KR100250889B1 - Motion detecting method using jpeg algorithm - Google Patents

Abstract

PURPOSE: A method of detecting motion is to detect the motion using a JPEG decompressing algorithm when detecting the motion in a CCTV system in which the JPEG is employed as a data compressing/decompressing algorithm, thereby exactly detecting the motion without a separate hardware for detecting the motion. CONSTITUTION: If power is applied to the CCTV, a main controlling portion(100) initializes each constructing portion. A video signal input from a video camera of a CCTV system is decoded by a decoder(20), and then stored at a memory portion(30) in an image unit. The CCTV employs JPEG as a data compressing/decompressing algorithm. Image data stored in the memory portion is read and compressed by a JPEG algorithm and then stored in a desired recording medium(80) provided at the CCTV system. The image data compressed and stored in the image unit is decompressed by the JPEG algorithm. Only the lowest frequency component is extracted from data generated in the decompressing process. The extracted lowest frequency component is compared with the lowest frequency component of a previous image.

Description

Motion Detection Method Using J-PEG Algorithm

The present invention relates to a CCTV system, and more particularly, to motion detection using a JPEG algorithm that performs accurate motion detection using a part of the JPEG decompression algorithm when detecting a motion of a CCTV system adopting JPEG as a data compression / extension algorithm. It is about a method.

Closed Circuit Tele-Vision (CCTV) is one of the security systems that is becoming more and more common in recent years. This closed-circuit monitoring system, or CCTV, prevents theft of valuables or money in places such as general houses, department stores, banks, exhibition halls, etc., and checks the operation status of the machinery in operation at the factory and the flow of processes in progress. As it is installed for the purpose, briefly described the configuration as follows. That is, the CCTV includes one or more video cameras for capturing a surveillance object, a monitor on which an image captured by the video camera is displayed, and a VCR for recording the image captured by the video camera for later playback.

Referring to the internal configuration of the CCTV as described above with reference to the system block diagram of Figure 1 as follows. As shown in FIG. 1, CCTV sequentially converts an analog video signal input through a video input unit 10 and a video input unit 10 into which video signals output from one or more video cameras are input, in units of fields or frames. Decoder that separates horizontal and vertical sync signals from sequentially converted video signals, separates them into luminance (Y) and color difference signals (BY, RY), and converts the separated luminance and color difference signals into digital signals. (20), a memory 30 for temporarily storing the luminance signal and the color difference signal digitally converted by the decoder 20, or the digital luminance signal and the color difference signal that are pre-compressed and re-extended by the decoder / decoder 70. Converting the digital luminance signal and the color difference signal read out at 30 into an analog signal and loading and resynthesizing the horizontal and vertical synchronizing signals coming through the memory control unit 60 to output them as video signals. Compressor 40, a video output unit 50 for outputting the video signal output from the encoder 40 in a split screen or a single screen on the monitor, and digital luminance and color difference signals read out from the memory 30 are compressed. To a recording / decoding unit 80 such as a video tape, or conversely, to a code / decoding unit 70 and a memory 30 that decompress and store a signal stored in the recording medium 80 and write the same to the memory 30. The subject on the surveillance screen by the video signal input from the memory control unit 60 and the video input unit 10 that control timing of writing and reading of the decoder 20, the encoder 40, and the encoder / decoder 70. The memory control unit 60, the encoder / decoder 70, the decoder based on the intrusion detection data input from the motion detection unit 90 to detect whether the outside of the intrusion into the monitoring place by detecting the movement of the 20, memory 30, encoder 40 to control Writing is configured to include a main control unit 100 for controlling to read out the data corresponding to a single divided screen and the screen is stored in each address of the memory 30.

In the encoding / decoding unit 70, which is one of the main components of the CCTV as described above, the video signal is compressed / extended, and the CCTV which can be implemented according to the present invention is a compression / extension algorithm such as JPEG (Joint Photographic Experts Group) It is required to adopt and use algorithms.

On the other hand, in the conventional CCTV, multi-channel video signals input from a plurality of video cameras are divided into 4 or 16 screens on one monitor screen at once, or only the screen for a desired surveillance area is displayed on the full screen of the monitor. It can also be displayed as a single screen. In addition, an intrusion detection sensor such as an optical sensor that monitors the intrusion of outsiders is installed in various places of the monitoring place, and when the intrusion of the outsider is detected through the intrusion detection sensor while displaying the monitoring area in the split screen, the corresponding monitoring area A function of automatically switching the display of the display to a single screen on the full screen of the monitor is added. In addition, even when an alarm does not occur due to an intrusion detection sensor error, by comparing and analyzing the pre-recorded images, the movement of the subject is detected by itself. There is also a function for automatic switching.

However, conventionally, when comparing the pre-recorded image and detecting the movement of the subject by itself, a signal corresponding to a specific part of the screen, which is output from the decoder 20, is used for the motion detection unit in the motion detection unit. In the memory control unit 60, a program for searching for a specific position on the two-dimensional space must be separately loaded. Therefore, in the related art, additional material costs are added as it is necessary to additionally install hardware components such as a motion detection memory for motion detection. In addition, since the motion detection algorithm itself is to compare only a specific part of the screen, there is a problem that frequent errors occur during motion detection.

