JPH0737100A - Moving object detection and judgement device - Google Patents

Abstract

PURPOSE:To provide a moving object detection and judgement device capable of surely detecting an object moving in an area and picking up the image of the moving object in a state required for judgement without providing a dead angle in a wide monitoring area with few cameras. CONSTITUTION:This device is provided with a detection camera device 10 for moving object detection capable of detecting a wide range area, a judgement camera device for picking up the image in a state sufficient for judging whether or not the moving object is an object based on position information or the like outputted from the detection camera device 10 and a judgement processing part 30 for performing a judgement processing based on picture data image picked up by the judgement camera device. The detection camera device 10 is provided with a detection camera 11 facing upwards and an approximately conical reflection mirror 12 arranged above the detection camera 11 and can monitor the wide range of 360 degrees. A judgement camera 21 is rotated based on the position information and seizes the moving object and even at the time, the detection camera 10 performs a detection processing for the other moving object.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving object detection / determination device, and more particularly to a device suitable for detecting an illegal intruder or the like from image data obtained by an image pickup device such as a camera.

[0002]

2. Description of the Related Art Conventionally, as a device (image monitoring device, etc.) for detecting a moving object such as an intruder, which is generally used, for example, a camera is installed on a pole or the like and a monitoring area is provided with a depression angle. Is imaged, a predetermined pre-processing is performed on the obtained image data, and then the difference processing is performed with respect to the background image of the monitoring area stored in advance. Then, the difference image becomes an intruder (moving object), and thus the moving object is detected. Further, a feature amount (area, shape, etc.) is extracted from the moving object, image recognition processing is performed based on the feature amount data, and what the moving object is is determined.

[0003]

By the way, in order to make the above determination accurately, for example, in the case of a human being, a size of about 200 pixels is required. This is much larger than the minimum 9 pixels (3 × 3) required to detect a moving object, and is an area that can be monitored on a single screen captured by a single camera (area where judgment can be made). Is relatively narrow.
Moreover, the area imaged by the camera spreads in a fan shape, and the moving object located at a position away from the camera has a small shape in the imaged image, and the number of pixels required for the determination cannot be obtained, so that it is accurate. I can't judge. On the other hand, if it is too close to the camera, the entire image cannot be captured, and accurate determination cannot be made. Therefore, even if it is within the field of view of the camera, the effective monitoring area for determination is a part of it. Becomes

Therefore, there is no problem if the intruder's passage / existence area is narrow, such as the entrance of a building, but a relatively large area such as a factory, a substation, a power plant, or the like becomes the monitoring range. In this case, if an attempt is made to install the cameras so as to eliminate the blind spot, a very large number of cameras are required, the control thereof is complicated, and the cost is increased.

On the other hand, in order to reduce the number of cameras to be installed, the cameras are installed on a pan which can be swung freely, and the cameras are swung by a swivel driving device, and the monitoring area is temporally divided for determination. Image data can be obtained. However, in this case, since there is a blind spot region in which the camera is not facing (not imaged) in the monitoring region, there is a possibility that detection may be omitted.

The present invention has been made in view of the above background, and an object of the present invention is to provide an image in a state necessary for determination of a wide monitoring area without providing a blind spot with a small number of cameras (imaging devices). An object of the present invention is to provide a moving object detection / determination device that can obtain data.

[0007]

In order to achieve the above object, a moving object detection / judgment apparatus according to the present invention includes a detection camera device for detecting a moving object capable of capturing an image of a wide area, and an output from the detection camera device. Based on the detected signal
Based on image data captured by the determination camera device capable of capturing an image of a narrow range region that is captured in a state sufficient to determine whether the detected moving object is an object, and image data captured by the determination camera device, And a determination processing unit that determines whether or not the moving object is the target object. The detection camera device includes at least a detection camera body facing in a predetermined direction, a substantially conical reflecting mirror arranged at a predetermined position in front of the detection camera body, and image data captured by the detection camera body. On the other hand, it comprises a moving object extracting means for performing a predetermined image processing to extract a moving object in the image data, and a position information extracting means for extracting at least position information of the extracted moving object, and further, the determination The camera device is composed of a judgment camera body and means for controlling the attitude of the judgment camera body.

