發(fā)布時(shí)間：2018-04-30 00:25

  本文選題：動(dòng)態(tài)背景 + 攝像機(jī)運(yùn)動(dòng)　； 參考：《延邊大學(xué)》2017年碩士論文

【摘要】：隨著信息技術(shù)的發(fā)展,視頻文件的數(shù)量飛速增長(zhǎng)。如何在海量的視頻文件中檢索出關(guān)鍵事件和關(guān)鍵目標(biāo)具有重要的研究意義和廣泛的應(yīng)用價(jià)值。尤其是在冗余信息較多的交通監(jiān)控和安防監(jiān)控視頻中進(jìn)行事件和目標(biāo)檢測(cè)更具有重要的現(xiàn)實(shí)意義。但是,如何在復(fù)雜場(chǎng)景視頻中實(shí)現(xiàn)快速準(zhǔn)確的目標(biāo)檢測(cè)與行人檢測(cè),仍是目前運(yùn)動(dòng)目標(biāo)檢測(cè)領(lǐng)域亟待解決的問(wèn)題。光流法是一種有效檢測(cè)運(yùn)動(dòng)物體與運(yùn)動(dòng)行人的算法,不需任何先驗(yàn)知識(shí),對(duì)背景模型不存在依賴,在運(yùn)動(dòng)目標(biāo)檢測(cè)領(lǐng)域有著廣泛的應(yīng)用。但光流法存在對(duì)噪聲敏感、算法復(fù)雜度較高等缺陷,本學(xué)位論文針對(duì)光流法對(duì)噪聲敏感和實(shí)時(shí)性差的缺點(diǎn)進(jìn)行改進(jìn),提出了一種改進(jìn)的光流法并將其應(yīng)用于運(yùn)動(dòng)目標(biāo)檢測(cè)。本文主要研究?jī)?nèi)容如下:首先,針對(duì)攝像機(jī)運(yùn)動(dòng)造成的全局動(dòng)態(tài)背景,采用改進(jìn)的灰度投影法進(jìn)行全局運(yùn)動(dòng)補(bǔ)償。針對(duì)灰度投影法存在累積誤差等缺陷,提出隔三幀更換一次參考幀的方法,減少了由于選擇固定參考幀導(dǎo)致的累積計(jì)算誤差;針對(duì)灰度投影法的投影區(qū)域存在運(yùn)動(dòng)目標(biāo)時(shí)計(jì)算誤差較大的問(wèn)題,本學(xué)位論文使用視頻圖像幀的邊角地帶作為投影區(qū)域來(lái)計(jì)算運(yùn)動(dòng)矢量,大大減少了運(yùn)動(dòng)目標(biāo)對(duì)運(yùn)動(dòng)補(bǔ)償?shù)挠绊�。其�?提出了一種基于梯度閾值和特征抑制的光流法進(jìn)行運(yùn)動(dòng)目標(biāo)檢測(cè)。將LK光流法和HS光流法進(jìn)行結(jié)合,對(duì)光流約束方程進(jìn)行改進(jìn),對(duì)梯度較大的像素點(diǎn)采用亮度約束,對(duì)梯度較小的像素點(diǎn)采用全局平滑約束,以確保光流約束方程的適用性。特征抑制作為輔助判斷有效光流點(diǎn)的手段,在運(yùn)動(dòng)目標(biāo)檢測(cè)算法中對(duì)噪聲和局部動(dòng)態(tài)背景進(jìn)行了有效地抑制。最后,在運(yùn)動(dòng)目標(biāo)檢測(cè)算法基礎(chǔ)上,采用改進(jìn)的形狀復(fù)雜度特征,對(duì)視頻中的運(yùn)動(dòng)目標(biāo)進(jìn)行分類。通過(guò)比較不同類型運(yùn)動(dòng)目標(biāo)的形狀復(fù)雜度取值范圍,確定能夠區(qū)分行人和其他類型運(yùn)動(dòng)目標(biāo)的閾值,從而有效地檢測(cè)出視頻中的行人目標(biāo)。測(cè)試結(jié)果表明,本文提出的動(dòng)態(tài)背景下行人檢測(cè)算法與其他算法相比,準(zhǔn)確率和穩(wěn)定性都有明顯的提高。此外,本文提出的算法能夠?qū)z像機(jī)運(yùn)動(dòng)、局部動(dòng)態(tài)背景具有良好的魯棒性,為下一步工作提供了良好的基礎(chǔ)。
[Abstract]:With the development of information technology, the number of video files is increasing rapidly. How to retrieve the key events and key objects from a large number of video files has important research significance and wide application value. Especially, the detection of events and targets in traffic surveillance and security surveillance video with more redundant information is of great practical significance. However, how to achieve fast and accurate target detection and pedestrian detection in complex scene video is still an urgent problem in the field of moving target detection. Optical flow method is an effective algorithm for detecting moving objects and pedestrians. It does not require any prior knowledge and does not depend on the background model. It is widely used in the field of moving target detection. However, the optical flow method is sensitive to noise and the complexity of the algorithm is high. In this dissertation, an improved optical flow method is proposed and applied to the detection of moving targets. In view of the shortcomings of the optical flow method for noise sensitivity and poor real-time performance, an improved optical flow method is proposed and applied to the detection of moving targets. The main contents of this paper are as follows: firstly, an improved gray projection method is used to compensate the global motion of the camera. Aiming at the defects of gray projection method, such as accumulative error, the method of replacing reference frame once every three frames is put forward, which reduces the accumulated calculation error caused by selecting fixed reference frame. In view of the problem that the projection region of gray projection method has the problem of large calculation error when moving object, this thesis uses the edge zone of the video frame as the projection region to calculate the motion vector. The effect of moving target on motion compensation is greatly reduced. Secondly, an optical flow method based on gradient threshold and feature suppression is proposed to detect moving targets. The LK optical flow method and HS optical flow method are combined to improve the optical flow constraint equation. The brightness constraint is applied to the pixel points with large gradient and the global smoothing constraint is applied to the smaller gradient pixel points to ensure the applicability of the optical flow constraint equation. Feature suppression is used as an auxiliary method to judge the effective optical flow point. The noise and local dynamic background are effectively suppressed in the moving target detection algorithm. Finally, based on the moving target detection algorithm, the improved shape complexity feature is used to classify the moving target in video. By comparing the range of shape complexity of different types of moving objects, the threshold can be determined to distinguish pedestrians from other types of moving objects, so as to effectively detect pedestrian targets in video. The test results show that the proposed pedestrian detection algorithm in dynamic background is more accurate and more stable than other algorithms. In addition, the proposed algorithm is robust to camera motion and local dynamic background, and provides a good basis for further work.
【學(xué)位授予單位】：延邊大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.41

