【摘要】：視頻車輛檢測(cè)是一種在視頻序列中提取運(yùn)動(dòng)車輛對(duì)象的技術(shù),其廣泛運(yùn)用于視頻監(jiān)控、智能交通檢測(cè)等系統(tǒng)中。由于運(yùn)動(dòng)車輛檢測(cè)技術(shù),特別是相機(jī)運(yùn)動(dòng)條件下的視頻車輛檢測(cè),具有復(fù)雜性、多變性等特點(diǎn),該技術(shù)仍處于起步階段,需要不斷的研究并加以改進(jìn)。為了在相機(jī)運(yùn)動(dòng)條件下準(zhǔn)確檢測(cè)出車輛,本文圍繞著以下幾個(gè)方面來(lái)開展研究工作:(1)對(duì)本文涉及到的理論基礎(chǔ)知識(shí)進(jìn)行學(xué)習(xí)研究�？偨Y(jié)概括涉及視頻圖像處理,尤其是視頻目標(biāo)檢測(cè)等相關(guān)國(guó)內(nèi)外文獻(xiàn)資料;篩選并學(xué)習(xí)基于相機(jī)運(yùn)動(dòng)條件下視頻車輛檢測(cè)的相關(guān)知識(shí),重點(diǎn)是近幾年關(guān)于動(dòng)攝像頭下視頻車輛檢測(cè)研究成果;針對(duì)動(dòng)攝像頭視頻車輛檢測(cè)研究現(xiàn)狀,并結(jié)合自身知識(shí)積累,構(gòu)建本文算法框架。(2)對(duì)視頻車輛檢測(cè)方法進(jìn)行分析研究。視頻車輛檢測(cè)方法可根據(jù)攝像頭是否運(yùn)動(dòng)進(jìn)行分類,本論文簡(jiǎn)單介紹了靜攝像頭下視頻運(yùn)動(dòng)車輛檢測(cè)方法,具體介紹相機(jī)運(yùn)動(dòng)條件下的視頻車輛檢測(cè)方法。(3)對(duì)相機(jī)運(yùn)動(dòng)條件下的視頻車輛檢測(cè)進(jìn)行算法設(shè)計(jì)。算法分為四部分。第一部分全局運(yùn)動(dòng)估計(jì)與補(bǔ)償算法,并對(duì)基礎(chǔ)算法進(jìn)行說(shuō)明,在這基礎(chǔ)上,通過(guò)分析不同的全局運(yùn)動(dòng)估計(jì)方法的優(yōu)缺點(diǎn),提出使用六參數(shù)仿射變換模型,估計(jì)仿射變換參數(shù),然后對(duì)仿射變換后的圖像進(jìn)行補(bǔ)償。第二部分高斯差分算法,改進(jìn)了運(yùn)動(dòng)補(bǔ)償后的差分步驟,以便獲得更好的檢測(cè)效果。第三部分非參數(shù)核密度估計(jì)算法,以此對(duì)檢測(cè)進(jìn)行優(yōu)化。第四部分使用矩形框?qū)z測(cè)出的車輛進(jìn)行目標(biāo)定位。(4)對(duì)本論文的算法進(jìn)行實(shí)驗(yàn)驗(yàn)證。利用VS2010和Matlab軟件平臺(tái),并結(jié)合OpenCV開源庫(kù),編寫本算法的實(shí)驗(yàn)仿真程序,以動(dòng)攝像頭拍攝的視頻為輸入,進(jìn)行實(shí)驗(yàn)。實(shí)驗(yàn)驗(yàn)證過(guò)程分為兩部分:第一部分實(shí)驗(yàn)用來(lái)驗(yàn)證本文算法的功能實(shí)現(xiàn);第二部分實(shí)驗(yàn)是對(duì)比試驗(yàn),用來(lái)驗(yàn)證本文算法的準(zhǔn)確性和高魯棒性。研究創(chuàng)新有兩點(diǎn):(1)為了減小攝像頭運(yùn)動(dòng)對(duì)視頻中運(yùn)動(dòng)目標(biāo)檢測(cè)的影響,提高運(yùn)動(dòng)估計(jì)的準(zhǔn)確度,在進(jìn)行仿射運(yùn)動(dòng)估計(jì)時(shí),對(duì)目標(biāo)幀的前后幀采用不同的仿射變換矩陣,計(jì)算量降低了,仿射變換效果提高了。(2)使用非參數(shù)核密度估計(jì)對(duì)得到的檢測(cè)目標(biāo)進(jìn)行優(yōu)化,減少各環(huán)節(jié)帶來(lái)的目標(biāo)空洞問題和噪聲靈敏度問題的影響。研究的不足是,仿射運(yùn)動(dòng)估計(jì)參數(shù)的獲取,計(jì)算量較大,時(shí)間較長(zhǎng),仿射運(yùn)動(dòng)估計(jì)參數(shù)的獲取速度仍有待提高,算法的天氣情況適應(yīng)性,需要進(jìn)一步研究及改進(jìn)。
[Abstract]:Video vehicle detection is a technique for extracting moving vehicle objects in video sequence. It is widely used in video surveillance and intelligent traffic detection systems. Because of the characteristics of vehicle detection technology, especially video vehicle detection under camera motion conditions, it is still in its infancy. In order to detect the vehicle accurately under the camera movement, this paper focuses on the following aspects: (1) study and study the theoretical basic knowledge involved in this article. Summarize and summarize the related domestic and foreign documents related to video image processing, especially the visual frequency target detection. And learn the related knowledge of video vehicle detection based on camera motion. The focus is on the research results of video vehicle detection in recent years. In view of the current situation of video vehicle detection research in mobile camera, and combining with the knowledge accumulation, this paper constructs the algorithm framework. (2) video vehicle detection methods are analyzed and studied. The vehicle detection method can be classified according to the motion of the camera. This paper briefly introduces the video moving vehicle detection method under the static camera and introduces the video vehicle detection method under the camera motion condition. (3) the algorithm is designed for the video vehicle detection under the camera motion condition. The algorithm is divided into four parts. The first part is global. The motion estimation and compensation algorithm and the basic algorithm are explained. On this basis, by analyzing the advantages and disadvantages of different global motion estimation methods, the six parameter affine transform model is used to estimate the affine transformation parameters, and then the affine transform image is compensated. The second part of the Gauss difference algorithm is improved after the motion compensation. The third part of the non parametric kernel density estimation algorithm is used to optimize the detection. The fourth part uses the rectangle frame to target the detection of the vehicle. (4) the experimental verification of the algorithm in this paper. Using the VS2010 and Matlab software platform, and combining the OpenCV open source library, write this The experimental simulation program of the algorithm is carried out with the video taken by the moving camera as input, and the experiment is divided into two parts: the first part is used to verify the function realization of the algorithm. The second part of the experiment is a contrast test, which is used to verify the accuracy and robustness of the algorithm. (1) to reduce the perturbation. Like the effect of head motion on moving target detection in video, the accuracy of motion estimation is improved. When carrying out the affine motion estimation, different affine transform matrices are used for the front and back frames of the target frame. The amount of computation is reduced and the effect of affine transformation is improved. (2) the detection targets are optimized by using non parametric kernel density estimation, and each of them is reduced. The problem of the target cavitation and the noise sensitivity caused by the link. The shortage of the research is that the parameters of the affine motion estimation are obtained, the computation is large and the time is long. The acquisition speed of the affine motion estimation parameters remains to be improved, and the weather conditions of the algorithm are adaptable, and further research and improvement are needed.
【學(xué)位授予單位】：山東師范大學(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10條