[0008]

When a moving object invades the monitoring area, the detection camera device detects it, the position information extracting means obtains an approximate position of the moving object, and the position information is sent to the corresponding determination camera device. send. The determination camera device determines the posture and magnification of the determination camera body based on the given information to capture an image of a moving object, and sends the captured image data to the determination processing means, where the moving object is an object or not. Make a decision.

By the way, while the determination camera main body is operated to capture an image of a moving object and a predetermined determination process is being performed, the detection camera device always captures an image of its own monitoring area to determine whether there is a new moving object. Detection is taking place. Therefore, there is no blind spot that occurs when the determination camera device zooms up to capture an image of a moving object for the determination process in the related art, and monitoring omission does not occur.

Since the detection camera device only needs to determine the presence or absence of a moving object, the moving object imaged may be small, the area that can be monitored by one device is wide, and there is almost no blind spot in all the monitoring areas. It also reduces the number of units required to install in. Similarly, the determination camera device does not always (frequently) operate, and even if a plurality of moving objects are present at the same time, the determination camera device can sequentially perform processing one by one (moving objects that have not undergone determination processing). Is supplemented by the detection camera device), the number of installed units can be reduced.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT A preferred embodiment of a moving object detection / judgment device according to the present invention will be described in detail below with reference to the accompanying drawings.
FIG. 1 shows a first embodiment of the present invention. As shown in the figure, it is determined whether or not the moving object is an object based on a detection camera device 10 for detecting a moving object capable of detecting a wide area and a detection signal output from the detection camera device 10. Determination camera device 20 capable of turning and moving and capturing an image of the moving object in a zoomed-up state
A determination processing unit 30 that performs determination processing based on image data captured by the determination camera device 20, and a determination processing unit 3
0, and an information processing unit 40 for grasping and managing the state of the moving object on the surveillance area plane based on the detection information sent from the detection camera device 10.

Next, each part will be described in detail. In the detection camera device 10, a detection camera 11 is installed upward on a pole (not shown), and a conical reflecting mirror 12 is installed above it. The installation height of the detection camera 11 and the cone angle of the reflecting mirror 12 are set according to the size of the monitoring area of the detection camera 11. And, in this conical reflecting mirror 12,
The periphery of the pole (camera) installation position is projected in the central part, and the farther the part is projected as it approaches the outer periphery. As a result, the detection camera 11 can obtain a wide monitoring area in the 360-degree direction.

Further, the image data picked up by the detection camera 11 is sent to the moving object extraction unit 13, and the moving object existing in the image data is extracted there. The specific configuration of the moving object extraction unit 13 is as shown in FIG. That is, similar to normal image processing, the image capturing unit 13a performs predetermined preprocessing and A / D conversion, and the brightness difference processing unit 13b stores the background image (reference image) of the monitoring area stored in the memory 13c in advance. ) And the difference image (moving object) is extracted. The difference image is sent to the binarization unit 13d, binarized by a predetermined threshold value, and the labeling unit 13e further performs a labeling process (numbering) on the binarized moving object. Attach).

Information on the moving object detected and labeled in this way is used as the position information extraction unit 1 in the next stage.
4, the information such as the position and area of the moving object, the moving direction and the moving speed of the moving object is extracted for each label. Here, the position information (particularly the direction) determines the direction in which the determination camera device 20 captures an image, and the distance L and the area in the position information determine the approximate magnification for lens control in the determination camera device 20, and further movement. The direction and the moving speed are used to eliminate the time lag from the time when the detection camera device 10 detects a moving object by the feedforward control to the time when the determination camera device 20 actually acquires an image.

As shown in FIG. 3, the position information is calculated by using the installation position (center of the reflecting mirror 12) O of the detection camera as a reference (origin) and the coordinate value (x, x) of the moving object I on the screen. y) is calculated. From the coordinate values, the distance L from the detected camera installation position (Xn, Yn) on the surveillance area plane to the moving object I (the distance from the reference on the camera image,
By obtaining the actual distance on the monitoring area in advance and storing it in a table, the distance can be easily obtained by referring to the table. Further, it may be calculated each time from the distance from the reference and the cone angle on the camera image) and the direction θ thereof.

