JPH0635443A - Monitor device - Google Patents

Abstract

PURPOSE:To set the same monitor area even when an image pickup device is moved and also set a monitor area by only single operation even when plural devices are installed. CONSTITUTION:The monitor device has the image pickup device 5 which is so installed that an area to be monitored is included in the visual field, a three- dimensional model constituting means 13 which constitutes a three-dimensional scene model of a scene to be monitored and a three-dimensional body model of a body to be monitored, a three-dimensional monitor area generating means 14 which generates the three-dimensional monitor area by computer graphics from a previously set plane monitor area and the three-dimensional model of the body to be monitored, a monitor area setting means 15 which calculates how the three-dimensional monitor area is seen frame the image pickup device 5 by using previously set installation parameters of the image pickup device 5, sets that as a monitor area to an image, and segments a monitor area image from an image in the visual field picked up by the image pickup device 5, and a moving body detecting means 16 which detects a moving body from the monitor area image segmented by the means 15.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a monitoring device, and more particularly to a monitoring device for monitoring a specific area using image processing technology.

[0002]

2. Description of the Related Art In recent years, as a safety measure, a computer and image processing technology have been used to monitor illegal intruders in places where human entry is prohibited and to monitor illegal parking in prohibited parking areas. There is increasing interest in surveillance devices that provide automatic surveillance. These conventional devices are generally used for input images obtained from image capturing devices such as television cameras,
Image processing such as monitoring area setting processing, image conversion processing, and feature extraction processing is performed to detect humans and objects invading the monitoring area. FIG. 4 shows such a conventional monitoring device.
It shows an example of the configuration of (hereinafter, also simply referred to as “apparatus”). In such an apparatus, instead of monitoring the entire image input from the monitoring camera, attention may be paid to a part of the image and monitoring may be performed within the specific area as the monitoring area.
With current technology, it is difficult to automatically set the watched area of interest.Therefore, the watched area of interest can be manually set on the two-dimensional image observed from the monitor camera displayed on the display. By detecting a moving object in the area, an illegal intruder or the like is monitored. Regarding this, for example, "Real-time measurement of human flow by ITV image" by Y. Y. and T. Takaba.
(Journal of the Institute of Electronics, Information and Communication Engineers D, August 1983 issue) can be referred to.

More specifically, computer 1, display unit
The computer 1 comprises a (CRT) 2, an information input device 3, a memory 4, and an image pickup device 5 such as a surveillance camera.
An arithmetic processing unit (CPU) 11, an image input unit 12, a monitoring area setting unit 15, and a moving object detection unit 16 are provided, and these are connected by a data bus 17. In the device configured as described above, the grayscale image of the monitoring area input from the image pickup device 5 is converted into a digital frame image by the image input unit 12. This digital frame image is input to the monitoring area setting unit 15, where only the image of the specific area (monitoring area) of interest is cut out. That is, the digital frame image is displayed on the display unit 2 as shown in FIG. 5, and the operator of the monitoring device looks at the image and specifies the contour of the region of interest with the information input device 3 such as a light pen. . The monitoring area setting unit 15 of the computer 1 cuts out the monitoring area 54 based on the designated contour. The monitoring target image cut out by the above-described processing is input to the moving object detection unit 16. The moving object detection unit 16 performs a process of detecting a moving object from the image to be monitored. If a moving object is detected, the moving unit detection unit 16 displays the information on the display unit 2 and displays the current special mark and The monitoring target image is stored.

[0004]

However, in the above-mentioned conventional technique, the operator of the monitoring device sets the monitoring region on the two-dimensional image obtained from the image pickup device 5, and the monitoring region setting section 15 cuts it out. Since the processing is performed, the appearance changes when the image pickup device 5 moves,
The monitoring area must be set again. That is, there is a problem that the image pickup device 5 must always be fixed. Further, when monitoring is performed using a plurality of image capturing devices, there is also a problem that it is necessary to set a monitoring area for each image capturing device. If you want to monitor a wide area outdoors, or if you can't image all the area you want to monitor with one monitoring device even indoors, move the surveillance camera (pivot,
Up and down movement) and installation of multiple surveillance cameras are required, but the above-mentioned problem is a major obstacle to this. The present invention has been made in view of the above circumstances, and an object of the present invention is to solve the above-mentioned problems in the related art, to set the same monitoring area even when the image capturing apparatus is moved, and to capture an image. An object of the present invention is to provide a monitoring device in which the operator of the monitoring device can set the monitoring area with a single operation even when a plurality of devices are installed.

