Edge detection algorithm of moving vehicle based on sequential image motion segmentation
-
摘要: 为了准确获得图像感兴趣区中运动车辆的形状特征, 提出了一种新的车辆边界轮廓提取算法。利用连续3帧图像, 对包含同一运动车辆的图像感兴趣区进行光流场分割, 以获取目标运动区域, 通过平移运动区域的左、右边界获得正确的车辆区域及其封闭边界轮廓, 通过放大运动矢量计算公式的阈值来提高其运行效率。试验结果表明: 该算法可从具有复杂自然场景的图像序列中检测出完整的运动车辆边界轮廓, 检测正确率在95%以上。Abstract: To obtain the shape feature of moving vehicle in interested region, a novel algorithm of edge detection was proposed.After inputting three consecutive frames with interested vehicle and stationary complex background, the motion region of vehicle was detected by image motion segmentation and median filtering operator.The complete region and enclosed contour edge of vehicle were gotten by displacing the left edge or the right one of the motion region in horizontal detection.The run time of the algorithm was reduced by enlarging the threshold of iterative equation for motion vectors.Experiment result shows that in a 30 min sequence of urban traffic, 95% of vehicles are accurately detected in complex background.
-
图 9 运动区域左、右边界经位置平移后的结果
Figure 9. Results after displacing the left edge and right one of motion region
图 10 不同运动车型边界提取结果
Figure 10. Edge detection results of moving vehicles with different types
-
[1] 张泽旭, 李金宗, 李宁宁. 基于光流场分割和Canny边缘提取融合算法的运动目标检测[J]. 电子学报, 2003, 31(9): 1299-1302. doi: 10.3321/j.issn:0372-2112.2003.09.005ZHANG Ze-xu, LI Jin-zong, LI Ning-ning. Detection ofmoving object using a fusion method based on segmentation ofoptical flow field and edge extracted by Canny s operator[J]. Acta Electronica Sinica, 2003, 31(9): 1299-1302. (in Chinese) doi: 10.3321/j.issn:0372-2112.2003.09.005 [2] TSUCHIYA H, UEMURA N, ITOHS. Optical flowdetectionusing segmentation and MDL criterion[J]. Systems andComputers in Japan, 2004, 35(9): 51-59. doi: 10.1002/scj.10335 [3] HARITAOGLU L, HARWOOD D, DAVIS L S. W4: real-ti me surveillance of people and their activities[J]. IEEETransactions on Pattern Analysis and Machine Intelligence, 2000, 22(8): 809-830. doi: 10.1109/34.868683 [4] WEI Zhi-qiang, JI Xiao-peng, WANG Peng. Real-ti me movingobject detection for video monitoring systems[J]. Journal ofSystems Engineering and Electronics, 2006, 17(4): 731-736. doi: 10.1016/S1004-4132(07)60007-3 [5] 屈有山, 田维坚, 李英才. 基于并行隔帧差分流场与灰度分析综合算法的运动目标检测[J]. 光子学报, 2003, 32(2): 182-186. https://www.cnki.com.cn/Article/CJFDTOTAL-GZXB200302014.htmQU You-shan, TIAN Wei-jian, LI Ying-cai. The moving targetdetecting based on the parallelable discontinuous frame differenceoptical field integrated with gray intensity analysis[J]. ActaPhotonica Sinica, 2003, 32(2): 182-186. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-GZXB200302014.htm [6] TEKALP A M. Digital Video Processing[M]. NJ: PrenticeHall, 1995. [7] 肖旺新, 张雪, 黄卫. 视频交通图像自适应阈值边缘检测[J]. 交通运输工程学报, 2003, 3(4): 104-107. http://transport.chd.edu.cn/article/id/200304020XI AO Wang-xin, ZHANG Xue, HUANG Wei. Adaptivethresholds edge detection of traffic i mage[J]. Journal of Trafficand Transportation Engineering, 2003, 3(4): 104-107. (in Chinese) http://transport.chd.edu.cn/article/id/200304020 [8] 初秀民, 严新平, 毛喆. 道路标志自动分类方法[J]. 交通运输工程学报, 2006, 6(4): 91-95, 111. http://transport.chd.edu.cn/article/id/200604021CHU Xiu-min, YAN Xin-ping, MAO Zhe. Automatic classifymethod of traffic sign[J]. Journal of Traffic and TransportationEngineering, 2006, 6(4): 91-95, 111. (in Chinese) http://transport.chd.edu.cn/article/id/200604021 [9] KUO C M, HSIEHC H, HUANG Y R. Automatic extractionof moving objects for head-shoulder video sequences[J]. Journal of Visual Communication and I mage Representation2005, 16: 68-92. [10] 连静, 王珂, 杨兆升. 结合信号配准技术的小波多尺度车辆边缘检测方法[J]. 中国公路学报, 2007, 20(5): 95-100. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL200705018.htmLIANJing, WANG Ke, YANGZhao-sheng. Method of waveletmulti-scale vehicle edge detection combined with signal regis-tration technology[J]. China Journal of Highway andTransport, 2007, 20(5): 95-100. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL200705018.htm -
下载:
点击查看大图
图(10) / 表(1)
计量
- 文章访问数: 338
- HTML全文浏览量: 147
- PDF下载量: 334
- 被引次数: 0
