RU2336659C1 - High-definition video surveillance system - Google Patents

Abstract

FIELD: physics, video technology.

SUBSTANCE: high-definition video surveillance system relates to real-time video stream transmission systems. The objective is achieved by including at least one camera with CCD matrix transmitting video stream to surveillance point device with coding unit which can capture video signal and convert it into frame stream in RAW or BMP format with further conversion of graphic frames into three formats. In the first format each frame is automatically shrunk in size at 25 fps rate for further displaying on surveillance point monitors. Simultaneously in the second format each frame undergoes contour compression of graphic files and is compressed at 5 fps rate, and sent to storage in compressed form at the surveillance point device. Simultaneously in the third format each frame is shrunk, undergoes contour compression of graphic files at 5-12 fps rate and is sent to mobile video display devices in stream mode over radio communication channels.

EFFECT: improved image quality due to contour compression of each frame.

3 cl, 2 dwg

Description

The invention relates to a communication system, in particular to systems for transmitting a video stream in real time over a considerable distance over radio channels of communication, and can be used in systems for transmitting and receiving compressed graphic files.

A hierarchical hybrid method for detecting frame changes for MPEG-compressed video is known from the prior art, including detection of changes, hierarchically applying the global color distribution difference among the color histogram structures, detection of the reduced frame change when using the global color difference difference of the histogram among the adjacent register, the structure sheet after additional the decryption of the structure sheet is detected in the area of the applicant, the decryption of the bit creates within the abbreviated detection applicant barter frame at the macroblock level in the bit sequence patterns separating two adjacent reference video frame, determine feature distribution intra coded block and copies each bit type structure in the region and detecting a reduction when using particular types and distribution links (US 7027509, 13.09.2001).

A known method for compressing video documents, comprising the steps of decomposing a video document into frames, extracting image samples of said frames, measuring distances between said image samples, grouping images whose distance is less than a predetermined threshold into a group of images, calculating the relative cost of encoding each image relative to other images of the group images, including the steps of calculating the minimum decoder for each group of images, minimizing the cost to encoding of each group of image encoding of each group of images, viewing the images in a given sequence by a decoder (US 2006002612, January 5, 2006).

A known system for processing a sequence of video structures, including: an encoding device for dynamically encoding a sequence of video structures to produce pseudo-contacts of bits in the structure of a compressed signal in a change in action within a sequence of video structures. The encoding device includes: means for detecting new actions in the sequence of video structures and means for dynamically determining the data group of the pixel of the frequency domain to be stored for the structure of the new action responsive to the detection of a new action; means for dynamically determining the inclusion of means using a frequency domain data management model (US 6914939, July 5, 2005).

The disadvantages of these solutions are poor image quality, slow transfer of compressed files.

The problem to which the proposed invention is directed, is to create such a high-resolution video surveillance system that would eliminate the above disadvantages.

The technical result achieved by the implementation of this invention is to improve the quality of the image in security systems due to the contour compression of each frame, which allows you to save the details necessary for observation, neglecting those frames or places in frames that are not of interest for observation.

The specified technical result is achieved in a high-resolution video surveillance system containing at least one camera with a CCD matrix, transmitting the video stream to the observation point device with an encoding unit, capable of capturing the video signal and converting it into a stream of frames in RAW or BMP format, with the subsequent conversion of graphic frames into three formats, in the first of which each frame is automatically reduced in size at a speed of 25 frames per second for subsequent display on point n of monitors observations, in parallel with this, in the second format, each frame undergoes contour compression of graphic files and is compressed at a speed of 5 frames per second and sent in compressed form for storage to the device of the observation point, and simultaneously in the third format, each frame is reduced, undergoes contour compression of graphic files from at a speed of 5-12 frames per second and is transmitted in streaming mode via radio communication channels to mobile video display facilities.

When contouring compression of graphic files, the frame of the graphic image is reduced by resizing at least 4 times, compressed and saved in a compressed file, while the compressed file is decompressed and expanded at least 4 times, and the original frame of the graphic image is superimposed decompression enlarged frame, then a search is made for the difference in pixel values based on contrast elements using arithmetic subtraction, and the frame with the values of the contour of the contrast elements is compressed and saved tsya in a compressed file.

In the third format, each frame is reduced to a size of 160 × 120 or 320 × 240, or 640 × 480 pixels.

The invention is illustrated by drawings, where figure 1 shows a diagram of a high-resolution video surveillance system HD SECYRITI - VKS; figure 2 - fragments of the input file in HD 1920 × 1080 pixels before compression and after decompression by standard JPEG and STU technologies.

The SECYRITI-VKS high-definition video surveillance system HD contains: at least one 1 HD video camera with a CCD (CCD) resolution of a matrix of 3 Mega pixels or more; device 2 observation points with a coding unit, made on the basis of a personal computer with a processor of at least 2 GHz; at least one monitor 3 for monitoring on the device 2 in the control room with a resolution of 1024 × 768 to 1920 × 1080 or more pixels; at least one mobile device 4 for receiving video signal via WiFi radio channels or others.

The observation point device 2 contains a memory unit (hard disk) for recording and storing compressed frames in a format with the extension * .stu (* - means replacing all previous characters) according to STU Technologies (Space Technology Universal - Space Universal Technology) up to 200 Gigabytes, from the calculation of recording compressed frames without interruption for 24 hours.

