Intelligent video security and protection system and signal processing method thereof

Abstract

The invention provides an intelligent video security and protection system and a signal processing method thereof. The system comprises a video analysis front end and a central video server, wherein the video analysis front end comprises a video input interface, a control module and a video output interface; the video input interface is used for deserializing a high-speed SDI (Serial Digital Interface) video signal into a digital video stream signal meeting BT.656 standard; the control module is used for performing preprocessing and video behavior analysis on the digital video stream signal and outputting an IP (Internet Protocol) video stream signal; the video output interface is used for transmitting the IP video stream signal to the central video server. Compared with the prior art, the system and the method have the advantages that multiple targets and multiple rules can be effectively realized aiming at real-time high-definition video and the video behavior analysis is simultaneously performed. In addition, great use of stream line operation for an algorithm is realized apart from the use of the parallel processing characteristic of an FPGA (Field Programmable Gate Array), and the data throughput of processing is improved by using a mode that the speed is increased by reducing the area.

Description

Technical field [0001] The invention relates to the technical fields of public safety monitoring, video image processing, computer vision, artificial intelligence and the like, and in particular to an intelligent video security system based on an FPGA-SOC chip and a signal processing method for the security system. Background technique [0002] At present, the security system has gradually shifted from analog security to digital security. The future security system trend is intelligent, high-definition digital, and networked. Among them, intelligent solutions are an important part and urgent requirement of future security. In short, intelligent security refers to the use of computer vision, artificial intelligence and other professional knowledge to process video images, analyze the attributes, characteristics, and movement direction of the target to determine whether the target meets the set rules and whether to send an alarm signal In order to deal with the incident in time a...

AI Technical Summary

Problems solved by technology

The disadvantage of the former solution is that when the number of video stream channels is large, the amount of video data that needs to be analyzed on the server side is also huge, and the analysis efficiency of pure software will be significantly reduced, which cannot meet the needs of real-time analysis and alarm of the security system

For example, for a 5M pixel video with a resolution of 2560*2048 and 15 frames per second (Frames Per Second, FPS), even if a high-end DSP chip is used, it cannot be processed due to the limitations of the DSP processing core processing capability and clock frequency.

In addition, for multi-objective, multi-rule and complex operation situations, due to the limited multi-task execution capability of the video processing unit integrated in the DSP core, coupled with many reasons of the serial processing mechanism, the data processing capability of the video stream signal is still limited

Images