Then, as shown in FIG. 4, the detection camera 11
Since the installation position (Xn, Yn) in the monitoring area is known in advance, the existing position (Xp, Yn) of the moving object I in the monitoring area is substituted by substituting the obtained L and θ into the following equation.
p) can be determined.

Xp = Xn + Lcos θ Yp = Yn + Lsin θ Further, the area information obtained by the detection camera 11 is obtained by counting the number of pixels of the detected moving object, and the moving direction and the moving speed are determined for a predetermined time. The position information of the moving object in the previous frame is stored and obtained from the position information and the current position information. Then, the information thus obtained is transmitted to the corresponding information processing unit 40 via the data transmission unit 15 by wired or wireless communication.

The determination camera device 20 includes a determination camera 21.
And a camera control unit 22 that controls the posture of the determination camera 21, and the detection camera 21 is mounted on the basis of the position information of the moving object sent from the detection camera device 10 via the information processing unit 40. The turning angle and the depression angle of the mounted platform are determined, and they are directed in a predetermined direction, and the magnification is determined from the area information.

That is, as shown in FIG. 4, since the position coordinates (Xj, Yj) of the determination camera 21 are known, the position coordinates (Xp, Yj) of the moving object I are used as a reference. Yp), and further, based on the area, moving speed, and moving direction of the moving object I, the moving object I having a predetermined size suitable for the judgment is located in the approximate center of the image captured by the judgment camera 21. The predictive control is performed so that the image is captured.

Further, the determination processing unit 30 has a moving object extraction unit 31 on its input side, receives image data captured by the determination camera 21, and extracts moving objects existing in the image data. ing. This moving object extraction unit 3
As shown in FIG. 5, the concrete configuration of No. 1 is that the video capturing unit 31a performs predetermined preprocessing and A / D conversion in the same manner as in normal image processing, and the brightness difference processing unit 31b preliminarily performs it. The difference image (moving object) is extracted by performing difference processing with the background image (reference video) of the monitoring area stored in the memory 31c. The difference image is sent to the binarization unit 31d, binarized by a predetermined threshold value, and the labeling unit 31e further performs the labeling process on the binarized moving object. .

The output of the moving object extracting section 31 is sent to the object determining section 32 and the position information extracting section 33. Then, the target object selection unit 32 performs feature amount extraction on the extracted moving object, determines whether the target object is a target object, and sends the determination result to the information processing unit 40 via the data transmission unit 34. It has become. Since various known methods can be used as a specific method of this determination process, detailed description thereof will be omitted.

Further, the position information extraction unit 33, like the position information extraction unit 14 of the detection camera device 10 described above, obtains the existing position and area of the moving object on the screen, as well as the moving direction and speed. ing. Then, when the moving object is not in the center or is not large enough for the determination, various information is sent to the camera control unit 22 via the data transmission unit 34 to obtain a determinable image. The posture of the determination camera 21 is controlled again. That is,
It is possible to perform follow-up control so that a moving object can be imaged at an appropriate size and position based on the image data obtained by itself.

Further, the information processing unit 40 uses the position information of the moving object obtained from the detection camera device 10 and the determination information of the moving object obtained from the determination processing unit 30 in the object mark unit 41 to detect both of them. Correlation is performed, an identification mark is added as a result of the determination of the moving object, and the position on the plan view of the moving object that changes from moment to moment is updated by the plan view position updating unit 42 so that a plurality of objects in the monitoring area can be displayed. The situation of moving objects is known. That is, in this example,
For the moving object that has once been subjected to the judgment processing, the judgment processing is not performed thereafter in principle, and the movement locus of the moving object is tracked by using the detection camera device 10 and the position information of the moving object is updated by the plan view position updating unit 42, thereby making the judgment. The number of times of processing is reduced so that the determination processing can be efficiently performed on a moving object existing in the monitoring area. Further, as described above, the predetermined information sent from the detection camera 10 is sent to the determination camera.

