KR100252468B1 - Method for storing a monitoring video information - Google Patents

Abstract

PURPOSE: A method for storing monitor video information is provided to select necessary video information by selecting a video according to a video recognition technique, and by digitally compressing and storing video information selected. CONSTITUTION: A video monitoring device is composed of a video camera(10), a video processing unit(12), a video display unit(14), a video recording unit(20). The video camera(10) picks up a video of a monitoring object and generates a video signal. The video processing unit(12) amplifies the video signal transmitted from the video camera(10) and processes a luminescence and a chrominance signal. The video display unit(14) inputs the video signal provided from video processing unit(12). The video recording unit(20) is composed of a digital video processing unit(22), a video compression unit(24), a video storage unit(26), a storage object judgement unit(28) and a control unit(30). The digital video processing unit(22) inputs the video signal provided from the video processing unit(12). The storage object judgement unit(28) inputs a monitor video provided from the video processing unit(22). The video compression unit(24) inputs video data of a storage object judged from the judgement unit(28). The video recording unit(26) is composed of a non-volatile semiconductor memory. The control unit(30) controls each block.

Description

How to store surveillance video information

The present invention relates to a video surveillance apparatus, and more particularly, to a surveillance video information storage method for compressing and storing surveillance video information captured under the condition that the feature information of the surveillance video is the largest.

In general, a conventional video surveillance apparatus picks up an image of a visitor with a video camera and stores a video signal of the picked up visitor on a storage medium such as a video tape. Therefore, in the conventional video surveillance apparatus, when a continuous video is to be stored due to a limited storage capacity of the video tape, it is troublesome to frequently replace the video tape, and when the replacement operation is careless, There was a problem. In addition, if the video tape is repeatedly used many times, the image quality is deteriorated and the visitor information becomes unfamiliar.

In order to solve the problem of frequent replacement, a method of using a single video tape for a long time has been introduced by sampling a moving image of a visitor at a predetermined time interval or storing a still image. However, in this method, when a visitor at the time of sampling is standing on the back of the camera or turning his head, important information can not be stored and thus the effect of monitoring becomes poor.

An object of the present invention is to provide a surveillance image information storage method capable of capturing and storing efficiently the images including the feature information of the surveillance target object most efficiently in order to solve the problems of the related art.

According to an aspect of the present invention, there is provided a method of monitoring an object to be monitored, comprising: Extracting a surveillance image from the picked-up image; Extracting a feature image from the extracted surveillance image; Calculating an area of the extracted feature image; Comparing the area of the current feature image with the area of the previous feature image; Performing the steps repeatedly at predetermined intervals to select a surveillance image corresponding to a feature image having the largest area as a storage object; Compressing the selected monitoring image data; And generating a compressed monitoring video data file and storing the compressed monitoring video data in a storage medium with the generated file name.

1 is a block diagram showing a configuration of a surveillance system according to the present invention;

2 is a flowchart illustrating a method of storing surveillance image information according to the present invention.

FIG. 3 is a diagram for explaining a selection process of a surveillance image by a feature image according to the present invention; FIG.

4 is a block diagram showing the configuration of another embodiment of the monitoring system according to the present invention;

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Reference will now be made in detail to the preferred embodiments of the present invention.

Referring to FIG. 1, an image monitoring apparatus includes a video camera 10, an image processing unit 12, an image display unit 14, and an image recording unit 20. The video camera 10 is a video camera using a CCD type solid-state image pickup device to pick up a video to be monitored and generate a video signal. The image processing unit 12 amplifies the image signal transmitted from the video camera 10 and processes luminance and color signals. The video display unit 14 includes a monitor for inputting the video signal provided from the video processing unit 12 and displaying the video to be monitored picked up on the screen of the CRT and its control circuit.

The image recording unit 20 includes a digital image processing unit 22, an image compression unit 24, an image storage unit 26, a storage object determination unit 28, and a control unit 30. Each block of the image recording unit 20 is divided into functional blocks. The specific hardware configuration includes a microprocessor, a system RAM, a system ROM, a digital signal processor, a video memory, and other clock generators. The digital image processing unit 22 receives an image signal provided from an image processing unit, digitally converts the image signal, stores the image signal in a field memory, and extracts a monitoring image from the digital image signal. The storage object determination unit 28 receives the monitoring image provided from the digital image processing unit 22, extracts the characteristic image from the monitoring image, and calculates the characteristic value from the extracted characteristic image. Then, the image to be stored is judged by comparing the calculated feature value with the previously calculated feature value. The image compressing unit 24 receives the image data to be stored determined by the storage target determining unit 28 and compresses the image data in accordance with a normal image data compression method such as AV1, JPEG, Motion JPEG, MPEG1, or MPEG2 Compress. The compressed image data is transmitted to and stored in the image storage unit 26. The image storage unit 26 is constituted by a nonvolatile semiconductor memory such as an EEPROM, a flash memory or the like. The control unit 30 controls each block of the image recording unit 20, generates a file of image data to be stored, and provides time information such as a storage date and a storage time.

