結(jié)合雙向光流約束的特征點(diǎn)匹配車輛跟蹤方法
發(fā)布時(shí)間:2018-11-15 13:51
【摘要】:針對(duì)復(fù)雜交通場(chǎng)景中動(dòng)態(tài)光照變化、目標(biāo)尺度變化和部分遮擋等因素帶來(lái)的影響,提出了一種基于特征點(diǎn)的穩(wěn)定可靠的車輛跟蹤方法.針對(duì)運(yùn)動(dòng)車輛高速行駛時(shí)具有較大幀間運(yùn)動(dòng)的特點(diǎn),構(gòu)造KLT算法的金字塔模型,根據(jù)前向和后向跟蹤偏移量,對(duì)穩(wěn)定性較差的特征點(diǎn)進(jìn)行剔除.同時(shí),采用SURF特征匹配算法對(duì)目標(biāo)特征點(diǎn)集進(jìn)行更新和校正.最后,利用特征點(diǎn)之間的位置信息,確定目標(biāo)的尺度和旋轉(zhuǎn)變化因子,從而實(shí)現(xiàn)當(dāng)前幀中目標(biāo)區(qū)域的定位.實(shí)驗(yàn)結(jié)果表明,提出的車輛跟蹤方法可以有效地解決復(fù)雜場(chǎng)景中目標(biāo)形變和部分遮擋等問(wèn)題,對(duì)尺度和旋轉(zhuǎn)變化也具有較強(qiáng)的魯棒性.
[Abstract]:A stable and reliable vehicle tracking method based on feature points is proposed to deal with the effects of dynamic illumination variation target scale change and partial occlusion in complex traffic scenarios. In view of the large inter-frame motion of moving vehicles at high speed, the pyramid model of KLT algorithm is constructed, and the feature points with poor stability are eliminated according to the forward and backward tracking offsets. At the same time, the target feature set is updated and corrected by SURF feature matching algorithm. Finally, using the position information between the feature points, the scale of the target and the rotation change factor are determined, so that the location of the target region in the current frame can be realized. Experimental results show that the proposed vehicle tracking method can effectively solve the problems of target deformation and partial occlusion in complex scenes, and is robust to scale and rotation changes.
【作者單位】: 西安石油大學(xué)計(jì)算機(jī)學(xué)院;
【基金】:國(guó)家自然科學(xué)基金(61572083)~~
【分類號(hào)】:TP391.41
[Abstract]:A stable and reliable vehicle tracking method based on feature points is proposed to deal with the effects of dynamic illumination variation target scale change and partial occlusion in complex traffic scenarios. In view of the large inter-frame motion of moving vehicles at high speed, the pyramid model of KLT algorithm is constructed, and the feature points with poor stability are eliminated according to the forward and backward tracking offsets. At the same time, the target feature set is updated and corrected by SURF feature matching algorithm. Finally, using the position information between the feature points, the scale of the target and the rotation change factor are determined, so that the location of the target region in the current frame can be realized. Experimental results show that the proposed vehicle tracking method can effectively solve the problems of target deformation and partial occlusion in complex scenes, and is robust to scale and rotation changes.
【作者單位】: 西安石油大學(xué)計(jì)算機(jī)學(xué)院;
【基金】:國(guó)家自然科學(xué)基金(61572083)~~
【分類號(hào)】:TP391.41
【相似文獻(xiàn)】
相關(guān)期刊論文 前10條
1 卞建勇;徐建閩;裴海龍;;基于強(qiáng)化學(xué)習(xí)的視頻車輛跟蹤[J];華南理工大學(xué)學(xué)報(bào)(自然科學(xué)版);2008年10期
2 張曉媚;陳偉海;劉敬猛;趙志文;;基于柔性曲桿的車輛跟蹤算法設(shè)計(jì)與實(shí)現(xiàn)[J];北京航空航天大學(xué)學(xué)報(bào);2011年07期
3 甘玲;潘小雷;;一種應(yīng)用于交通環(huán)境中的運(yùn)動(dòng)車輛跟蹤方法[J];重慶郵電大學(xué)學(xué)報(bào)(自然科學(xué)版);2013年03期
4 郭鋒;王秉政;楊晨暉;;復(fù)雜背景下車輛跟蹤的改進(jìn)算法及逆行檢測(cè)[J];圖學(xué)學(xué)報(bào);2013年04期
5 魏玉強(qiáng);王成儒;;多車輛跟蹤時(shí)分割粘連車輛的方法[J];電視技術(shù);2009年11期
6 王文龍;李清泉;;基于蒙特卡羅算法的車輛跟蹤方法[J];測(cè)繪學(xué)報(bào);2011年02期
7 李t,
本文編號(hào):2333483
本文鏈接:http://sikaile.net/kejilunwen/ruanjiangongchenglunwen/2333483.html
最近更新
教材專著