Next, the operation of the above embodiment will be described. First, as a premise, as shown in FIG. 6, there is almost no blind spot in all of the monitoring areas (strictly speaking, the area immediately below the detection camera 11 cannot be imaged, but a moving object does not suddenly appear at such a position, so there is no blind spot in use. It is assumed that 10N detection camera devices are installed at a predetermined position and M determination camera devices 20 are also installed at appropriate positions in order to capture an image (no problem). When a moving object exists in the monitoring area of any one of the detection camera devices 10, the detection camera device 10 detects it and obtains the approximate position and size of the moving object, so that the moving object can be captured best. Predetermined information such as its position information is sent to the determination camera device 20. Then, the determination camera device 20 determines the posture and magnification of the determination camera 21 based on the given information to capture an image of the moving object, and sends the captured image data to the determination processing unit 30 where the moving object is the target object. It is determined whether or not.

By the way, while the determination camera 21 is operated to capture an image of a moving object and a predetermined determination process is being performed, the detection camera device 10 constantly captures an image of the inside of the monitoring area,
Since the presence of a new moving object is monitored, there is no blind spot that occurs when zooming up to capture an image of a moving object as in the related art, and monitoring omission does not occur.

Further, since the detection camera device only needs to determine the presence / absence of a moving object, the imaged moving object may be small, so that the area that can be monitored by one unit is wide,
It also reduces the number of units required to install in all surveillance areas with almost no blind spots.

Furthermore, it is unlikely that a moving object actually exists in the monitoring area, and even if another moving object appears during the turning motion in order to capture an image of the moving object with the determination camera 21, other Since the position of the moving object is captured by the detection camera device 10, the moving object being processed is first imaged, and then the next moving object is imaged by the determination camera device 20 and the determination processing unit 30. Since the determination process may be performed, the number of determination camera devices 20 installed can be reduced. Furthermore, the determination camera device 20 and the determination processing unit 3
0 does not have a one-to-one correspondence, and one determination processing unit may process image data captured by a plurality of determination camera devices 20.

Next, a case where a moving object as a target object is a vehicle having a large moving speed and a large size and a human being having a small moving speed and a small size (a case where a plurality of moving objects are simultaneously present in the monitoring area) will be described. To do.

First, as a detection example of a moving object, it is roughly classified into a case where one moving object exists in the surveillance area of one detection camera as shown in FIG. 6 and a case where one moving object exists as shown in FIG.
Two moving objects (human M and vehicle S) may exist within the same monitoring area of the detection camera.

As shown in FIG. 6, the moving objects M and S exist in different areas of the detection camera device 10, and in addition, in order to obtain the image data for the determination processing for each moving object M and S. If the determination camera devices 20 used for the same are also different, the determination process is performed by performing parallel processing independently in each of the camera devices 10, 20 and the determination processing unit 30.

On the other hand, as shown in FIG. 7, when there are a plurality of moving objects M and S in the monitoring area of one detection camera device 10 (normally, the determination is made to image both moving objects). In principle, the determination camera device 20 cannot simultaneously image a plurality of moving objects (for example, a plurality of moving objects of the same type are close to each other, and the determination camera device 20 is sufficient for determination in the image data of the position). Therefore, the determination camera device 20 sequentially captures images and performs the determination process.

At this time, the order of the judgment is to carry out the judgment in descending order of importance, such as "Which is more likely to approach a dangerous area or the like (need to make an urgent judgment)". Become. The determination can be made based on information such as the moving direction, speed, and distance to the dangerous area. It should be noted that it is also possible to make a determination from an arbitrary moving object, for example, in the order of label numbers, without prioritizing according to the degree of importance.

Incidentally, such a situation is not limited to the case where a plurality of moving objects exist in the same surveillance area as shown in FIG. 7, and even if the surveillance areas of different detection cameras have the same judgment camera. It also applies when the device needs to be used.

FIG. 8A shows a second embodiment of the present invention. As shown in the figure, in the present embodiment, the shape of the reflecting mirror 12 'used in the detection camera device is changed to a curved surface in which the cone angle gradually decreases from the tip (center side) toward the bottom side. did. That is, in the case of the conical reflecting mirror of the first embodiment, a moving object passing near the position where the camera is installed appears large, and the cone angle is extremely large so as to extend the radius that can be monitored by one detection camera. If it is set to a small value, it tends to be difficult to capture an image near the installation position of the camera as shown in FIG. This is also clear from the relationship between the near-end distance and the far-end distance of the imageable monitoring area with respect to the cone angle shown in FIG. In view of this, in this example, the imageable region is enlarged by forming the shape as shown in FIG.

