本文選題：視頻圖像處理　切入點(diǎn)：車輛檢測(cè)　出處：《長(zhǎng)安大學(xué)》2014年碩士論文

【摘要】：隨著經(jīng)濟(jì)的發(fā)展和社會(huì)的進(jìn)步，對(duì)交通運(yùn)輸?shù)母鞣N需求也迅猛增長(zhǎng)，現(xiàn)有交通路網(wǎng)的管理能力與管理設(shè)施已經(jīng)不能滿足日益增長(zhǎng)的交通流量需求，交通運(yùn)輸問(wèn)題日益嚴(yán)重，交通擁擠，時(shí)常有車輛隨意變道的情況發(fā)生，引起了許多不必要的交通事故。為了解決這些交通問(wèn)題，智能交通系統(tǒng)（ITS）的研究被提到了重要位置。運(yùn)動(dòng)車輛檢測(cè)與跟蹤系統(tǒng)作為ITS的重要組成部分，成為了目前的研究熱點(diǎn)。車輛跟蹤技術(shù)是實(shí)現(xiàn)視頻車輛監(jiān)測(cè)的關(guān)鍵技術(shù)之一，如何實(shí)現(xiàn)基于視頻圖像處理的車輛檢測(cè)，以及車輛跟蹤的算法研究成為了國(guó)內(nèi)外研究的重要問(wèn)題。 本文將車輛跟蹤系統(tǒng)分為車輛檢測(cè)、車輛跟蹤兩個(gè)主要的技術(shù)模塊，，并對(duì)各模塊的主要技術(shù)方法進(jìn)行了介紹。在車輛檢測(cè)模塊，通過(guò)對(duì)幾種常用的檢測(cè)方法進(jìn)行分析比較，確定采用基于背景差分法的檢測(cè)方法進(jìn)行車輛檢測(cè)，重點(diǎn)研究了該算法中背景的提取和更新、閾值的選取方法等關(guān)鍵技術(shù)。在車輛跟蹤模塊，重點(diǎn)對(duì)車輛跟蹤的幾種算法進(jìn)行分析比較，確定使用基于目標(biāo)車輛質(zhì)心的跟蹤算法進(jìn)行跟蹤標(biāo)記，最后在跟蹤標(biāo)記的基礎(chǔ)上得到追蹤目標(biāo)車輛的運(yùn)行軌跡。 運(yùn)用MATLAB分析軟件及實(shí)例對(duì)所選擇的算法進(jìn)行驗(yàn)證，確定所選擇的算法是否可以達(dá)到預(yù)期效果。
[Abstract]:With the development of economy and social progress, but also the rapid growth of demand for the various transportation facilities, management ability and management of existing traffic network has been unable to meet the growing traffic demand, traffic problems have become increasingly serious, traffic congestion, vehicles often have occurred change situation, caused a lot of unnecessary traffic accident. In order to solve these traffic problems, intelligent traffic system (ITS) research has been raised to an important position. The moving vehicle detection and tracking system as an important part of ITS, has become a research hotspot at present. Vehicle tracking is one of the key technologies of video vehicle monitoring, how to realize vehicle detection based on video image processing research on vehicle tracking algorithm, and has become an important problem of research at home and abroad.
The vehicle tracking system for vehicle detection, vehicle tracking two technical modules, and the main technical methods of each module are introduced. The vehicle detection module, based on several commonly used detection methods of analysis and comparison, determine the vehicle detection background difference detection method based on the method, focus on background extraction and update the algorithm, the key technology of threshold selection methods. In the vehicle tracking module, focuses on several algorithms of vehicle tracking analysis and comparison, determine the use of the target vehicle centroid tracking algorithm based on tracking markers, finally get the trajectory tracking target vehicle based on tracking markers.
MATLAB analysis software and examples are used to verify the selected algorithms and determine whether the selected algorithm can achieve the desired results.

【學(xué)位授予單位】：長(zhǎng)安大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：U495

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 肖梅;韓崇昭;;基于在線聚類的背景減法[J];模式識(shí)別與人工智能;2007年01期

本文編號(hào)：1677447