【摘要】：隨著視頻監(jiān)控系統(tǒng)的迅速普及,計(jì)算機(jī)視覺正逐漸為大眾所熟知,特別是計(jì)算機(jī)視覺領(lǐng)域的運(yùn)動目標(biāo)檢測與定位技術(shù),近年來受到越來越多的重視,已廣泛應(yīng)用于安防監(jiān)控、入侵檢測、無人駕駛汽車等領(lǐng)域。傳統(tǒng)的視頻監(jiān)控系統(tǒng)需要較多的人工參與,無法應(yīng)對越來越復(fù)雜多變的監(jiān)控環(huán)境,而基于視頻圖像的智能視頻監(jiān)控系統(tǒng)不需要或者只需要很少的人工參與,可以同時(shí)對多個(gè)場景進(jìn)行監(jiān)控,能夠在極短的時(shí)間內(nèi)分析、發(fā)現(xiàn)監(jiān)控中的異常行為并實(shí)時(shí)發(fā)出告警。本文對視頻監(jiān)控中的核心目標(biāo)檢測技術(shù)進(jìn)行了深入研究,分析當(dāng)今主流的目標(biāo)檢測算法,并針對具體的應(yīng)用場景,使用了多種目標(biāo)檢測算法構(gòu)建目標(biāo)檢測系統(tǒng),在多臺攝像機(jī)上應(yīng)用該系統(tǒng),構(gòu)建了目標(biāo)空間定位系統(tǒng)。本文的主要內(nèi)容有:1.提出一種基于幀間差分和相似度校驗(yàn)的多目標(biāo)檢測方法。該方法采用幀間差分法得到差分圖像,經(jīng)過形態(tài)學(xué)處理得到目標(biāo)輪廓,針對非剛體目標(biāo)形變引起的目標(biāo)分裂等問題,采用多種方法進(jìn)行合并,通過計(jì)算區(qū)域相關(guān)性辨別多目標(biāo),可以快速確定目標(biāo)在圖像中的粗略位置,同時(shí)具有較高的運(yùn)行效率,為目標(biāo)的精確檢測提供位置參考。2.研究基于可變形部件模型的目標(biāo)檢測算法,在差分算法得到目標(biāo)粗略位置的基礎(chǔ)上,通過使用該算法可以獲得指定類型目標(biāo)的精確位置�？勺冃尾考Ｐ褪且环N基于目標(biāo)特征統(tǒng)計(jì)學(xué)習(xí)的檢測算法,該方法使用HOG描述子作為模型特征,具有良好的幾何和光學(xué)轉(zhuǎn)換不變性;同時(shí)由于形變模型的引入,使得該方法對于物體的非剛體形變具有極好的魯棒特性,特別使用于行人等非剛體目標(biāo)檢測的應(yīng)用場景。3.研究基于圖像的目標(biāo)空間定位方法,在兩點(diǎn)前向交會的基礎(chǔ)上,提出了一種基于聯(lián)合概率分布的多路交會算法,使用概率模型根據(jù)多臺攝像機(jī)的檢測結(jié)果得到目標(biāo)在空間位置的估計(jì)值,這種方法比傳統(tǒng)的多次使用兩路交會然后取幾何中心的方法可靠性和精度都有了極大提高;通過數(shù)學(xué)模型對定位系統(tǒng)在水平方向和垂直方向的最小理論誤差進(jìn)行了分析,得出了系統(tǒng)實(shí)施環(huán)境下定位的理論精度極限。
[Abstract]:With the rapid popularization of video surveillance system, computer vision is gradually becoming known to the public, especially in the field of computer vision moving object detection and location technology, in recent years, more and more attention has been paid to, has been widely used in security surveillance, intrusion detection, driverless vehicles and other fields. Traditional video surveillance system needs more. Artificial participation can not cope with the increasingly complex and changeable monitoring environment. However, intelligent video surveillance system based on video images does not need or only need a small amount of manual participation. It can simultaneously monitor multiple scenes. It can analyze in a very short time, discover abnormal behavior in monitoring and give real-time alarm. In this paper, the core target detection technology in frequency monitoring is deeply studied, and the main target detection algorithms are analyzed. Aiming at the specific application scenarios, a target detection system is constructed by using a variety of target detection algorithms, and a target location system is constructed by using the system on multiple cameras. The main contents of this paper are as follows: 1. A multi-target detection method based on inter-frame difference and similarity checking is proposed. The difference image is obtained by inter-frame difference method, and the object contour is obtained by morphological processing. In order to solve the problem of target splitting caused by non-rigid object deformation, a variety of methods are used to merge. The multi-target can be identified quickly by calculating the region correlation. The rough position of the target in the image is determined, and the operation efficiency is high, which can provide a reference for the accurate detection of the target. 2. The target detection algorithm based on deformable component model is studied. On the basis of the rough position of the target obtained by the difference algorithm, the precise position of the specified type of target can be obtained by using this algorithm. The shape component model is a detection algorithm based on target feature statistical learning, which uses HOG descriptors as model features and has good geometric and optical transformation invariance. At the same time, due to the introduction of deformation model, this method has excellent robustness for non-rigid deformation of objects, especially for pedestrians and other non-rigid deformation. 3. The application scene of object detection is studied. Based on the two-point forward rendezvous, a multi-path rendezvous algorithm based on joint probability distribution is proposed. The estimation value of object position in space is obtained by using probability model according to the detection result of multiple cameras. This method is more than the traditional method. The reliability and accuracy of the method of using two paths to intersect and then take the geometric center are greatly improved. The minimum theoretical error of the positioning system in the horizontal and vertical directions is analyzed by mathematical model, and the theoretical accuracy limit of the positioning system in the implementation environment is obtained.
【學(xué)位授予單位】：電子科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2015
【分類號】：TP391.41;TN948.6