With this configuration, the area (monitoring area) that can be monitored by one detection camera can be expanded, so that the number of installed cameras can be further reduced. The rest of the configuration and operation are similar to those of the above-described first embodiment, so description thereof will be omitted.

FIG. 10 shows a third embodiment of the present invention. As shown in the figure, in this example, unlike the above-described examples, the detection camera is installed in the horizontal direction. That is, when there is a restricted space A as shown in the figure, the reflecting mirrors 12 ″ a, b (two positions in this example) are provided at a constant distance from the boundary wall surface H of the restricted space A. At this time, the reflecting mirrors 12 ″ aa, b and the boundary wall surface H on the bottom face each other and are arranged parallel to each other. Then, the detection cameras 11 "a, b" are arranged opposite to the tip ends of the reflecting mirrors 12 "a, b. further,
The cone angle δ of the reflecting mirrors 12 ″ a, b is set as follows, where ψ is the vertical angle of view of the detection camera 11 ″. That is, it is necessary to set an angle when the peripheral edges of the bottom surfaces of the reflecting mirrors 12 ″ a, b are imaged from the optical axis of the camera, but in this example, as shown by hatching in FIGS. Since the reflecting mirrors 12 ″ a and 12b are adjusted to be imaged in the entire screen imaged by the detection camera 11, the required angle is the vertical angle of view.

Δ = (1/2) (π + ψ) By setting the cone angle in this way, the detection camera 1
On the boundary of the reflecting mirrors 12 "a, b picked up by 1" a, b, a danger warning is issued which is an extension of the bottom surface of the reflecting mirrors 12 "a, b indicated by the symbol K in FIG. Boundary of information (boundary wall surface 11
And image data on a parallel surface with a constant interval).

Therefore, for example, when a moving object (human) M is present at the position shown in FIG. 9A, the screen imaged by the detection camera 11 "a at that time is as shown in FIG. The moving object M exists at any position, and when the human moves so as to approach the boundary K of the danger warning, as indicated by the arrow in FIG.
The object M in the video imaged at 1 ″ a moves so as to approach the outer peripheral edge of the surveillance area, as indicated by the arrow in FIG. It is necessary to issue an alarm, etc., or to notify the supervisor.

Therefore, in this example, the detection camera 11 "
When a moving object is detected in the monitoring area of a (shown by hatching in FIGS. (B) and (C)), its position information and the like are determined on the camera device side (not shown) as in the above embodiments. The moving object is imaged by the determination camera device based on the information, and the determination processing unit determines whether or not the target object needs to issue an alarm, and if necessary, then The moving condition of the moving object imaged by the detection camera 11 ″ a is monitored, and when the moving object approaches the boundary K, a predetermined process such as an alarm is performed.

By the way, since the normally closed space is partitioned by a plurality of wall surfaces H as shown in FIG. 1A, the detection camera devices 10 "a and 10b are installed in front of the respective wall surfaces as shown in the drawing. (Determining camera device can be shared in some cases.) Then, for example, a person M shown in FIG. Considering the case of moving to the side and further bending along the wall surface H, the image is initially captured by the detection camera 11 ″ a, but then the extension surface (not the boundary K) of the bottom surface of the reflecting mirror 12 ″ is detected. If it crosses, it will come out from the outer peripheral edge of the monitoring area of the detection camera 11 ″ a (which should be an alarm target in the original case). However, in such a case, at the same time, another detection camera 11 ″ b. Superintendent of The moving object detected by the detection camera 11 "b is detected by comparing the position that has entered the area and that has entered and exited the monitoring area of both detection cameras 11" a, b. 1
Since it is possible to determine whether or not the moving object comes out from 1 ″ a, it is possible to accurately determine whether or not there is a need to issue an alarm.

In this example, it is not always necessary to issue an alarm or the like, and the point is that a moving object approaching the boundary K can be detected and it can be determined whether or not it is an object. Further, as the reflecting mirror 12 ″, any of the ones shown in the above-mentioned first and second embodiments can be used.

