KR100420620B1 - Object-based digital video recording system) - Google Patents

Abstract

The present invention extracts a background frame from an object image photographed from a camera installed to monitor an arbitrary restricted area, such as a bank and a mall building, and sets a background frame from the object image, and compares the object image with the background frame. After extracting the image, the object is to record and store the compressed image of the external contour and the internal image of the subject through a specific compression algorithm.

As a means for achieving the present invention, a pre-processing frame memory for storing the object image photographed through the surveillance camera (C); Background setting means (20) for setting a background frame of the object image photographed through the surveillance camera (C); A background frame memory 30 which stores the background frame set by the background setting means 20 and transmits the background frame to the monitoring unit 90; Image extracting means (40) for extracting only the subject image from the images stored in the preprocessing frame memory (10) and the background frame memory (30); A contour contrast unit 70 for dividing the outer contour and the inner image of the subject image extracted by the image extracting means 40; And a code conversion unit 80 encoding the image segmented by the contour contrast unit 70 using a specific compression algorithm and transmitting the encoded image to the monitoring unit 90.

Description

Object-based digital video recording system

The present invention relates to an object-based video surveillance system, and more particularly, to an object image obtained by photographing a specific area from a surveillance camera installed to monitor a restricted area of a specific place frequently visited by several people, such as a bank or a mall building. The present invention relates to a surveillance system for setting a background screen from the object image, comparing only the object image with the background screen, extracting only a subject image, and then encoding and storing the same by a specific compression algorithm.

In general, the video surveillance system is installed to install a surveillance camera in a specific place, such as a bank or a mall building, to shoot a specific limited area, the surveillance camera is to transmit the image obtained by photographing a specific area to the monitoring unit output device such as a monitor It outputs to monitor specific area.

In addition, in a specific place where a plurality of surveillance cameras are set, playback may be performed by multiple monitors to output images captured by each surveillance camera, or by dividing and reproducing the display area to output multiple images on one monitor. For example, monitoring a limited area of a specific place.

The monitoring unit is configured to record an image transmitted from each camera through an image recording means such as a VTR or a VCR. The respective camera operating power is controlled by the manager or the monitor of the monitoring unit.

As described above, the manager or the monitor visually monitors the image displayed by the monitor, and the captured image is stored and stored in an image recording medium such as a video tape of an image recording means such as a VTR or a VCR.

However, since the surveillance system using the analog system is recorded continuously by the video tape of the video recording means, etc., the administrator or the supervisor must replace the video tape at any time unit at any time and at the same time Have a problem of having to record the recorded date and time for each field, and in order to find the necessary scene from the recorded image, the administrator or the monitor must find and search the corresponding tape according to the recorded date and time. I have a problem.

In view of the uncomfortable problems inherent in the analog method due to the recent development of the technology, the digital method monitors a specific area by a monitoring unit and captures an image captured by a digital camera or a camcorder. The recorded image is stored in the computer hardware of the monitoring unit.

Here, the image stored in the computer is a method of storing each screen (frame) consecutive from the image taken for a certain time or by compressing each of the screen and each frame consecutive for the arbitrary time A method of compressing and storing only video portions that are changed between neighboring screens is used.

However, when the necessary frames are compressed and stored in the image photographed for a certain time as described above, since the background frame of each frame is compressed and stored in the image photographed for a certain time, unnecessary background frames are stored and unnecessary. Background frame is repeatedly stored.

In addition, since the continuous shooting image for a certain time is stored in a hardware having a limited capacity, the surveillance system has a problem that a large amount of hardware is required to store a large amount of image.

In order to solve the above problems, the present invention is set to the background frame by extracting the background frame from the object image taken from the camera installed to monitor any restricted area, such as banks and mall buildings, The object of the present invention is to extract the subject image by comparing the object image with the background frame, and to record and store the compressed image of the subject by using a specific compression algorithm.

1 is a block diagram showing the configuration of an object-based video surveillance system according to the present invention.

2 is a reference diagram for extracting only a subject image from an image photographed through a surveillance camera of an object-based image surveillance system according to the present invention.