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 賈云偉,王聃,楊金勇;復(fù)雜動(dòng)態(tài)背景中的人臉檢測(cè)[J];黑龍江科技學(xué)院學(xué)報(bào);2005年03期

2 李倩;李金屏;;動(dòng)態(tài)背景下具有偽裝色的運(yùn)動(dòng)目標(biāo)檢測(cè)[J];濟(jì)南大學(xué)學(xué)報(bào)(自然科學(xué)版);2011年02期

3 洪宇,高廣珠,余理富,何智勇;一種基于動(dòng)態(tài)背景的運(yùn)動(dòng)車輛跟蹤算法[J];計(jì)算機(jī)應(yīng)用研究;2004年06期

4 徐克虎;王天召;陳金玉;張波;;動(dòng)態(tài)背景下的運(yùn)動(dòng)目標(biāo)檢測(cè)定位算法研究[J];計(jì)算機(jī)測(cè)量與控制;2013年12期

5 宋春玉;;動(dòng)態(tài)背景下圖像的分割[J];計(jì)算機(jī)應(yīng)用與軟件;2014年04期

6 王天召;徐克虎;黃大山;;動(dòng)態(tài)背景下的運(yùn)動(dòng)目標(biāo)檢測(cè)[J];計(jì)算機(jī)技術(shù)與發(fā)展;2013年07期

7 毛文;朵天林;;電視節(jié)目片頭中常用動(dòng)態(tài)背景制作解析[J];中國(guó)有線電視;2007年Z2期

8 劉英霞;賀長(zhǎng)偉;王欣;;基于貝葉斯模型的動(dòng)態(tài)背景檢測(cè)[J];系統(tǒng)仿真學(xué)報(bào);2007年21期

9 宦若虹;王浙滬;唐曉梅;陳慶章;;動(dòng)態(tài)背景下基于粒子濾波的運(yùn)動(dòng)目標(biāo)跟蹤方法[J];計(jì)算機(jī)應(yīng)用與軟件;2011年05期

10 崔智高;李艾華;姜柯;;雙目協(xié)同動(dòng)態(tài)背景運(yùn)動(dòng)分離方法[J];紅外與激光工程;2013年S1期

相關(guān)會(huì)議論文 前2條

1 白建華;李翠華;吳琦穎;;基于動(dòng)態(tài)背景的運(yùn)動(dòng)車輛檢測(cè)算法[A];第十二屆全國(guó)圖象圖形學(xué)學(xué)術(shù)會(huì)議論文集[C];2005年

2 李金屏;李倩;;基于Laws能量和變化頻次的偽裝色移動(dòng)目標(biāo)檢測(cè)研究[A];第十五屆全國(guó)圖象圖形學(xué)學(xué)術(shù)會(huì)議論文集[C];2010年

相關(guān)重要報(bào)紙文章 前1條

1 MXC;讓網(wǎng)頁(yè)背景飄起來(lái)[N];電腦報(bào);2005年

相關(guān)碩士學(xué)位論文 前10條

1 趙坤;基于模型共享機(jī)制的前景檢測(cè)算法[D];北京理工大學(xué);2015年

2 馮艷;動(dòng)態(tài)背景下基于SIFT特征匹配的目標(biāo)檢測(cè)算法[D];西安電子科技大學(xué);2014年

3 于少波;面向虛實(shí)合作的動(dòng)態(tài)背景分離技術(shù)研究[D];北方工業(yè)大學(xué);2016年

4 韓立媛;移動(dòng)機(jī)器人運(yùn)動(dòng)圖像捕捉、識(shí)別與跟蹤[D];燕山大學(xué);2016年

5 李恒年;動(dòng)態(tài)背景下的視頻目標(biāo)檢測(cè)與跟蹤研究[D];鄭州大學(xué);2016年

6 蘇杭;動(dòng)態(tài)背景多目標(biāo)檢測(cè)與跟蹤技術(shù)研究[D];電子科技大學(xué);2016年

7 張?jiān)沸?動(dòng)態(tài)背景下運(yùn)動(dòng)目標(biāo)的檢測(cè)與跟蹤[D];西安電子科技大學(xué);2015年

8 王芬芬;動(dòng)態(tài)背景下的視頻目標(biāo)跟蹤[D];杭州電子科技大學(xué);2015年

9 劉永鵬;高清視頻監(jiān)控系統(tǒng)動(dòng)態(tài)背景去除研究[D];天津理工大學(xué);2014年

10 范五東;動(dòng)態(tài)背景下的多目標(biāo)檢測(cè)與跟蹤的研究[D];重慶大學(xué);2012年

，

本文編號(hào)：1822242