processing of an intruding object, and if necessary, the main memory. 21. The result of the image recognition processing of the intruding object is C
The output is displayed on the RT display 26. Mouse 24,
The keyboard 25 and the like are used as input means for data selection and data input in the processing stage. By configuring the intruding object recognition device in this manner, the above-described functional configuration can be realized, and the intruding object can be recognized with high accuracy.

(Effect of First Embodiment): According to the first embodiment described above, in the process of detecting a specific intruding object from an image taken by a television camera or the like, the intruding object is as shown in FIG. Even when the change areas of multiple intruding objects appear to overlap with each other, using the areas at both ends where the change areas extracted by the difference processing and the binarization processing do not overlap,
By moving the area symmetrically inward, the changed area of a single intruding object is restored, and by performing detection processing based on that changed area, it is possible to detect the intruding object even when the intruding objects appear to overlap. It can be

Further, the feature amount of the changing region is obtained from the area of the changing region or the aspect ratio of the circumscribing polygon, and the feature amount of the changing region is compared with the reference feature amount of the intruding object that is the recognition target. By analyzing whether or not it is displayed, the recognition result can be easily output if the overlap is simple.

Further, as an analysis of the overlapping state of the intruding objects, a silhouette function is generated from the change region, the number of maximum points (or minimum points) of this silhouette function is analyzed, and the intruding objects are overlapped from the analysis result. By estimating whether or not there is an overlapping intruding object, the presence or absence of the overlapping can be accurately recognized.

Furthermore, from the position of the maximum point or the minimum point of the silhouette function of the change area where the intruding objects are estimated to overlap, one or more axes are set around the maximum point or the minimum point, and By analyzing the symmetry of the change area around the axis and restoring the change areas of some intruding objects in the change area from this symmetry analysis, even if the intruding objects overlap in a complicated manner, the recognition target can be recognized. It is possible to restore a change area of a certain single intruding object and facilitate recognition determination.

[Second Embodiment]: In the first embodiment described above, in the single intruding object determination process 6, the change area feature amount such as the area and the aspect ratio of the circumscribed quadrangle is calculated from the cut change area, and this feature is calculated. Whether the change area in the input image is a specific intruding object to be detected by comparing the amount of change with a reference feature value of the changing area of a specific intruding object (for example, a person) to be detected that is set in advance. The process of determining is performed.

Therefore, an intruding object that is far from the installation location of the television camera that detects the intruding object from the image picked up by the television camera or the like has a small area of change area on the image. On the other hand, near the TV camera installation location,
Even if a small object moves, the area of the change region increases. Therefore, if you try to detect an intruding object that is far from the installation location of the TV camera and lower the area threshold setting, the change area caused by the movement of a small object other than a person near the TV camera or the change area caused by noise. Therefore, there is a possibility that an intruding object may be overlooked, or a non-intruding object that should be detected may be erroneously detected as an intruding object.

The aspect ratio of the circumscribed quadrangle, which is another parameter, is about 4 when a person normally walks from the side and when the legs and arms are closed compared to when they are opened.
~ 5 times larger. Therefore, when the intruder is determined from the aspect ratio, the same problem as in the case of determining the area can occur.

From the above, there is a possibility that the intruding object cannot be detected completely accurately from the area of the changing region and the aspect ratio of the circumscribing quadrangle.

Therefore, in the second embodiment, an intruding object recognizing device for accurately recognizing a person by utilizing the shape peculiar to the person generated when a person walks and for extremely reducing false detection is constructed.

Therefore, FIG. 9 is a functional block diagram of the intruding object recognition apparatus of the second embodiment. In FIG. 9, the intruding object recognition apparatus includes a video camera 11, a video interface circuit 12, a background image memory circuit 1, an input image memory circuit 2, a difference processing circuit 3, a binarization processing circuit 4,
A change area cutout processing circuit 5, a single intruding object shape determination processing circuit 106, an intruding object weight determination processing circuit 7,
It is composed of a single intruding object restoration processing circuit 8, a single intruding object shape determination processing circuit 107, and an output processing circuit 19.