The present invention has been made to solve the above problems, and does not have a separate motion detection hardware by using a part of the JPEG decompression algorithm when detecting a motion of a CCTV system adopting JPEG as a data compression / extension algorithm. It is an object of the present invention to provide a motion detection method using a JPEG algorithm capable of performing accurate motion detection without errors.

Motion detection method using the JPEG algorithm of the present invention for achieving this object is a CCTV system adopting JPEG as a data compression / extension algorithm, decoding the video signal input from the video camera of the CCTV system, one screen unit Sequentially converting the data into the memory; Reading data stored in the memory, compressing the data through the JPEG algorithm, and storing the data in a predetermined recording medium provided in the CCTV system; Decompressing and storing the data compressed by the screen unit through the JPEG algorithm, and extracting only the lowest frequency component among the data generated in the decompression process; And comparing and analyzing the extracted lowest frequency component with the lowest frequency component of the previous screen.

1 is a system block diagram of a general CCTV,

2 is a system block diagram of CCTV to which a motion detection method using the JPEG algorithm of the present invention is applied.

*** Explanation of symbols for the main parts of the drawing ***

10: video input, 20: decoder,

30: memory, 40: encoder,

50: video output unit, 60: memory control unit,

70: copy / decryption unit, 80: recording medium,

90: motion detection unit, 100: main control unit

Hereinafter, a motion detection method using the JPEG algorithm of the present invention will be described in detail with reference to the accompanying drawings.

2 is a system block diagram of CCTV to which a motion detection method using the JPEG algorithm of the present invention is applied. As shown in FIG. 2, unlike the conventional CCTV, the CCTV to which the method of the present invention is applied does not have a separate motion detection unit 90 and is recompressed and re-extended through the encoder / decoder 70. Is directly input to the main control unit 100. In addition, the main control unit 100 is equipped with a program for implementing a motion detection method using the JPEG algorithm of the present invention.

Hereinafter, the process of detecting the movement of the subject itself through the method of the present invention in the CCTV having the configuration as shown in FIG. First, when power is applied to the CCTV, the main control unit 100 initializes each component of the system. After that, the analog video signal photographed from the video camera of CCTV is sequentially converted to one screen unit through the decoder 20, and the horizontal and vertical synchronization signals are separated and separated, and the luminance (Y) and the color difference signals (BY, RY). ), And the separated luminance and chrominance signals are converted into digital signals. The separated and digitally converted video signal is stored in the memory 30 and then delivered to the encoder / decoder 70 for compression, and the video signal compressed by the JPEG algorithm in the encoder / decoder 70 is recorded on a recording medium. In addition to being stored in (80) it is input to the main control unit (100).

On the other hand, in order to implement the motion detection method using the JPEG algorithm of the present invention, the main control unit 100 is sequentially converted to one screen unit and then re-extended the JPEG compressed video signal using the JPEG decompression algorithm. The JPEG compression / extension process is described in detail as follows. First, the compressed data of one JPEG-compressed screen is composed of 64 frequency components including the lowest frequency DC component, which is the average value of the frequency within the screen. As divided into square blocks of 8 and 8 vertical pixels, 64 frequency components are displayed for one screen. After performing a DCT (Discrete Cosine Transform), which is a specific operation for implementing the JPEG algorithm, for each divided square pixel block, the DCT operation is divided into a specific number of quantization steps. do. According to such a series of JPEG compression processes, one screen is represented as compressed data consisting of predetermined frequency components distributed in a zigzag order from a low frequency term to a high frequency term, and the lowest frequency component is a direct current component. The total frequency average value. On the other hand, the main control unit 100 is generated through the Huffman decoding (Huffman decoding) process and the reverse DCT (Inverse Discrete Cosine Transform) process in the JPEG release, that is, re-extension the pre-compressed video signal through the above process Among the 64 frequency components, only the lowest frequency DC component and the end-of-block code (EOB), which means the end of 64 data, are stored.

Therefore, the main control unit 100 compares and analyzes the lowest frequency signal of each screen generated as the JPEG-compressed video signal is re-extended as described above, thereby subjecting the subject through the entire region instead of any specific region of each sequentially switched screen. Detect movements. When the motion is detected in the comparative analysis process, the main control unit 100 controls the video output unit 50 to display the corresponding surveillance area where the movement of the subject is detected as a single screen on the full screen of the monitor.

According to the motion detection method using the JPEG algorithm of the present invention as described above, a separate motion detection hardware by using a part of the JPEG decompression algorithm in the motion detection operation of the CCTV system adopting JPEG as a data compression / extension algorithm It is possible to perform accurate motion detection without error without having a.

Claims (1)

In CCTV system that adopts JPEG as data compression / extension algorithm, Decoding the video signal inputted from the video camera of the CCTV system, sequentially converting one screen unit and storing the image signal in a memory; Reading data stored in the memory, compressing the data through the JPEG algorithm, and storing the data in a predetermined recording medium provided in the CCTV system; Decompressing and storing the data compressed by the screen unit through the JPEG algorithm, and extracting only the lowest frequency component among the data generated in the decompression process; And And comparing and analyzing the extracted lowest frequency component with the lowest frequency component of the previous screen.

Images