* Description of the symbols for the main parts of the drawings *

10: frame memory for preprocessing 20: background setting means

30: background frame memory 40: image extraction means

50: image correction means 60: switching unit

70: outline contrast unit 80: code conversion unit

90: monitoring unit 91: monitor

93: video recording unit C: surveillance camera

In order to achieve the present invention, by monitoring the specific area by transmitting the image taken by the surveillance camera installed in order to monitor a specific restricted area, such as a bank or a shopping mall building to an output device such as a monitor to monitor the specific area at the same time A surveillance system for recording and storing an image of a specific region photographed through an image recording means, the surveillance system comprising: a frame memory for preprocessing an object image photographed through the surveillance camera; Background setting means for setting a background frame of the object image photographed through the surveillance camera; A background frame memory which stores the background frame set by the background setting means and transmits the background frame to the monitoring unit; Image extraction means for extracting only a subject image from the images stored in the preprocessing frame memory and the background frame memory; A contour contrast unit for dividing the outer contour and the inner image of the subject image extracted by the image extracting means; And a code conversion unit encoding the image divided by the contour contrast unit by a specific compression algorithm and transmitting the encoded image to the monitoring unit.

When the surveillance camera has a panning function, the surveillance camera is set to be converted into a predetermined angle at a predetermined time interval within a preset movement range, and includes information converted to a predetermined angle according to the predetermined time interval. .

On the other hand, the image output to the contour contrast unit by the image correction means for comparing the threshold value of the image converted to the predetermined angle according to a predetermined time interval and the absolute value in the subject image extracted from the image extraction means Characterized in that further comprises a switching unit to control the.

Hereinafter, a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram showing the configuration of an object-based video surveillance system using a camera having a panning function according to the present invention.

2 is a reference diagram for extracting only a subject image from an image photographed by a surveillance camera of a surveillance system through object-based image compression reproduction according to the present invention.

As shown in FIG. 1 and FIG. 2, the surveillance camera C installed to monitor a specific limited area such as a bank or a mall building may use a camera having a fixed focus or a panning function. When (C) is a fixed focus camera, the output of the camera is input to the preprocessing frame memory 10 and the background setting means 20.

In addition, in the case of the camera C having a panning function, the camera C is set to be converted to a predetermined angle (for example, 1 °) at a predetermined time interval within a preset movement range, and is converted to a predetermined angle according to the predetermined time interval. The image including the information is input to the preprocessing frame memory 10, the background setting means 20, and the background frame memory 30, respectively (the present invention described below uses a camera having a panning function).

The image captured by the surveillance camera C is input to the preprocessing frame memory 10 and the background setting means 20. The preprocessing frame memory 10 is a pre-processing frame memory 10 is obtained by photographing an arbitrary region of the image taken by the camera set to be converted to a predetermined angle at a predetermined time interval according to a predetermined time interval Temporarily stored and then output to the image extraction means (40).

The background setting means 20 is a means for setting a background frame in the image photographed from the camera C. The background setting means 20 sets a predetermined period which is converted to a predetermined angle according to a predetermined time unit, and is continuously photographed at the predetermined period. An area before the subject region starts in the frame and a screen having no difference in chromaticity and brightness for each pixel of the image screen from the time when the subject region ends are defined and divided as a background screen.

The background frame set by the background setting means 20 is stored in the background frame memory 30, and the stored background frame is transmitted to the monitoring unit 90 by encoding the texture and stored in the image recording means 93.

The preprocessing frame memory 10 and the background frame memory 30 are preferably constituted by a memory having an R / W (read / white) function.

In addition, image extracting means 40 for matching images having the same time and angle in images stored in the preprocessing frame memory 10 and the background frame memory 30, and extracting only a subject image from the matched image. ) Extracts only the subject image according to the chromaticity and brightness difference of the pixel area of a predetermined range (or unit; the image of the minimum unit that can detect the change in the RGB signal of the pixel of the image) and then to the image correction means 50 Enter it.