[0005]

SUMMARY OF THE INVENTION The above object of the present invention is a monitoring device for monitoring a specific area by using image processing technology, and image pickup is installed so that the monitoring target area is included in the visual field. A device, a three-dimensional model construction means for constructing a three-dimensional scene model of a scene to be monitored and a three-dimensional object model of an object to be monitored, and a computer graphic from a preset plane monitoring area and a three-dimensional model of the object to be monitored. By using a three-dimensional monitoring area generating means for generating a three-dimensional monitoring area by a monitor and a preset installation parameter of the image capturing apparatus, the appearance of the three-dimensional monitoring area when viewed from the image capturing apparatus is determined. Monitoring area setting means for calculating and setting it as a monitoring area on the image, and cutting out the monitoring area image from the image in the visual field picked up by the image pickup device; It is achieved by the monitoring device, characterized in that it comprises a moving object detecting means for detecting a moving object from the extracted surveillance area image by said means.

[0006]

In the surveillance device according to the present invention, the operator of the surveillance device constructs a three-dimensional (hereinafter also referred to as "3D") model of the scene to be monitored in advance, and sees the scene from directly above. 3D object of the monitoring target object such that the plane monitoring area, the 3D object model of the monitoring target object and the installation parameters of the image pickup device are set, and the computer draws a spiral from the contour of the planar monitoring area toward the inside. A simulated surveillance scene in which the model moves is generated by computer graphics (CG), the three-dimensional contour space of the movement trajectory is calculated as a 3D surveillance area, and the 3D scene model and the installation parameters of the image pickup device are used. Then, the appearance of the 3D monitoring area when viewed from the image pickup device is calculated,
It is set as a monitoring area on the image. As a result, even if the image pickup device is moved, the monitoring area is automatically set by the computer, and even when a plurality of image pickup devices are installed, the computer also sets the monitoring area of each image pickup device. Since this is performed automatically, the operator of the monitoring device can set the monitoring regions of all the image capturing devices substantially by only the first operation.

[0007]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 1 is a block diagram showing the configuration of a monitoring device according to an embodiment of the present invention. The monitoring device according to the present embodiment is provided with a computer 1, a display unit (CRT) for displaying information such as characters and images, a position of the monitor by inputting information such as characters and pointing on the CRT 2 of the display unit. An information input device 3 for inputting information, a memory 4 for storing information such as characters and images, and an image pickup device 5 constituted by a television camera or the like are provided. The computer 1 includes an arithmetic processing unit 11, an image input unit 12, and a 3D model construction unit 1.
A 3D monitoring area generation unit 14, a monitoring area setting unit 15, and a moving object detection unit 16 are provided, and these units are connected by a data bus 17. In this case, the display unit 2 is composed of, for example, a CRT display device, and the memory 4 is composed of a main storage device of the computer 1 and an auxiliary storage device such as a magnetic disk device and an optical disk device. The information input device 3 is composed of, for example, a keyboard, a light pen, a mouse and the like.
The image pickup device 5 is arranged so that its field of view faces the monitoring area.

Further, the image input section 12 includes an image pickup device 5
The 3D model construction unit 13 converts the grayscale image including the monitoring region input from the digital frame image into a digital frame image. The 3D model construction unit 13 is based on the information input by the operator. , Constructs a 3D object model of the monitored object. It should be noted that these models are manually constructed in advance by the operator of the monitoring device, and the method is described, for example, in Fujio Yamaguchi "Graphic Processing Engineering by Computer Display" (Nikkan Kogyo Shimbun, Showa 56). It is described in detail in (Annual). The 3D monitoring area generation unit 14 generates a 3D monitoring area by computer graphics from a preset planar monitoring area and a 3D model of an object to be monitored. The monitoring area setting unit 15 calculates the appearance of the 3D monitoring area when viewed from the monitoring camera using the installation parameters of the monitoring camera, sets it as the monitoring area on the image, and cuts it out. Further, the moving object detection unit 16 detects a moving object such as an illegal intruder in the monitoring area. Here, as a method of detecting a moving object, for example, the method described in "Motion Image Processing" by Saburo Tsuji (Information Processing Society of Japan, June 1979 issue) is used.

Next, the operation of the present embodiment configured as described above will be described in detail with reference to the flow chart shown in FIG. First, the operator of the monitoring device according to the present embodiment constructs a 3D scene model of a scene to be monitored and inputs it from the information input device 3. The input 3D scene model is stored in the memory of the computer 1 (step S01). Here, the construction of the 3D scene model of the scene to be monitored by the operator is performed by using a technique such as CAD based on the plan view and the 3D data, as shown in the above-mentioned technique. However, it is also possible to apply an automatic surface shape acquisition technique using a laser beam to this. The monitoring device, as shown in FIG.
An image when the scene model is viewed from directly above is displayed on the display unit 2, and the operator looks at this display and designates the area to be monitored from the information input device 3 with a light pen or the like. As a result, the area is input to the 3D monitoring area generation unit 14 as the plane monitoring area 41 (step S02). Further, the operator inputs the shape and size of the 3D object model 42 of the monitored object from the information input device 3 as shown in FIG. The input 3D object model 42 is input to the 3D monitoring area generation unit 14 (step S03).