FIG. 11 shows a fourth embodiment of the present invention. As shown in the figure, in this example, unlike the above-described embodiments, two sets of the detection camera and the reflecting mirror are provided, and the existence position of the moving object I in the three-dimensional space can be detected. That is, the first detection camera 11a and the reflecting mirror 12a are installed with the camera facing upward (the optical axis of the camera is aligned with the axis of the reflecting mirror) as in the first and second embodiments. On the other hand, the second detection camera 11b is arranged below the first detection camera 11a so that the optical axes thereof coincide with each other. Then, the second reflecting mirror 12b is arranged below the second detection camera 11b. Thereby, the first and second detection cameras 11a and 11b and the first and second reflecting mirrors 12 are provided.
The a and 12b are arranged on the same axis extending vertically from the ground.

Then, assuming that the vertical line extending upward as shown in the drawing represents the spatial coordinates of the object to be detected and the Cartesian coordinate system has the X and Y axes in the horizontal direction orthogonal thereto, the moving object I is detected by both detection cameras. When existing in the monitoring areas 111a and 11b, the directions θ on the X and Y coordinate planes obtained by imaging with both cameras are the same in principle. Further, the distance L from the center of the image captured by the camera to the moving object M
1, L2 are associated with the directions φ1 and φ2 in which the moving object is located. Therefore, on the plane including the optical axes of the detection cameras 11a and 11b and the position of the moving object M, a triangle is formed from the angles φ1 and φ2 formed by the extended surfaces of the bottom surfaces of the respective reflecting mirrors 12a and 12b and the position of the moving object M. Moving object I
The spatial position coordinates (Xm, Ym, Zm) of can be obtained.

The specific calculation method is as shown in the following formula, and the calculation is obtained by the position information extraction unit (not shown) connected to each of the cameras 11a and 11b.

If the detection cameras 11a and 11b and the reflecting mirrors 12a and 12b are installed on the optical axis of the camera as shown in FIG. 11, the optical axis of the camera need not be perpendicular to the ground. It may be horizontal. That is, the orientation of installation is not limited in three-dimensional position detection.

[0046]

[Equation 1] Then, the position coordinate information and the like thus obtained are sent to the determination camera device side (not shown) in the same manner as in each of the above-described embodiments, and based on the information, the determination camera device images the moving object and makes a determination. The processing device determines whether or not the target object needs to issue an alarm or the like. That is, in this example, when the target object to be detected moves in the air or there is a height difference in the monitoring area and it is necessary to detect a moving object moving on a hill or the like, the position is accurately specified. This can be done, and the determination camera can surely supplement. In addition,
The reflector used also in this example may have a shape as shown in the second embodiment.

[0047]

As described above, in the moving object detection / judgment apparatus according to the present invention, the moving object is detected based on the image data picked up by the detection camera device capable of picking up a wide area. If so, the acquisition of the image data necessary for the determination can be obtained by imaging with the determination camera device.Therefore, the detection camera device can monitor the presence or absence of a new moving object during the determination process. It never happens.

Since the detection camera device only has to determine the presence or absence of a moving object, the moving object imaged can be small, so that the area that can be monitored by one unit is wide, and there is almost no blind spot in all the monitoring areas. It is possible to reduce the number of units required for installation in. Further, the determination camera device does not always (frequently) operate, and even if there are a plurality of moving objects at the same time, it is possible to perform processing one by one (moving objects that have not undergone determination processing are Since it is supplemented by the detection camera device), the number of installed units can be reduced. In particular, a determination camera that makes a turning motion or the like is expensive because it requires a drive mechanism (including control). Therefore, reducing the number of determination camera devices to be installed can reduce the cost of the entire system. .

Further, when the area, the moving direction and the moving speed are extracted as the information about the moving object detected by the detecting camera device, the attitude control of the judging camera body in the judging camera device, etc. Can be performed more accurately, and the image data required for the determination can be obtained accurately and at high speed.

Further, when the installation direction of the detection camera body or the like is horizontal, it is possible to easily detect the presence or absence of a moving object approaching a predetermined place such as a restricted space.

Further, when two sets of the detection camera body and the reflecting mirror are provided, the position information of the moving object in the three-dimensional space can be obtained, so that the space can be moved or the position can be changed to a different height. It is possible to accurately detect an existing moving object.