The image correction means 50 is a means for correcting the movement of the subject image and an error caused by the panning control means P, and the error according to the image of the subject moved to the left and right by the panning control means P. The absolute value of the subject image (difference image in the previous frame of the image to be compared) and the threshold for correcting Converting a gray level image into a binary image (black and white image) to separate an object of interest from a background], but when the threshold value is smaller than an absolute value, the contact of the switching unit 60 is converted to the image. The subject image extracted by the extracting means 40 is output to the outline contrast unit 70.

When the threshold value is larger than the absolute value, the contact of the switching unit 60 is converted to ground, thereby bypassing the subject image extracted by the image extracting means 40 to ground. Remove unnecessary images.

The contour contrast unit 70 separates the outer contour and the inner image of the subject image, and the separated outer contour and the inner image are input to the code converter 80.

The code conversion unit 80 compresses the external contour and the internal image of the subject image by a specific compression algorithm and transmits the same to the monitoring unit 90. The monitoring unit 90 includes a monitor 91 and an image recording unit 93 for monitoring an image of a specific region captured by a camera.

In addition, the image recording unit 93 receives and stores the background frame from the background frame memory 30 that sets the background frame (or screen) in the image of the specific region captured by the surveillance camera C.

The output of the subject image stored in the image recording means 93 outputs the background frame set by the background setting means 20 to convert the background image into a predetermined angle according to a predetermined time interval, and monitors the moving subject image. To print).

In the case of reproducing the subject image stored in the image recording means 93, first, the image set as the background frame is output, and then the subject image is synthesized, and only the subject image which is continuous for a certain time is output.

The code conversion unit 80 is a means for compressing the subject image. The code converter 80 compresses the outer contour and the inner image of the subject image into a texture by using a motion J-pack or JPEG.

As described above, the present invention automatically sets a background image from the images taken through a plurality of cameras installed to monitor a specific limited area such as a bank or a shopping mall building, and synthesizes the images. By extracting only the subject image by detecting the difference and making an outline along the surface of the subject image, the external contour is extracted and the compressed image encoded in the texture format is transmitted and stored as data, thereby eliminating unnecessary images. This is because the amount of data stored in the hardware is minimized.

In addition, it has the effect of enhancing the convenience of the administrator or supervisor who monitors certain restricted areas.

Claims (5)

The image captured by the surveillance camera installed in order to monitor a specific restricted area such as a bank or a mall building is transmitted to an output device such as a monitor for display. In the surveillance system for recording and keeping by video recording means, A pre-processing frame memory 10 for storing the object image photographed through the surveillance camera C; The background frame of the object image photographed by the surveillance camera C is set, and a predetermined period is converted to a predetermined angle according to an arbitrary time unit, and the subject region is started in the frames continuously photographed at the predetermined period. Background setting means (20) for setting a background frame by defining a screen having a chromaticity and brightness difference of each pixel of the image screen within a predetermined range from a time when the previous area and the subject area ends, as a background screen; A background frame memory 30 which stores the background frame set by the background setting means 20 and transmits the background frame to the monitoring unit 90; Image extracting means (40) for extracting only the subject image from the images stored in the preprocessing frame memory (10) and the background frame memory (30); A contour contrast unit 70 for dividing the outer contour and the inner image of the subject image extracted by the image extracting means 40; And And a code conversion unit (80) for encoding the image divided by the contour contrast unit (70) using a specific compression algorithm to transmit the image to the monitoring unit (90). delete The method of claim 1, wherein the absolute value of the image according to the movement of the subject image extracted from the image extracting means 40 and the threshold value are compared by the image correcting means 50. And a switching unit 60 for outputting the subject image to the contour contrast unit 70 when the threshold value is larger than the threshold value. The object-based video surveillance system according to claim 1, wherein the contour contrast unit (50) divides the external contour and the internal image of the subject image through the operation of creating an external contour along the surface of the subject image. According to claim 1, The monitor (91) of the monitoring unit 90 includes a decoding unit, the image of the moving image by the motion estimation or motion information, etc. during playback of the image compressed by the code conversion unit 80 An object-based video surveillance system characterized in that to correct the deviation.