In the second embodiment, the characteristic configuration is the single intruding object shape determination processing circuits 106 and 107.
Since the same functional components as those in the first embodiment are designated by the same reference numerals, detailed description thereof will be omitted. That is, up to the change area cutout processing circuit 5, the change area is extracted as in the first embodiment. After that, in the first embodiment, the feature amount of the change region such as the area and the aspect ratio is calculated from this change region, and the determination is performed based on this feature amount.

Therefore, specifically, the single intruding object shape determination processing circuit 106 of the second embodiment is realized by the circuit configuration as shown in FIG. That is, the single intruding object shape determination processing circuits 106 and 107 are composed of a change area division processing circuit 86, a projection histogram creation processing circuit 87, a feature amount extraction circuit 88, and an intruding object determination processing circuit 89.

The change area division processing circuit 86 vertically divides the extracted change area into three equal parts, and the projection histogram creation processing circuit 87 projects the divided change area on the horizontal axis. That is, the projection histogram is obtained by vertically adding the pixels having the value 1 representing the change area. Then, the feature amount extraction processing circuit 88 obtains an average value or a standard deviation from the projection histogram, and the feature amount and the change area of a specific intruding object (for example, a person) to be detected which is set in advance are also included. And the standard feature
By comparing and detecting the intruding object in the intruding object determining circuit 89, it is possible to detect the intruding object with higher accuracy even if the intruding objects appear to overlap each other.

An experiment of detecting 370 frames of overlapping image data of an intruding object photographed with the configuration of the second embodiment was successful and succeeded in detecting the intruding object with an accuracy of 94% or more.
There is no intruding object to be detected but other intruding objects are present, or false detection is made in 100 frames of image data in which noise has occurred (it is erroneous if there is no intruding object to be detected). When I conducted an experiment,
False detection could be suppressed to an accuracy of 20% or less.

(Effect of Second Embodiment): According to the configuration of the second embodiment described above, the single intruding object shape determination processing circuit 1
06 and 107, the changing area division process, the projection histogram creation, and the feature amount extraction can be performed to detect the intruding object with high accuracy by using the shape feature of the intruding object, and more accurately detect the intruding object. You can do it.

(Other Embodiments) (1) It should be noted that it is possible to combine the above embodiment with the configuration for detecting an intruding object with high accuracy by utilizing the shape feature of the intruding object.
With such a combination, the intruding object can be detected with higher accuracy.

(2) Further, the input image signal can be recognized not only from the television camera but also from a recording / reproducing signal from a video recording / reproducing device or the like. In addition to the display output, the recognition output is preferably printed out.

(3) Further, as the intruding object recognition, not only the person recognition but also any object such as a vehicle, an aircraft, an animal can be applied. Further, it is also possible to apply it to the overlapping of a plurality of face images.

(4) Furthermore, FIG. 1, FIG. 2, FIG. 9, FIG.
Although the figure of 0 is shown as showing the functional configuration, it can be seen as a flowchart showing the procedure (flow) of the processing. (5) Further, the intruding object recognition device of the present invention may be applied to a video conference system, various monitoring camera devices, a front door camera device, a collision prevention device mounted on a moving body such as a vehicle or an aircraft, and the like. You can also

[0054]

As described above, according to the constitution of the present invention,
Extraction means for extracting a change area from an image signal captured in a state where a plurality of intruding objects are partially overlapped, and analysis means for analyzing whether or not the change area represents a single intruding object that is a recognition target, When it is determined by the analysis of this analysis means that the single intruding object that is the recognition target is not represented, the overlapping state of the intruding objects is analyzed, and some of the change areas estimated to overlap the intruding objects are analyzed. A restoration unit that restores the changed area of the intruding object is provided, and by recognizing the presence or absence of the intruding object of the recognition target from the restored changed area, even when the intruding objects are imaged in an overlapping manner,
It is possible to realize an intruding object recognition device that accurately recognizes an intruding object that is a recognition target without erroneously detecting the intruding object.