[Brief description of drawings]

FIG. 1 is a diagram showing a first embodiment of a moving object detection / determination device according to the present invention.

FIG. 2 is a diagram showing an internal configuration of a detection camera device.

FIG. 3 is a diagram illustrating an operation of a position information extraction unit of the detection camera device.

FIG. 4 is a diagram illustrating an operation of a position information extraction unit of the detection camera device.

FIG. 5 is a diagram illustrating an internal configuration of a determination processing unit.

FIG. 6 is a diagram illustrating the operation of the present embodiment.

FIG. 7 is a diagram for explaining the operation of the present embodiment.

FIG. 8A is a diagram showing a main part of a second embodiment of the moving object detection / determination apparatus according to the present invention. (B) is a figure which shows the comparative example.

FIG. 9 is a graph illustrating the operating principle.

FIG. 10A is a diagram showing a main part of a third embodiment of the moving object detection / determination apparatus according to the present invention. (B) is a figure which shows an example of the image data imaged by the detection camera 11 "a. (C) is a figure which shows an example of image data imaged by the detection camera 11" b.

FIG. 11 is a diagram showing a main part of a fourth embodiment of the moving object detection / determination device according to the present invention.

[Explanation of symbols]

10, 10 "a, 10" b detection camera device 11, 11a, 11b, 11 "a, 11" b detection camera 12, 12a, 12b, 12 ', 12 "a, 12" b
Reflecting mirror 13 Moving body extraction unit 14 Position information extraction unit 20 Judgment camera device 30 Judgment processing unit 40 Information processing unit

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁶ Identification number Internal reference number FI Technical indication location H04N 7/18 E (72) Inventor Yoshiaki Yamada 1-4-10 Irifune, Chuo-ku, Tokyo TEPCO Riki Co., Ltd. System Research Center

Claims (7)

[Claims] 1. A detection camera device for detecting a moving object capable of capturing an image of a wide area, and whether or not the detected moving object is an object based on a detection signal output from the detection camera device. Based on the determination camera device capable of capturing an image of a narrow range area to be captured in a sufficient state for determination, and the image data captured by the determination camera device,
A detection processing unit that determines whether or not the moving object is the target object, wherein the detection camera device is disposed at a predetermined position in front of the detection camera body facing at least a predetermined direction and the detection camera body. A substantially conical reflecting mirror, and moving object extraction means for performing a predetermined image processing on the image data captured by the detection camera body to extract a moving object in the image data,
A position information extraction unit that extracts at least position information of the extracted moving object, and the determination camera device includes a determination camera body and a unit that controls the attitude of the determination camera body. A moving object detection / determination device characterized. 2. The position information extracting means has a function of calculating an area of a moving object from image data taken by the detection camera body, and takes the calculated area into consideration in addition to the position information. The moving object detection / determination device according to claim 1, wherein the magnification of the determination camera can be set. 3. The position information extracting means has a function of calculating a moving direction and a moving speed of a moving object from image data taken by the detection camera body, and the calculated movement in addition to the position information. 3. The moving object detection / determination device according to claim 1, wherein the posture of the determination camera is determined in consideration of the direction and the moving speed. 4. The moving object detection determination according to claim 1, wherein the detection camera and the reflecting mirror which form the detection camera device are arranged in a vertical direction. apparatus. 5. The detection camera and the reflector, which constitute the detection camera device, are arranged such that an optical axis of the camera is a normal line with respect to a monitoring boundary surface. The moving object detection and determination device according to any one of items 1 to 3. 6. Two sets of the detection camera and the reflection mirror which constitute the detection camera device are used, the detection camera and the reflection mirror are arranged on the same axis in the optical axis direction of the camera, and the first camera is arranged in order. A reflecting mirror, a first detecting camera, a second detecting camera, and a second reflecting mirror are arranged, and further, the position information extracting means is a moving object in a three-dimensional space based on image data obtained from the both detecting cameras. The moving object detection determination device according to claim 1, wherein the presence position of the moving object is extracted. 7. The moving object detection / determination device according to claim 1, wherein the reflector has a curved tip end shape.