The operation and effect of the present invention will be described with reference to FIG. 2 and FIG. 3 as an example of a visitor monitoring apparatus.

First, when the visitor presses the doorbell, the system is activated to pick up the image of the visitor through the video camera 10 (202). The digital image processing unit 22 extracts only the face image, which is the surveillance image, (204). The extraction of the facial image uses a known method used in the conventional image recognition field, so a detailed description thereof will be omitted. The storage object determining unit 28 extracts the eyes and mouth from the provided facial image as shown in FIG. 3 (206), and calculates the area ESO and MSO of the extracted eye and mouth feature images as feature values (208). As shown in FIG. 3A, if the feature value is the feature value obtained first by ESO1 and MSO1, the feature value register stores the value (210) because there is no previous feature value. If there is a previous feature value in step 210, the current feature value is compared with the previous feature value (step 214). If the current facial image is obtained as shown in FIG. 3B, since the current feature values ESO2 and MSO2 are greater than the stored ESO1 and MSO1, the value stored in the feature value register is updated to the current feature value (step 216). If it is smaller in step 214, the previous feature value is maintained and step 202 is performed. If the feature values ESO3 and MSO3 for the face image as shown in FIG. 3C are obtained through steps 202 to 210, the time is checked through steps 214 and 206 to check if a predetermined time has elapsed (step 218 ). When the predetermined time has elapsed, the monitoring image corresponding to the stored characteristic value, that is, the face image of FIG. 3 (c), is compressed by the image compression unit 24 (220). The compressed image data is stored in the image storage unit 26 together with the storage date and the storage time information as the data file name generated by the controller 30 (222).

4 shows another embodiment of the video surveillance apparatus according to the present invention. In another embodiment, the present invention further includes a microphone 40, a voice processing unit 42, and a voice compression unit 44 for storing voice data of visitors as well as video data. The voice processing unit 420 amplifies the input voice signal and outputs the digital voice signal to the voice compressing unit 44. The voice compressing unit 44 transmits the compressed voice data to the image storing unit 26, Video and audio are associated with each other and stored in a storage medium.

The present invention can be applied not only to a visitor monitoring system such as a home visitor monitoring system or a bank surveillance system, but also to all the images to be monitored by an automobile or the like.

Also, in the present invention, data stored in the image storage unit 26 can be output by a request from the police, or by checking a secret number input by a limited security person, that is, a restricted person. And is connected to the video reproducing apparatus by a person who can connect to the video reproducing apparatus to output the video data. In the video reproducing apparatus, the compressed video data inputted by connection is expanded and the extended video data is displayed on the LCD or CRT screen. Therefore, in the present invention, since the stored image data is stored in a compressed state, it is difficult for anyone to damage or reproduce the stored image data without a decoder, and it is possible to safely provide image information of a suspect in a crime incident.

As described above, according to the present invention, the image information of the surveillance object picked up by the video camera is selected by the image recognition technology and the image information of the selected surveillance object is the largest, Only the information can be selected and the selected image data can be compressed and stored with a minimum amount of data, so that it can be easily stored in a storage medium having a relatively small storage capacity such as a nonflammable semiconductor memory. In addition, since the stored image information is stored in the semiconductor memory as digital data, there is no fear that image quality will be deteriorated even if it is repeatedly reproduced a number of times.

Claims (4)

A step of picking up a surveillance target image with a video camera; Extracting a surveillance image from the picked-up image; Extracting a feature image from the extracted surveillance image; Calculating an area of the extracted feature image; Comparing the area of the current feature image with the area of the previous feature image; Performing the steps repeatedly at predetermined intervals to select a surveillance image corresponding to a feature image having the largest area as a storage target; Compressing the selected monitoring image data; And generating a compressed monitoring video data file and storing the compressed monitoring video data in a storage medium with the generated file name. The method of claim 1, wherein the monitored image is a face image of a visitor and the feature image is an eye and mouth image of the face image. The method as claimed in claim 1, wherein the method of compressing the supervised image is one of AV1, JPEG, Motion JPEG, MPEG1, and MPEG2. The method of claim 1, wherein the storage medium is a nonvolatile semiconductor memory.