1 高勇鋼;;改進(jìn)幀差法和背景差法的多目標(biāo)跟蹤[J];巢湖學(xué)院學(xué)報(bào);2013年06期

2 王斌;何中市;伍星;賈媛媛;;基于高斯金字塔的圖像運(yùn)動(dòng)估計(jì)算法[J];計(jì)算機(jī)工程與應(yīng)用;2015年07期

3 欒慶磊;趙為松;;動(dòng)背景下幀差分法與邊緣信息融合的目標(biāo)檢測(cè)算法[J];光電工程;2011年10期

4 劉彬;嚴(yán)京旗;施鵬飛;;高斯差分的AdaBoost車牌定位方法[J];智能系統(tǒng)學(xué)報(bào);2010年06期

5 孫劍芬;;基于高斯核密度估計(jì)的運(yùn)動(dòng)目標(biāo)檢測(cè)新方法[J];計(jì)算機(jī)技術(shù)與發(fā)展;2010年08期

6 付青青;張春海;;高斯模糊圖像的復(fù)原處理與研究[J];長(zhǎng)江大學(xué)學(xué)報(bào)(自然科學(xué)版)理工卷;2010年02期

7 盛旭鋒;朱方文;李校祖;莊俊;;基于三幀時(shí)間差分法的獨(dú)居老人運(yùn)動(dòng)檢測(cè)[J];計(jì)算機(jī)工程與應(yīng)用;2010年13期

8 王久陽(yáng);何廣軍;朱福軍;;基于線性高斯濾波的多傳感器管理算法[J];探測(cè)與控制學(xué)報(bào);2009年05期

9 李寧;黃山;張先震;李秀君;;基于背景差分的人體運(yùn)動(dòng)檢測(cè)[J];微計(jì)算機(jī)信息;2009年21期

10 鄭雅羽;田翔;陳耀武;;基于運(yùn)動(dòng)矢量對(duì)消和差分原理的快速全局運(yùn)動(dòng)估計(jì)[J];電子與信息學(xué)報(bào);2009年04期

相關(guān)碩士學(xué)位論文 前2條

1 張金花;運(yùn)動(dòng)目標(biāo)跟蹤算法的研究[D];蘭州大學(xué);2009年

2 方穎;運(yùn)動(dòng)目標(biāo)的檢測(cè)、定位與跟蹤研究[D];山東大學(xué);2008年

，

本文編號(hào)：2166930