本文關(guān)鍵詞：室內(nèi)監(jiān)控中移動(dòng)檢測(cè)與跟蹤算法的改進(jìn)與實(shí)現(xiàn)　出處：《東南大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

  更多相關(guān)文章： 目標(biāo)檢測(cè) 人體識(shí)別 目標(biāo)跟蹤

【摘要】：基于智能視頻監(jiān)控的移動(dòng)目標(biāo)檢測(cè)、識(shí)別與跟蹤是計(jì)算機(jī)視覺領(lǐng)域研究的熱點(diǎn),在現(xiàn)代室內(nèi)安全防護(hù)系統(tǒng)中有著越來(lái)越多的應(yīng)用,使用這些技術(shù),我們可以快速獲取監(jiān)控區(qū)域中感興趣的前景目標(biāo)、識(shí)別前景目標(biāo),并對(duì)前景目標(biāo)跟蹤形成跟蹤軌跡,為后續(xù)目標(biāo)的行為分析與理解打下良好的基礎(chǔ)。本文以室內(nèi)監(jiān)控環(huán)境為研究場(chǎng)景,以單目標(biāo)人體為研究對(duì)象,并將單目標(biāo)人體的檢測(cè)、識(shí)別以及跟蹤作為本文研究的主要內(nèi)容,旨在通過對(duì)比分析現(xiàn)有的移動(dòng)目標(biāo)檢測(cè)與跟蹤算法,改進(jìn)現(xiàn)有算法的某些不足,避免監(jiān)控過程中常見的干擾,以提升室內(nèi)智能監(jiān)控系統(tǒng)的魯棒性。本文的主要工作如下:首先,在移動(dòng)目標(biāo)檢測(cè)階段,針對(duì)背景減差算法對(duì)光線變化比較敏感的缺點(diǎn),本文提出了基于GMM算法的背景減差算法,利用GMM算法良好的穩(wěn)定性以及對(duì)光線緩慢變化不敏感的特點(diǎn),為靜態(tài)背景圖像建立背景模型。另外,針對(duì)GMM算法對(duì)光照突變適應(yīng)性差的缺陷,則通過定義前景目標(biāo)所占的面積比率以及光線突變持續(xù)的幀數(shù)來(lái)檢測(cè)室內(nèi)光線是否發(fā)生突變。實(shí)驗(yàn)表明,改進(jìn)的移動(dòng)目標(biāo)檢測(cè)算法不僅可以完整檢測(cè)出前景目標(biāo),而且對(duì)于傳統(tǒng)的背景減差算法光線緩慢變化以及突變情況引起的檢測(cè)誤差也可以很好地解決,從而大大提升了移動(dòng)目標(biāo)檢測(cè)的準(zhǔn)確率以及查全率。其次,在人體目標(biāo)識(shí)別階段,針對(duì)室內(nèi)監(jiān)控環(huán)境下,不同前景目標(biāo)的分類問題,本文提出基于HOG特征的SVM分類器算法對(duì)前景目標(biāo)進(jìn)行分類,通過借助公共數(shù)據(jù)集INRIA提供的正負(fù)樣本進(jìn)行分類器訓(xùn)練。最后通過仿真實(shí)驗(yàn)驗(yàn)證了該分類算法具有較高的準(zhǔn)確率。最后,在單目標(biāo)人體跟蹤階段,針對(duì)傳統(tǒng)Camshift移動(dòng)目標(biāo)跟蹤算法抗遮擋性差以及目標(biāo)尺度變化過大敏感性的缺點(diǎn),本文提出了一種改進(jìn)的Camshift目標(biāo)跟蹤算法。采用對(duì)目標(biāo)分塊跟蹤的方式來(lái)處理目標(biāo)遮擋的問題,并通過定義目標(biāo)匹配率來(lái)判斷目標(biāo)不同程度的遮擋。另外,針對(duì)目標(biāo)尺度變化過大引入的跟蹤誤差,本文通過將目標(biāo)的幾何特征和目標(biāo)的顏色特征結(jié)合起來(lái),以更充分地描述目標(biāo),提高目標(biāo)的識(shí)別率。實(shí)驗(yàn)表明,改進(jìn)的移動(dòng)目標(biāo)跟蹤算法在保證系統(tǒng)實(shí)時(shí)性的前提下,對(duì)于傳統(tǒng)的Camshift跟蹤算法抗遮擋性差以及尺度變化過大帶來(lái)的跟蹤誤差都能很好地解決,提高了移動(dòng)目標(biāo)跟蹤階段的魯棒性。
[Abstract]:Moving target detection based on intelligent video surveillance, recognition and tracking is a hot research field of computer vision, it has more and more application in modern interior safety protection system, the use of these techniques, we can quickly get the foreground object interested in the monitoring area, identify the foreground objects, and the prospect of the target tracking tracking trajectory lay a good foundation for the follow-up behavior analysis and understanding of the target. This paper takes the indoor monitoring environment of the scene, with a single target body as the research object, and the detection of single target recognition and tracking of the human body, as the main content of this paper is to, through the comparative analysis of moving target detection and tracking algorithm in the existing, some overcome the shortcomings of the existing algorithms, avoid common interference in the process of monitoring and control, to enhance the robustness of indoor intelligent monitoring system. The main work of this paper is as follows: first of all In moving target detection, background subtraction algorithm, aiming at the light sensitive shortcomings, proposed subtraction algorithm GMM algorithm based on the background, using the GMM algorithm and good stability to light slow change characteristics is not sensitive to the static background images to establish the background model. In addition, according to the light mutation low adaptability of the GMM algorithm, through the definition of the area occupied by the foreground object and the ratio of the number of frames to detect the light mutation for interior light mutation. Experimental results show that the moving target detection algorithm can not only detect foreground objects, but also for the traditional background subtraction detection error caused by poor light slow change and mutation algorithm the situation can be solved very well, thus greatly enhance the accuracy of moving target detection and recall. Secondly, the human target recognition stage According to the monitoring, indoor environment, classification of different objects, this paper proposes a SVM classification algorithm based on HOG feature of foreground object classification, classifier training by means of positive and negative samples of public data sets provided by INRIA. The simulation experiment verifies the accuracy of the algorithm has a high classification. Finally, in the single target tracking the body, according to the traditional Camshift mobile target tracking algorithm for anti block difference and the scale change of target large sensitivity tracking algorithm is proposed in this paper an improved Camshift target. To deal with object occlusion problem using the target block tracking, and the matching rate to determine the occlusion target different degrees by definition of target. In addition, the tracking error for the target scale change is too large is introduced, this paper will combine the color features and geometric features of the target Up to more fully describe the target, improve the recognition rate. Experimental results show that the algorithm not only guarantees the real-time system under moving target tracking is improved, the traditional Camshift tracking algorithm of anti occlusion and scale variation of the tracking error caused by the large can be a good solution to improve the robustness of mobile the target tracking stage.

