摘要
针对目前显著性检测算法在复杂多目标遥感图像中检测能力不足的问题,提出一种结合显著性检测和超像素分割的遥感信息提取算法。该算法通过GBVS(graph-based visual saliency)方法检测出原始影像中部分显著性较高的区域,然后利用SLIC(simple linear iterative clustering)方法分割显著区域,并修正显著区域边缘得到训练样本数据,进一步对训练样本进行统计学习,构造显著目标提取的阈值区间,最后实现对整幅超像素图像的显著目标提取。实验结果表明,该算法具有较高的准确率和召回率,能更加有效地检测出遥感图像中的显著目标,比目前主流的显著区域检测算法提取效果更好,可以很好地应用于具有明显显著区域的复杂多目标遥感图像信息提取中。
Due to the weakness of complex multi-target detection using some saliency detection algorithms in remote sensing images,this paper proposed an extraction algorithm based on saliency detection and superpixel segmentation. Firstly,it used graph-based visual saliency( GBVS) method to extract some high saliency regions correctly,then applied simple linear iterative clustering( SLIC) method to segment these saliency regions into some superpixels,and meanwhile amended part of the edge superpixels,and as the training sample. Secondly,by computing the statistical feature parameters of the training sample and constructing a reasonable statistic,it established a saliency objects extraction threshold range. Finally,it used the threshold range to extract the salient objects from the whole superpixels successfully. Compared with other saliency regions detection methods,experimental results show that the proposed algorithm has more higher precision and recall values and can detect the saliency objects more efficient,and can be better applied to extract remote sensing saliency information from the complex multitarget remote sensing images.
引文
[1]Rego L F G,Ueffing C,Vianna S B.Automatic land-cover classification derived from high-resolution IKONOS satellite imagery in the Urban Atlantic Forest of Rio de Janeiro,Brazil,by means of an objectoriented approach[M]//Applied Remote Sensing for Urban Planning,Governance and Sustainability.Berlin:Springer,2003:25-36.
[2]王荣.高分辨率遥感影像信息提取方法的研究[D].兰州:兰州交通大学,2013.
[3]Desimone R,Duncan J.Neural mechanisms of selective visual attention[J].Annual Review of Neuroscience,1995,18(1):193-222.
[4]温奇,李苓苓,刘庆杰,等.基于视觉显著性和图分割的高分辨率遥感影像中人工目标区域提取[J].测绘学报,2013,42(6):831-837.
[5]王鑫,王斌,张立明.基于图像显著性区域的遥感图像机场检测[J].计算机辅助设计与图形学学报,2012,24(3):336-344.
[6]孙晓帅,姚鸿勋.视觉注意与显著性计算综述[J].智能计算机与应用,2014,34(5):14-18.
[7]Achanta R,Hemami S,Estrada F,et al.Frequency-tuned salient region detection[C]//Proc of the IEEE Conference on Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE Press,2009:1597-1604.
[8]Itti L,Koch C,Niebur E.A model of saliency-based visual attention for rapid scene analysis[J].IEEE Trans on Pattern Analysis&MachineIntelligence,1998,20(11):1254-1259.
[9]Achanta R,Estrada F,Wils P,et al.Salient region detection and segmentation[C]//Proc of the Computer Vision Systems Conference.Berlin:Springer,2008:66-75.
[10]Hou Xiaodi,Zhang Liqing.Saliency detection:a spectral residual approach[C]//Proc of IEEE Conference on Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE Press,2007:1-8.
[11]Harel J,Koch C,Perona P.Graph-based visual saliency[C]//Advances in Neural Information Processing Systems.Vancouver:NIPS,2006:545-552.
[12]宋熙煜,周利莉,李中国,等.图像分割中的超像素方法研究综述[J].中国图象图形学报,2015,20(5):599-608.
[13]王春瑶,陈俊周,李炜.超像素分割算法研究综述[J].计算机应用研究,2014,31(1):6-12.
[14]张明哲,张红,王超,等.基于超像素分割和多方法融合的SAR图像变化检测方法[J].遥感技术与应用,2016,31(3):481-487.
[15]Shi Jianbo,Malik J.Normalized cuts and image segmentation[J].IEEE Trans on Pattern Analysis&Machine Intelligence,2000,22(8):888-905.
[16]Felzenszwalb P F,Huttenlocher D P.Efficient graph-based image segmentation[J].International Journal of Computer Vision,2004,59(2):167-181.
[17]Moore A P,Prince S J D,Warrell J,et al.Superpixel lattices[C]//Proc of IEEE Conference on Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE Press,2008:1-8.
[18]Vincent L,Soille P.Watersheds in digital spaces:an efficient algorithm based on immersion simulations[J].IEEE Trans on Pattern Analysis&Machine Intelligence,1991,13(6):583-598.
[19]Comaniciu D,Meer P,Member S.Mean-Shift:a robust approach toward feature space analysis[J].IEEE Trans on Pattern Analysis&Machine Intelligence,2002,24(5):603-619.
[20]Vedaldi A,Soatto S.Quick shift and kernel methods for mode seeking[C]//Proc of European Conference on Computer Vision.Berlin:Springer-Verlag,2008:705-718.
[21]Levinshtein A,Stere A,Kutulakos K N,et al.Turbo Pixels:fast superpixels using geometric flows[J].IEEE Trans on Pattern Analysis&Machine Intelligence,2009,31(12):2290-2297.
[22]Achanta R,Shaji A,Smith K,et al.SLIC superpixels,EPFL Technical Report 149300[R].Switzerland:EPFL,2010.
[23]Achanta R,Shaji A,Smith K,et al.SLIC superpixels compared to state-of-the-art superpixel methods[J].IEEE Trans on Pattern Analysis&Machine Intelligence,2012,34(11):2274-2282.
[24]汪成,陈文兵.基于SLIC超像素分割显著区域检测方法的研究[J].南京邮电大学学报:自然科学版,2016,36(1):89-93.
[25]Cheng Mingming,Zhang Guoxin,Mitra N J,et al.Global contrast based salient region detection[C]//Proc of the IEEE Conference on Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE Press,2011:409-416.
[26]王海罗,汪渤,周志强,等.基于超像素融合算法的显著区域检测[J].北京理工大学学报,2015,35(8):836-841.
[27]Sahli S,Lavigne D A,Sheng Yunlong.Saliency region selection in large aerial imagery using multiscale SLIC segmentation[C]//Proc of SPIE-The International Society for Optical Engineering.Bellingham:SPIE,2012.
[28]Sharma A,Ghosh J K.Saliency based segmentation of satellite images[C]//ISPKS Annals of the Photogrammetry,Remote Sensing and Spatial Information Sciences.Munich:ISPRS,2015:207-214.
[29]朱丹,王斌,张立明.基于直线邻近平行性和GBVS显著性的遥感图像机场目标检测[J].红外与毫米波学报,2015,34(3):375-384.
[30]胡志立,郭敏.基于SLIC的改进Grab Cut彩色图像快速分割[J].计算机工程与应用,2016,52(2):186-190,270.
[31]Otsu N.A threshold selection method from gray-level histograms[J].IEEE Trans on Systems Man&Cybernetics,1979,9(1):62-66.
[32]周涛,陆惠玲.数据挖掘中聚类算法研究进展[J].计算机工程与应用,2012,48(12):100-111.
[33]涂小坡.图像显著性算法和评价研究[D].南京:南京航空航天大学,2011.