Remote sensing image segmentation based on Parzen window density estimation of super-pixels
-
摘要: 图像分割是高分辨率遥感图像分析中的关键步骤,对信息提取精度起到重要作用。为提高传统基于像素的遥感图像分割算法性能,提出一种在超像素上进行Parzen窗密度估计的分割算法。包括超像素初始分割、特征测量、密度估计并重新聚类3个主要步骤。在超像素初始分割阶段,采用简单线性迭代聚类算法将图像进行超像素粗分割,并将每个超像素块标记为图结构中的一个顶点; 然后测量每个超像素块的Gabor纹理特征,构建高维特征向量并计算纹理间的相似度,作为图中连接2个顶点的边的权值,并在该图的最小生成树上计算2个顶点之间的距离; 接着将此距离用于Parzen窗,估计每个顶点的密度,并重新聚类得到最终结果。采用多幅多光谱高分辨遥感图像验证本文提出的算法,基于目视判别以及基于准确率和召回率的定量评价,将该方法与其他分割算法的结果进行比较,验证了提出算法的有效性。Abstract: Image segmentation is a key step in the analysis of high-resolution remote sensing images and plays an important role in improving information extraction accuracy. To improve the performance of traditional pixel-based image segmentation methods, this study proposed a new algorithm based on Parzen window density estimation of super-pixel blocks. The new algorithm includes three main steps, namely super-pixel initial segmentation, feature measurement, and density estimation and re-clustering. In the first step, an image is coarsely divided using the simple linear iterative clustering (SLIC) algorithm, and each super-pixel block is marked as a node in the graph structure of the image. In the second step, the Gabor texture features of each super-pixel block are measured to construct high-dimension feature vectors. Meanwhile, the similarity of the image textures is calculated as the weight of the edge connecting two nodes in the graph. Then, the distance between the two nodes is calculated on the minimum spanning tree (MST) of the graph. In the third step, the calculated distance is used for Parzen window density estimation of each node, and re-clustering of the density values is conducted to obtain the final results. In the experiments, multiple multispectral high-resolution remote sensing images were adopted to verify the algorithm proposed in this study. Using visual discrimination and the quantitative evaluation based on precesion rate and recall rate, the segmentation results of the algorithm proposed in this study were compared with those of other algorithms. The experiments verified that the algorithm proposed in this study is effective.
-
-
[1] Chehata N, Orny C, Boukir S, et al. Object-based change detection in wind storm-damaged forest using high-resolution multispectral images[J]. International Journal of Remote Sensing, 2014, 35(13):4758-4777. [2] Gao L P, Shi W Z, Miao Z L, et al. Method based on edge constraint and fast marching for road centerline extraction from very high-resolution remote sensing images[J]. Remote Sensing, 2018, 10(6):900. [3] Porter S, Linderman M . Historic land cover change in the agricultural midwest using an object-based approach for classification of high-resolution imagery[J]. Journal of Applied Remote Sensing, 2013, 7(1):073506. [4] 黄鹏, 郑淇, 梁超. 图像分割方法综述[J]. 武汉大学学报(理学版), 2020,(6):519-531.[4] Huang P, Zheng Q, Liang C. Overview of image segmentation metho-ds[J]. Journal of Wuhan University(Natural Science Edition), 2020, 66(6):519-531.[5] Peng B, Zhang L, Zhang D. A survey of graph theoretical approaches to image segmentation[J]. Pattern Recognition, 2013, 46(3):1020-1038. [6] Fan S, Sun Y, Shui P. Region-merging method with texture pattern attention for SAR image segmentation[J]. IEEE Geoscience and Remote Sensing Letters, 2020, 18(1):112-116. [7] Zhou C, Wu D, Qin W, et al. An efficient two-stage region merging method for interactive image segmentation[J]. Computers and Electrical Engineering, 2016, 54:220-229. [8] Lassalle P, Inglada J, Michel J, et al. A scalable tile-based framework for region-merging segmentation[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53(10):5473-5485. [9] 黄亮, 姚丙秀, 陈朋弟, 等. 高分辨率遥感影像超像素的模糊聚类分割法[J]. 测绘学报, 2020, 49(5):589-597.[9] Huang L, Yao B X, Chen P D, et al. Superpixel segmentation method of high resolution remote sensing image based on fuzzy clustering[J]. Acta Geodaetica et Cartographica Sinica, 2020, 49(5):589-597.