【學(xué)位授予單位】：東南大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前8條

1 王蒙;戴亞平;;基于FAST角點(diǎn)檢測(cè)的局部魯棒特征[J];北京理工大學(xué)學(xué)報(bào);2013年10期

2 ;Adaptive multi-feature tracking in particle swarm optimization based particle filter framework[J];Journal of Systems Engineering and Electronics;2012年05期

3 梁艷菊;李慶;陳大鵬;顏學(xué)究;;一種快速魯棒的LOG-FAST角點(diǎn)算法[J];計(jì)算機(jī)科學(xué);2012年06期

4 張濤;費(fèi)樹岷;王麗麗;;基于色彩相關(guān)直方圖和粒子濾波的目標(biāo)跟蹤[J];東南大學(xué)學(xué)報(bào)(自然科學(xué)版);2011年S1期

5 裴立志;王潤(rùn)生;;基于多個(gè)顏色分布模型的粒子濾波跟蹤算法[J];電路與系統(tǒng)學(xué)報(bào);2011年01期

6 甘明剛;陳杰;劉勁;王亞楠;;一種基于三幀差分和邊緣信息的運(yùn)動(dòng)目標(biāo)檢測(cè)方法[J];電子與信息學(xué)報(bào);2010年04期

7 劉惟錦;章毓晉;;基于Kalman濾波和邊緣直方圖的實(shí)時(shí)目標(biāo)跟蹤[J];清華大學(xué)學(xué)報(bào)(自然科學(xué)版);2008年07期

8 宋紅,石峰;基于人臉檢測(cè)與跟蹤的智能監(jiān)控系統(tǒng)[J];北京理工大學(xué)學(xué)報(bào);2004年11期

相關(guān)碩士學(xué)位論文 前1條

1 駱勇;復(fù)雜場(chǎng)景下的運(yùn)動(dòng)目標(biāo)檢測(cè)關(guān)鍵技術(shù)研究[D];上海交通大學(xué);2012年

，

本文編號(hào)：1390891