基于目標(biāo)分離的視頻濃縮技術(shù)在安防行業(yè)的應(yīng)用和研究

馬婷婷

摘要：隨著安防行業(yè)的發(fā)展，從早期的模擬監(jiān)控、數(shù)字監(jiān)控、網(wǎng)絡(luò)監(jiān)控逐步向高清化、智能化方向發(fā)展。海量的視頻資源信息為公安部門(mén)提供案件偵破線(xiàn)索依據(jù)的同時(shí)也給查閱觀看帶來(lái)了巨大的困難。文章主要介紹了視頻智能分析的技術(shù)特點(diǎn)、研究并實(shí)現(xiàn)了在視頻監(jiān)控中引入目標(biāo)分離技術(shù)從而實(shí)現(xiàn)對(duì)目標(biāo)對(duì)象的快速瀏覽。文中充分結(jié)合現(xiàn)有安防行業(yè)的業(yè)務(wù)需求并利用視頻分析的技術(shù)，在提高涉案線(xiàn)索獲取效率方面提出了有效的解決方案。從實(shí)際運(yùn)行效果來(lái)看，智能分析在視頻安防行業(yè)應(yīng)用中可有效提升事后調(diào)查研判的效率。

關(guān)鍵詞：智能分析；模式識(shí)別；目標(biāo)分離；云計(jì)算

中圖分類(lèi)號(hào)：TP31 文獻(xiàn)標(biāo)識(shí)碼：A 文章編號(hào)：1009-3044（2014）35-8529-02

The Application and Study of Target Separation Based Video Synopsis Technology in Security Industry

MA Ting-ting

（Hefei University， Hefei 230601， China）

Abstract： The technology used in security industry has gradually developed from early analog monitoring， digital monitoring， network monitoring to high definition and intelligent direction. Massive quantities of video resources not only provide clues for case detections for police departments， but also bring enormous difficulties for investigators to view and browse such big amount of videos. This paper describes the technical features of intelligent video analysis and the implementation of introduction of the target separation technology into video surveillance， which can achieve quick browsing of target objects. This paper fully integrates existing security industry needs and uses video analysis technology to propose effective solutions in order to improve the efficiency of obtaining case clues. From the practical point of view that is based on real operating results， intelligent analysis applications in the video security industry can effectively improve the efficiency of the later case investigation and judgment.

Key words： Intelligent analysis； pattern recognition； target separation； cloud computing

1 概述

隨著社會(huì)的不斷發(fā)展，越來(lái)越多的視頻監(jiān)控點(diǎn)位分布在城市的各個(gè)角落，在接到民眾報(bào)警后警方絕大部分會(huì)考慮從已經(jīng)安裝部署的攝像頭采集的視頻資源中尋找破案線(xiàn)索，但由于視頻采集內(nèi)容較多，依靠偵破人員人工進(jìn)行視頻的瀏覽調(diào)閱耗費(fèi)的人力物力巨大，而且通常要消耗較長(zhǎng)的時(shí)間。利用智能分析技術(shù)讓計(jì)算機(jī)將視頻內(nèi)容進(jìn)行預(yù)先處理呈現(xiàn)處理的結(jié)果再加上人工的確認(rèn)可大大提高破案線(xiàn)索排查的效率[1]。

2 研究擬采用的技術(shù)

2.1 目標(biāo)分離技術(shù)

目標(biāo)分離技術(shù)是視頻智能分析技術(shù)最核心基礎(chǔ)的技術(shù)之一，根據(jù)圖像視覺(jué)分析技術(shù)，將目標(biāo)從視頻場(chǎng)景的背景中分離出來(lái)，分析并監(jiān)控在攝像機(jī)視頻場(chǎng)景內(nèi)出現(xiàn)的目標(biāo)[2]。……