[10] An J, Shi Y, Han Y, et al. Extract and merge:Superpixel segmentation with regional attributes[C]// European Conference on Computer Vision.Springer, 2020:155-170.[11] Xu H, Zhang H, He W, et al. Superpixel-based spatial-spectral dimension reduction for hyperspectral imagery classification[J]. Neurocomputing, 2019, 360:138-150. [12] Achanta R, Shaji A, Smith K, et al. SLIC Superpixels compared to state-of-the-art superpixel methods[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(11):2274-2282. [13] Karydas C, Jiang B. Scale optimization in topographic and hydrographic feature mapping using fractal analysis[J]. International Journal of Geo-Information, 2020, 9(11):631.[14] Comaniciu D M P. A robust approach toward feature space analysis[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(5):313-329. [15] Park J H, Lee G S, Park S Y. Color image segmentation using adaptive mean shift and statistical model-based methods[J]. Computers and Mathematics with Applications, 2009, 57(6):970-980. [16] Wang S, Chung F, Xiong F. A novel image thresholding method based on Parzen window estimate[J]. Pattern Recognition, 2008, 41(1):117-129. [17] 向日华, 王润生. 一种基于高斯混合模型的距离图像分割算法[J]. 软件学报, 2003, 14(7):1250-1257.[17] Xiang R H, Wang R S. A range image segmentation algorithm based on Gaussian mixture model[J]. Journal of Software, 2003, 14(7):1250-1257.[18] 赵泉华, 石雪, 王玉, 等. 可变类空间约束高斯混合模型遥感图像分割[J]. 通信学报, 2017, 38(2):34-43.[18] Zhang Q H, Shi X, Wang Y, et al. Remote sensing image segmentation based on spatially constrained Gaussian mixture model with unknown class number[J]. Journal on Communications, 2017, 38(2):34-43.[19] Li W, Mao K, Zhang H, et al. Selection of Gabor filters for improved texture feature extraction[C]// 2010 IEEE International Conference on Image Processing.IEEE, 2010:361-364.[20] Parzen E. On estimation of a probability density function and mode[J]. Annals of Mathematical Statistics, 1962, 33(3):1065-1076. [21] Scott D W. Multivariate density estimation:Theory,practice,and visualization[M]. John Wiley and Sons, 2015.[22] Jones M C, Marron J S, Sheather S J. A brief survey of bandwidth selection for density estimation[J]. Journal of the American Statistical Association, 1996, 91(433):401-407. [23] Raykar V C, Duraiswami R. Fast optimal bandwidth selection for kernel density estimation[C]// Proceedings of the 2006 SIAM International Conference on Data Mining.Society for Industrial and Applied Mathematics, 2006:524-528.[24] Botev Z I, Kroese D P. Non-asymptotic bandwidth selection for density estimation of discrete data[J]. Methodology and Computing in Applied Probability, 2008, 10(3):435-451. [25] Trudeau R J. Introduction to graph theory[M]. Courier Corporation, 2013.[26] Foulds L R. Graph theory applications[M]. Springer Science and Business Media, 2012.[27] Unnikrishnan R, Pantofaru C, Hebert M. Toward objective evaluation of image segmentation algorithms[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(6):929-944. [28] Gong C, Zhou P, Han J . Learning rotation-invariant convolutional neural networks for object detection in VHR optical remote sensing images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, 54(12):7405-7415. [29] Fang Y, Wang J. Selection of the number of clusters via the bootstrap method[J]. Computational Statistics and Data Analysis, 2012, 56(3):468-477. [30] Haslbeck J M B, Wulff D U. Estimating the number of clusters via a corrected clustering instability[J]. Computational Statistics, 2020(35):1879-1894.[31] 青海玉树震后GeoEye-1卫星地图[EB/OL].(2010-04-20)[2021-02-15].http://www.godeyes.cn/html/2010/04/20/download_9519.html.
[31] GeoEye-1 satellite map after Yushu earthquake in Qinghai Province[EB/OL].(2010-04-20)[2021-02-15].http://www.godeyes.cn/html/2010/04/20/download_9519.html.
-
计量
- 文章访问数: 597
- PDF下载数: 35
- 施引文献: 0
下载: