一种鲁棒的运动目标阴影去除及修复算法
|
摘要
针对阴影区域对运动前景的干扰问题及归一化互相关系数阴影检测方法的局限性,提出一种归一化互相关系数融合对称交叉熵的阴影检测方法,并根据阴影区域和误检区域各自邻域的前景分布规律,提出一种基于阴影轮廓像素点邻域信息的阴影误检去除方法,对阴影去除形成的前景空洞及断裂进行恢复,最终得到完整准确的运动前景。前景保留率相比归一化互相关系数方法提高40.4%,相比SVM支持向量机方法提高6.3%,耗时约为SVM支持向量机方法的1/153。
For the problems of shadow region's interference to the motion foreground and the limitations of the normalized crosscorrelation shadow detection method,this paper proposes a shadow detection method that normalizes the cross-correlation coefficient fusion symmetry cross entropy,and according to the distribution law of the foreground of each neighborhood based on the shadow area and misdetection,a shadow misdetection removal method based on shadow contour pixel neighborhood information is proposed. The foreground voids and fractures formed by shadow removal are recovered,and the complete and accurate motion foreground is finally obtained. Compared with the normalized cross-correlation method and the SVM support vector machine method,the foreground reservation rate is respectively increased by 40. 4% and 6. 3%,and the time-consuming is approximately 1/153 of the SVM support vector machine method.
For the problems of shadow region's interference to the motion foreground and the limitations of the normalized crosscorrelation shadow detection method,this paper proposes a shadow detection method that normalizes the cross-correlation coefficient fusion symmetry cross entropy,and according to the distribution law of the foreground of each neighborhood based on the shadow area and misdetection,a shadow misdetection removal method based on shadow contour pixel neighborhood information is proposed. The foreground voids and fractures formed by shadow removal are recovered,and the complete and accurate motion foreground is finally obtained. Compared with the normalized cross-correlation method and the SVM support vector machine method,the foreground reservation rate is respectively increased by 40. 4% and 6. 3%,and the time-consuming is approximately 1/153 of the SVM support vector machine method.
引文
[1]Cucchiara R,Grana C,Piccardi M,et al.Improving shadow suppression in moving object detection with HSV color information[C]//Proceedings of IEEE Intelligent Transportation Systems.2002:334-339.
[2]Sanin A,Sanderson C,Lovell B C.Shadow detection:A survey and comparative evaluation of recent methods[J].Pattern Recognition,2012,45(4):1684-1695.
[3]高广珠,李忠武,余理富,等.归一化互相关系数在图像序列目标检测中的应用[J].计算机工程与科学,2005,27(3):38-40.
[4]Cucchiara R,Grana C,Piccardi M,et al.Detection moving objects,ghosts,and shadows in video streams[J].IEEE Transactions on Pattern Analysis&Machine Intelligence,2003,25(10):1337-1342.
[5]Bonini C,Ferrari G,Verzeletti S,et al.HSV-TK gene transfer into donor lymphocytes for control of allogeneic graft-versus-leukemia[J].Science,1997,276(5319):1719-1724.
[6]Mcfeely R,Glavin M,Jones E.Shadow identification for digital imagery using colour and texture cues[J].IET Image Processing,2012,6(2):148-159.
[7]朱周,路小波,卫朋,等.基于超像素和支持向量机的车辆阴影检测算法[J].东南大学学报(自然科学版),2015,45(3):443-447.
[8]Achanta R,Shaji A,Smith K,et al.SLIC superpixels compared to state-of-the-art superpixel methods[J].IEEE Transactions on Pattern Analysis&Machine Intelligence,2012,34(11):2274-2282.
[9]Wang She,Lu Huchuan,Yang Fan,et al.Superpixel tracking[C]//IEEE International Conference on Computer Vision.2011:1323-1330.
[10]Liu Ming-yu,Tuzel O,Ramalingam S,et al.Entropy rate superpixel segmentation[C]//Proceedings of 2011 IEEE Conference on Computer Vision and Pattern Recognition.2011:2097-2104.