[Brief description of drawings]

FIG. 1 is a functional configuration diagram of an intruding object recognition apparatus according to a first embodiment of the present invention.

FIG. 2 is a functional configuration diagram of a single intruding object determination processing circuit according to the first embodiment.

FIG. 3 is an explanatory diagram (No. 1) of a change area and a silhouette function of the embodiment.

FIG. 4 is an explanatory view (No. 1) of restoration of a single intruding object according to the embodiment.
Is.

FIG. 5 is an explanatory diagram (part 2) of a change area and a silhouette function according to the embodiment.

FIG. 6 is an explanatory view (No. 2) of restoration of a single intruding object in the example of the embodiment.

FIG. 7 is a hardware configuration diagram of an example of an intruding object recognition device according to the embodiment.

FIG. 8 is an explanatory diagram of a change area of an intruding object that is seen in an overlapping manner according to the embodiment.

FIG. 9 is a functional configuration diagram of an intruding object recognition device of a second embodiment.

FIG. 10 is a detailed functional configuration diagram of a single intruding object shape determination processing circuit according to the second embodiment.

[Explanation of symbols]

1 ... Background image memory circuit, 2 ... Input image memory circuit, 3
... difference processing circuit, 4 ... binarization processing circuit, 5 ... change area cutout processing circuit, 6, 9 ... single intruding object determination processing circuit,
7 ... Intruder weight judgment processing circuit, 8 ... Single intruding object restoration processing circuit, 10 ... Output processing circuit, 11 ... Video camera, 12 ... Video interface circuit, 61 ... Single change area feature amount calculation processing circuit, 62 ... Single intrusion Object determination processing circuit, 106, 107 ... Single intruding object shape determination processing circuit.

 ─────────────────────────────────────────────────── ─── Continued Front Page (72) Inventor Yuji Kuno 1-7-12 Toranomon, Minato-ku, Tokyo Oki Electric Industry Co., Ltd.

Claims (5)

[Claims] 1. An extracting unit for extracting a change region from an image signal captured by a plurality of intruding objects partially overlapped with each other, and analyzing whether or not the change region represents a single intruding object to be recognized. When the analysis means and the analysis means determine that the object does not represent a single intruding object that is a recognition target, the overlapping state of the intruding objects is further analyzed, and it is determined that the intruding objects overlap in this analysis. An intruding object recognition apparatus, comprising: a restoring unit that restores some changed areas of an intruding object from the estimated changed area, and recognizes the presence or absence of an intruding object that is a recognition target from the restored changed areas. . 2. The analyzing means obtains the feature amount of the change region from the area of the change region or the aspect ratio of the circumscribing polygon, and compares the feature amount with the reference feature amount of the intruding object which is the recognition target obtained in advance. The intruder recognizing device according to claim 1, wherein the intruder recognizing device analyzes whether or not a single intruding object which is a recognition target is represented. 3. The analyzing means divides the extracted change area, performs projection processing on the divided change area to obtain a projection histogram, analyzes the projection histogram, and obtains from this analysis. 2. The change area feature amount and a reference feature amount of an intruding object that is a recognition target obtained in advance are compared to analyze whether or not a single intruding object that is a recognition target is represented. The intruding object recognition device described. 4. The restoring means generates a silhouette function from the change region as an analysis of the overlapping state of the intruding object, analyzes the number of maximum points or minimum points of this silhouette function, and analyzes the intruding object from the analysis result. The intruding object recognition device according to any one of claims 1 to 3, wherein it is analyzed whether or not the two overlap. 5. The restoring means obtains the position of the maximum point or the minimum point of the silhouette function of the change area A estimated to be overlapping with the intruding object, and 1 or 2 or more around the maximum point or the minimum point. Is set, the symmetry of the change area A is analyzed around the axis, and the change areas of some intruding objects in the change area A are restored from the analysis of the symmetry. The intruding object recognition device according to claim 1.