The contour compression of STU image files (STU file is a compressed file with the * .stu extension) is as follows: a graphic image frame is reduced by resizing at least 4 times, it is compressed and saved in a compressed STU file; the compressed STU file is decompressed and expanded at least 4 times; the original frame of the graphic image is superimposed on the decompression enlarged frame and a search is made for the difference in pixel values based on contrasting elements using arithmetic subtraction; the frame with the contour values of the contrasting elements is compressed and saved in a compressed STU file.

The system may contain separate devices for transmitting to mobile means of WiFi communication or specialized devices for transmitting video images via radio channels.

The encoding unit was created on the basis of STU Technologies for processing, compression and transmission of high-definition video HD SECURITY.

Unlike standard solutions of security systems, in this system, the processing, compression and transmission of video information is carried out in three formats:

1. The video stream received from the CCD matrix of the camera is captured as a stream of uncompressed frames in RAW or BMP format at a speed of 25 frames per second.

In real time, each frame is compressed by the contour compression method according to STU Technologies, with options for the optimal ratio of image quality and compression ratio, providing the maximum compression ratio and maximum forward stream transmission rate at 25 frames per second, which is transmitted to the monitors of the observation site.

2. In parallel, every 5 frames are captured from a direct stream of uncompressed frames at a speed of 25 frames per second.

In real time, every fifth frame is compressed by the contour compression method according to STU Technologies, with options for the optimal ratio of image quality and compression ratio, ensuring the best possible image quality of every fifth frame, and sent to the memory unit of the device 2 observation points for storage.

3. Simultaneously with these two actions, each uncompressed frame is captured from the stream at a speed of 5 frames per second and frame sizes are reduced to formats with a resolution of 640 × 480; 320 × 240; 160 × 120 pixels.

In real time, each frame is compressed by the contour compression method according to STU Technologies, with options for the optimal ratio of image quality and compression ratio and is transmitted in streaming mode via radio channels to mobile video devices for various purposes.

The video signal received from at least one high-resolution camera 1 is supplied to the input of the device 2 of the observation point via a high-speed line 1394. The STU HD SECURITY encoding unit captures the video signal, turns it into a stream of frames in * .RAW or *. BMP format. The stream of graphic format frames is divided into 3 ways in parallel:

1. Each frame is automatically reduced in size depending on the installed monitors in the control and monitoring station, at a speed of 25 frames per second.

2. Each frame with the resolution received from the camera used is compressed by STU Technology at a speed of 5 frames per second and sent for storage to the memory unit in compressed * .stu form.

3. Each frame is reduced to a size of 160 × 120; 320 × 240; 640 × 480 pixels, STU Technology is compressed at a speed of 5-12 frames per second (depending on the speed of mobile communications) and sent in compressed form to mobile video displays (video transmission devices).

The encoding unit is able to work with all of the listed formats and devices and has various functions for the optimal operation of the entire system as a whole, both in automatic mode and with manual settings.

The basis was adopted by the new developing standard HD format with a resolution of 1920 × 1080 pixels or more.

To maximize the preservation of HD image quality, processing and compression is performed only with each frame individually.

Inter-frame conversions are completely eliminated, unlike the existing systems where various video codecs are used.

The use of video codecs leads to blurred image quality and with an increase in impossibility to consider the necessary details of the observed objects.

Using STU Technology contour compression of each frame, it becomes possible to preserve the details necessary for observation as much as possible, neglecting those frames or places in frames that are not of interest for observation.

The use of VKS HD SECURITY high-definition video surveillance systems represents a transition to a new, advanced level in the field of security.

Due to the use of digital, multi-pixel cameras with a resolution from 1440 × 1080 pixels to 470 megapixels in VKS systems, image quality is increased by at least 4 times, compared with the current CIF and VGA format standards with a resolution of 640 × 480 pixels and 720 × 576 pixels.

The contour compression method according to STU Technologies is capable of simultaneously processing, compressing and transmitting a video stream in three formats with high image quality simultaneously in real time.

Storage of compressed files is a sequence of high-resolution frames at a speed of 5 frames per second, while the storage capacity for frames with continuous recording 24 hours a day is only 160-200 Gigabytes.

Extremely important in video surveillance systems is the ability to transmit to wireless mobile devices. This VKS system is capable of transmitting a video stream in real time over a considerable distance over radio channels of communication.

VKS high-resolution multifunctional surveillance system has a wide range of uses and applications.

Claims (3)

1. A high-resolution video surveillance system, characterized in that it contains at least one camera with a CCD matrix, transmitting the video stream to the observation station device with an encoding unit, capable of capturing the video signal and converting it into a stream of frames in RAW or BMP format, with subsequent conversion of graphic frames into three formats, in the first of which each frame is automatically reduced in size at a speed of 25 frames per second for subsequent display on the monitors of the observation point, parallel about this, in the second format, each frame undergoes contour compression of graphic files and is compressed at a speed of 5 frames per second and sent in a compressed form for storage to the device of the observation point, and simultaneously in the third format, each frame is reduced, contour is compressed in graphic files at a speed of 5 -12 frames per second and is transmitted in streaming mode via radio communication channels to mobile video display facilities. 2. The system according to claim 1, characterized in that during the contour compression of graphic files, the frame of the graphic image is reduced by resizing at least 4 times, compressed and stored in a compressed file, while the compressed file is decompressed and expanded at least at least 4 times, and the original frame of the graphic image is superimposed on the decompression enlarged frame, then the difference in pixel values is searched for by the sign of contrasting elements using arithmetic subtraction, and the frame with the values of the contour The contrasting elements are compressed and saved in a compressed file. 3. The system according to claim 1, characterized in that in the third format, each frame is reduced to a size of 160 × 120 or 320 × 240 or 640 × 480 pixels.

Images