In the 3D surveillance area generator 14, the input plane surveillance area 41 and the 3D object model 42 are processed by computer graphics as shown in FIG. 3B.
The 3D monitoring area 43 is generated. Specifically, the 3D object model 42 is designed such that a certain point on the contour of the plane monitoring area 41 is used as a starting point and a spiral is drawn inward along the contour line.
Is generated by computer graphics, and the three-dimensional contour space of the movement locus is calculated as the 3D monitoring area 43 (step S04). The 3D monitoring area 43 calculated here is input to the monitoring area setting unit 15. At the same time, the operator inputs the installation parameters such as the three-dimensional installation position of the surveillance camera, the direction of the viewpoint, and the focal length from the information input device 3 (step S05). The input installation parameters are input to the monitoring area setting unit 15, and the appearance on the image when the 3D monitoring area 43 is viewed from the monitoring camera is calculated (step S06). Regarding the calculation method of such appearance,
This is described in detail in the above-mentioned "Graphics Engineering by Computer Display". The calculated area is shown in FIG.
As shown in (c), it is set as a monitoring area 44 on the image obtained from the monitoring camera.

On the other hand, the image in the visual field at the current time is input to the monitoring area setting unit 15 and the image input unit 12. The image input unit 12 converts this image into a digital frame image. This digital frame image is input to the surveillance area setting unit 15, where a moving object such as an illegal intruder in the surveillance area is detected. The method of detecting a moving object is as described above. By repeating the moving object detection process from the image input process described above, the monitoring area is monitored at a predetermined time interval, and if a moving object enters the monitoring area during the monitoring, a suspicious object is detected. A signal indicating that the intruder has entered is issued, and the monitor is notified to the monitor by an illustration or alarm on the display unit 2. According to the above-described embodiment, the operator of the monitoring device previously builds a 3D scene model of the scene to be monitored, and the plane monitoring area when the scene is viewed from directly above, the 3D model of the object to be monitored, And, setting the installation parameters of the position of the surveillance camera, after that, the computer, a pseudo-scene to move the 3D model of the monitored object so as to draw a spiral inward from the contour of the planar surveillance region, It is generated by graphics, and a three-dimensional contour space of the movement locus is calculated as a 3D monitoring area.

Then, using the 3D scene model and the installation parameters of the surveillance camera, the appearance of the above-mentioned 3D surveillance area when viewed from the surveillance camera is calculated and set as the surveillance area on the image. Therefore, even if you move the surveillance camera, the computer automatically sets the same surveillance area, and even if you install multiple surveillance cameras, the computer automatically sets the surveillance area of each surveillance camera. Therefore, the operator can set the monitoring areas of all the monitoring cameras with only one initial operation. It is needless to say that the above embodiment shows one example of the present invention, and the present invention should not be limited to this. For example, the present invention can be used in a wide range, such as detection of illegal intruders, detection of illegally parked vehicles, counting of gate visitors, and the like, as a monitoring device for monitoring a specific area by using image processing. , Very effective.

[0013]

As described above in detail, according to the present invention, the same monitoring area is set even when the image pickup device is moved, and even when a plurality of image pickup devices are installed, the monitoring device is operated. This is a remarkable effect that a person can realize a monitoring device capable of setting a monitoring area with only one operation.

[0014]

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a monitoring device according to an embodiment of the present invention.

FIG. 2 is a flowchart showing the operation of the monitoring device according to the embodiment.

FIG. 3 is an explanatory diagram of a process of calculating a 3D monitoring area according to the embodiment.

FIG. 4 is a block diagram showing a configuration example of a conventional monitoring device.

FIG. 5 is a diagram illustrating an example of setting a monitoring area in a conventional monitoring device.

[Explanation of symbols]

1: computer, 11: arithmetic processing unit, 12: image input unit, 13: 3D model construction unit, 14: 3D monitoring region generation unit, 15: monitoring region setting unit, 16: moving object detection unit, 1
7: data bus, 2: display unit, 3: information input device, 4:
Memory, 41: plane monitoring area, 42: 3D object model,
43: 3D monitoring area, 44: monitoring area, 5: image pickup device.

Claims (1)

[Claims] 1. A monitoring device for monitoring a specific area by using image processing technology, comprising: an image pickup device installed so that the monitoring target area is included in a field of view; and a three-dimensional scene to be monitored. A three-dimensional model building means for building a three-dimensional object model of the scene model and the monitored object,
Using a preset plane monitoring region and a three-dimensional monitoring region generating means for generating a three-dimensional monitoring region by computer graphics from a three-dimensional model of a monitoring target object, using preset installation parameters of the image capturing device, Calculate the appearance of the three-dimensional monitoring area when viewed from the image capturing device, and set it as a monitoring area on the image,
Surveillance characterized by having a monitoring area setting means for cutting out a monitoring area image from an image in the visual field picked up by the image pickup device, and a moving object detecting means for detecting a moving object from the monitoring area image cut out by the means. apparatus.