联合显著性与MRF的SAR建筑物分割算法

doi:10.19678/j.issn.1000-3428.0063218

计算机工程 ›› 2022, Vol. 48 ›› Issue (4): 284-291,298. doi: 10.19678/j.issn.1000-3428.0063218

联合显著性与MRF的SAR建筑物分割算法

张磊^1,2, 王小龙¹, 刘畅^1,2

1. 中国科学院空天信息创新研究院, 北京 100190;
2. 中国科学院大学电子电气与通信工程学院, 北京 100049

收稿日期:2021-11-12 修回日期:2021-12-27 发布日期:2022-04-14
作者简介:张磊(1997—),男,硕士研究生,主研方向为SAR图像目标检测;王小龙,副研究员、博士;刘畅,研究员、博士。
基金资助:
国家重点研发计划（2017YFB0503001）。

SAR Building Segmentation Algorithm Combining Saliency and MRF

ZHANG Lei^1,2, WANG Xiaolong¹, LIU Chang^1,2

1. Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100190, China;
2. School of Electronic, Electrical and Communication Engineering, University of Chinese Academy of Sciences, Beijing 100049, China

Received:2021-11-12 Revised:2021-12-27 Published:2022-04-14

摘要/Abstract

摘要： 针对经典马尔可夫随机场（MRF）在进行高分辨率SAR图像分割时存在容易受到斑点噪声干扰等问题，提出一种基于建筑物指数相似度距离及MRF模型（BISD-MRF）的高分辨率SAR建筑物分割算法。基于较复杂SAR场景下建筑物目标可能呈现多种形态结构的问题，设计一种多尺度显著性建筑物指数（MSBI）方案来提取建筑物目标的显著性特征，并通过强度信息重构、纹理显著性提取、频谱显著性信息统计来分别提取不同类型区域的显著性信息，构建适用于SAR建筑物目标的显著性模型。在此基础上，将MSBI值引入到改进的基于改进余弦函数的势函数模型中，利用余弦函数对邻域像素MSBI值进行相似性度量，同时利用特征空间语义信息对像素及其邻域像素标签信息进行有效约束，以提升势函数模型对高分辨率SAR建筑物目标的表征能力。不同平台下的建筑物分割实验结果表明，与MRF、MBI、FRFCM等算法相比，本文算法分割性能平均提升了4.3~10.7个百分点，更适用于较复杂场景下高分辨率SAR建筑物的分割任务。

关键词: 高分辨率SAR图像, 多尺度显著性建筑物指数, 改进MRF模型, 建筑物提取, 势函数

Abstract: Aiming at the problem that classic Markov Random Field(MRF) problem is susceptible to speckle noise during high-resolution Synthetic Aperture Radar(SAR) image segmentation, this study proposes a high-resolution SAR building segmentation algorithm based on a Building Index Similarity Distance and MRF(BISD-MRF).Considering that building targets in a more complex SAR scene may show multiple morphological structures, a Multi-scale Saliency Building Index(MSBI) scheme is designed to extract the saliency of building target features using intensity information reconstruction texture saliency extraction and spectral saliency information statistics;the saliency information of different types of target areas is extracted, and a saliency model suitable for SAR building targets is constructed.On this basis, the MSBI value is introduced to the improved potential function model based on the improved cosine function, which is used to measure the similarity of the MSBI value of the neighborhood pixels, and the feature space semantic information is used to measure the pixel and its neighborhood pixel labels.The information can be effectively constrained to improve the ability of the potential function model to characterize high-resolution SAR building targets.The results of building segmentation under different platforms show that the segmentation performance of the proposed algorithm improved by an average of 4.3~10.7 percentage points compared to traditional algorithms, such as MRF, MBI, and FRFCM, and it is more suitable for high-resolution SAR building segmentation tasks in more complex scenes.

Key words: high-resolution SAR image, Multi-scale Saliency Building Index(MSBI), improved MRF model, building extraction, potential function

中图分类号:

TP391.41

张磊, 王小龙, 刘畅. 联合显著性与MRF的SAR建筑物分割算法[J]. 计算机工程, 2022, 48(4): 284-291,298.

ZHANG Lei, WANG Xiaolong, LIU Chang. SAR Building Segmentation Algorithm Combining Saliency and MRF[J]. Computer Engineering, 2022, 48(4): 284-291,298.