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 李延軍;黃國勇;;基于幀間差分的火災(zāi)預(yù)警研究[J];微計(jì)算機(jī)應(yīng)用;2010年10期

2 樊曉亮;楊晉吉;;基于幀間差分的背景提取與更新算法[J];計(jì)算機(jī)工程;2011年22期

3 武懷金;王武江;;基于幀間差分方法的運(yùn)動目標(biāo)檢測[J];黑龍江科技信息;2012年10期

4 張丹丹;婁煥;;幀間差分法中閾值的選擇[J];科技信息;2013年34期

5 張秋仙;;幀間差分法與平均背景法在運(yùn)動檢測中應(yīng)用的研究[J];企業(yè)科技與發(fā)展;2010年18期

6 薛麗霞;羅艷麗;王佐成;;基于幀間差分的自適應(yīng)運(yùn)動目標(biāo)檢測方法[J];計(jì)算機(jī)應(yīng)用研究;2011年04期

7 廖馬騰;李全;;基于System Generator的幀間差分運(yùn)動目標(biāo)檢測算法仿真[J];電子質(zhì)量;2013年04期

8 崔星;閆清東;;基于幀間差分方法的道路車輛檢測系統(tǒng)[J];微計(jì)算機(jī)信息;2007年10期

9 楊慶華;李薇;舒蘭英;賀超;何先波;;基于雙高斯平均似然度和幀間差分的人臉視頻圖像膚色提取[J];數(shù)據(jù)采集與處理;2013年01期

10 李惠松;王小銘;張玉霞;;一種基于幀間差分與時(shí)空相關(guān)性分析的運(yùn)動目標(biāo)檢測算法[J];計(jì)算機(jī)與數(shù)字工程;2007年12期

相關(guān)會議論文 前4條

1 金黎明;周曉光;蘇志遠(yuǎn);;一種基于幀間差分背景重建的動目標(biāo)檢測算法[A];2009年先進(jìn)光學(xué)技術(shù)及其應(yīng)用研討會論文集（上冊）[C];2009年

2 熊衛(wèi)華;向磊;李俊峰;趙新龍;;背景減除與幀間差分相結(jié)合的運(yùn)動目標(biāo)檢測方法[A];中國自動化學(xué)會控制理論專業(yè)委員會A卷[C];2011年

3 王靜靜;林明秀;魏穎;;基于灰度相關(guān)的幀間差分和背景差分相融合的實(shí)時(shí)目標(biāo)檢測[A];2009年中國智能自動化會議論文集（第六分冊）[中南大學(xué)學(xué)報(bào)（增刊）][C];2009年

4 林佳乙;于哲舟;張健;馬安娜;楚葉峰;;基于背景差分法和幀間差分法的視頻運(yùn)動檢測[A];2008中國儀器儀表與測控技術(shù)進(jìn)展大會論文集（Ⅲ）[C];2008年

相關(guān)碩士學(xué)位論文 前6條

1 謝松;多攝像頭非剛體目標(biāo)檢測與空間定位系統(tǒng)[D];電子科技大學(xué);2015年

2 馮堯文;基于幀間差分的運(yùn)動目標(biāo)穩(wěn)健檢測方法[D];哈爾濱工業(yè)大學(xué);2011年

3 胡敬舒;基于幀間差分的運(yùn)動目標(biāo)檢測[D];哈爾濱工程大學(xué);2013年

4 劉靜;智能視頻監(jiān)控關(guān)鍵技術(shù)研究[D];北京化工大學(xué);2012年

5 曾旭;基于DM6446的智能視頻監(jiān)控系統(tǒng)開發(fā)[D];杭州電子科技大學(xué);2012年

6 陳文鼎;基于嵌入式系統(tǒng)的液滴流速監(jiān)測設(shè)計(jì)與實(shí)現(xiàn)[D];南京理工大學(xué);2013年

，

本文編號：2250147