[11]Adankon M M,Cheriet M.Support vector machine[J].Computer Science,2002,1(4):1-28.
[12]Baesens B,Viaene S,Van Gestel T,et al.Least squares support vector machine classifiers:An empirical evaluation[J].Access&Download Statistics,2000,2376(3):1-16.
[13]Furey T S,Cristianini N,Duffy N,et al.Support vector machine classification and validation of cancer tissue samples using microarray expression data[J].Bioinformatics,2000,16(10):906-914.
[14]Lewis J P.Fast normalized cross-correlation[J].Circuits Systems&Signal Processing,1995,82(2):144-156.
[15]Lin Jianhua.Divergence measures based on Shannon entropy[J].IEEE Transactions on Information Theory,1991,37(1):145-151.
[16]Boer P T D,Kroese D P,Mannor S,et al.A tutorial on the cross-entropy method[J].Annals of Operations Research,2005,134(1):19-67.
[17]Yu Jun,Zhou Jun,Yin Xuefeng.Two-dimensional tsallis symmetric cross entropy image threshold segmentation[C]//Proceedings of 2012 4th International Symposium on Information Science and Engineering.2012:362-366.
[2]Sanin A,Sanderson C,Lovell B C.Shadow detection:A survey and comparative evaluation of recent methods[J].Pattern Recognition,2012,45(4):1684-1695.
[3]高广珠,李忠武,余理富,等.归一化互相关系数在图像序列目标检测中的应用[J].计算机工程与科学,2005,27(3):38-40.
[4]Cucchiara R,Grana C,Piccardi M,et al.Detection moving objects,ghosts,and shadows in video streams[J].IEEE Transactions on Pattern Analysis&Machine Intelligence,2003,25(10):1337-1342.
[5]Bonini C,Ferrari G,Verzeletti S,et al.HSV-TK gene transfer into donor lymphocytes for control of allogeneic graft-versus-leukemia[J].Science,1997,276(5319):1719-1724.
[6]Mcfeely R,Glavin M,Jones E.Shadow identification for digital imagery using colour and texture cues[J].IET Image Processing,2012,6(2):148-159.
[7]朱周,路小波,卫朋,等.基于超像素和支持向量机的车辆阴影检测算法[J].东南大学学报(自然科学版),2015,45(3):443-447.
[8]Achanta R,Shaji A,Smith K,et al.SLIC superpixels compared to state-of-the-art superpixel methods[J].IEEE Transactions on Pattern Analysis&Machine Intelligence,2012,34(11):2274-2282.
[9]Wang She,Lu Huchuan,Yang Fan,et al.Superpixel tracking[C]//IEEE International Conference on Computer Vision.2011:1323-1330.
[10]Liu Ming-yu,Tuzel O,Ramalingam S,et al.Entropy rate superpixel segmentation[C]//Proceedings of 2011 IEEE Conference on Computer Vision and Pattern Recognition.2011:2097-2104.
[11]Adankon M M,Cheriet M.Support vector machine[J].Computer Science,2002,1(4):1-28.
[12]Baesens B,Viaene S,Van Gestel T,et al.Least squares support vector machine classifiers:An empirical evaluation[J].Access&Download Statistics,2000,2376(3):1-16.
[13]Furey T S,Cristianini N,Duffy N,et al.Support vector machine classification and validation of cancer tissue samples using microarray expression data[J].Bioinformatics,2000,16(10):906-914.
[14]Lewis J P.Fast normalized cross-correlation[J].Circuits Systems&Signal Processing,1995,82(2):144-156.
[15]Lin Jianhua.Divergence measures based on Shannon entropy[J].IEEE Transactions on Information Theory,1991,37(1):145-151.
[16]Boer P T D,Kroese D P,Mannor S,et al.A tutorial on the cross-entropy method[J].Annals of Operations Research,2005,134(1):19-67.
[17]Yu Jun,Zhou Jun,Yin Xuefeng.Two-dimensional tsallis symmetric cross entropy image threshold segmentation[C]//Proceedings of 2012 4th International Symposium on Information Science and Engineering.2012:362-366.