http://www.ecice06.com/CN/Y2022/V48/I4/284

图/表 12

20230131202622

20230131202625

20230131202629

20230131202632

20230131202635

20230131202638

20230131202641

20230131202644

20230131202647

20230131202650

20230131202653

20230131202656

参考文献

[1] 李威, 梁军, 张桢, 等.基于ARM GPU的机载SAR成像算法并行优化策略[J].计算机工程, 2020, 46(10):240-247. LI W, LIANG J, ZHANG Z, et al.Parallel optimization strategy of airborne SAR imaging algorithm based on ARM GPU[J].Computer Engineering, 2020, 46(10):240-247.(in Chinese)
[2] COOPER D B.New models and fast algorithms for natural and urban clutter with applications[EB/OL].[2021-10-10].https://www.researchgate.net/publication/235102603_New_Models_and_Fast_Algorithms_for_Natural_and_Urban_Clutter_with_Applications.
[3] TUPIN F, ROUX M.Detection of building outlines based on the fusion of SAR and optical features[J].Journal of Photogrammetry and Remote Sensing, 2003, 58(1/2):71-82.
[4] MICHAELSEN E, SOERGEL U, THOENNESSEN U.Perceptual grouping for automatic detection of man-made structures in high-resolution SAR data[J].Pattern Recognition Letters, 2006, 27(4):218-225.
[5] ZHANG F L, SHAO Y, ZHANG X, et al.Building L-shape footprint extraction from high resolution SAR image[C]//Proceedings of 2011 Joint Urban Remote Sensing Event.Washington D.C., USA:IEEE Press, 2011:273-276.
[6] ZHAO L J, ZHOU X G, KUANG G Y.Building detection from urban SAR image using building characteristics and contextual information[J].EURASIP Journal on Advances in Signal Processing, 2013(1):1-16.
[7] 李响, 苏娟, 杨龙.基于改进YOLOv3的合成孔径雷达图像中建筑物检测算法[J].兵工学报, 2020, 41(7):1347-1359. LI X, SU J, YANG L.A SAR image building detection algorithm based on improved YOLOv3[J].Acta Armamentarii, 2020, 41(7):1347-1359.(in Chinese)
[8] CHEN J K, QIU X L, DING C B, et al.CVCMFF net:complex-valued convolutional and multifeature fusion network for building semantic segmentation of SAR images[J].IEEE Transactions on Geoscience and Remote Sensing, 2022, 60:1-14.
[9] WANG Z C, DU L, SU H T.Target detection via bayesian-morphological saliency in high-resolution SAR images[J].IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(10):5455-5466.
[10] HUANG X, ZHANG L P.A multidirectional and multiscale morphological index for automatic building extraction from multispectral GeoEye-1 imagery[J].Photogrammetric Engineering & Remote Sensing, 2011, 77(7):721-732.
[11] 侯一民, 郭雷.一种基于马尔可夫随机场的SAR图像分割新方法[J].电子与信息学报, 2007, 29(5):1069-1072. HOU Y M, GUO L.A novel SAR image segmentation method based on Markov random field[J].Journal of Electronics & Information Technology, 2007, 29(5):1069-1072.(in Chinese)
[12] KUTYNIOK G, LIM W Q, REISENHOFER R.ShearLab 3D:faithful digital shearlet transforms based on compactly supported shearlets[J].ACM Transactions on Mathematical Software, 2016, 42(1):1-12.
[13] BRIDGE C P, NOBLE J A.Object localisation in fetal ultrasound images using invariant features[C]//Proceedings of the 12th International Symposium on Biomedical Imaging.Washington D.C., USA:IEEE Press, 2015:156-159.
[14] HOU X D, ZHANG L Q.Saliency detection:a spectral residual approach[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2007:1-8.
[15] 傅兴玉, 尤红建, 付琨.基于改进Markov随机场的高分辨率SAR图像建筑物分割算法[J].电子学报, 2012, 40(6):1141-1147. FU X Y, YOU H J, FU K.Building segmentation from high-resolution SAR images based on improved Markov random field[J].Acta Electronica Sinica, 2012, 40(6):1141-1147.(in Chinese)
[16] CHEN X H, ZHENG C, YAO H T, et al.Image segmentation using a unified Markov random field model[J].IET Image Processing, 2017, 11(10):860-869.
[17] ZHENG C, PAN X X, CHEN X H, et al.An object-based Markov random field model with anisotropic penalty for semantic segmentation of high spatial resolution remote sensing imagery[J].Remote Sensing, 2019, 11(23):28-38.
[18] BESAG J.On the statistical analysis of dirty pictures[J].Journal of the Royal Statistical Society:Series B (Methodological), 1986, 48(3):259-279.
[19] OMATI M, SAHEBI M R.Change detection of polarimetric SAR images based on the integration of improved watershed and MRF segmentation approaches[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2018, 11(11):4170-4179.
[20] CRESSIE N, VERZELEN N.Conditional-mean least-squares fitting of Gaussian Markov random fields to Gaussian fields[J].Computational Statistics & Data Analysis, 2008, 52(5):2794-2807.
[21] LEI T, JIA X H, ZHANG Y N, et al.Significantly fast and robust fuzzy C-means clustering algorithm based on morphological reconstruction and membership filtering[J].IEEE Transactions on Fuzzy Systems, 2018, 26(5):3027-3041.
[22] DICE L R.Measures of the amount of ecologic association between species[J].Ecology, 1945, 26(3):297-302.
[23] JACCARD P.The distribution of the flora in the alpine zone.1[J].New Phytologist, 1912, 11(2):37-50.
[24] XIONG D H, HE C, LIU X L, et al.An end-to-end Bayesian segmentation network based on a generative adversarial network for remote sensing images[J].Remote Sensing, 2020, 12(2):216.
[25] ZHANG C L, ZOU K L, PAN Y.A method of apple image segmentation based on color-texture fusion feature and machine learning[J].Agronomy, 2020, 10(7):972-981.

选择文件类型/文献管理软件名称

选择包含的内容

联合显著性与MRF的SAR建筑物分割算法

SAR Building Segmentation Algorithm Combining Saliency and MRF

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献

相关文章 2

编辑推荐

Metrics

本文评价

[1]	孙新德, 方桂珍, 李玲玲, 薄树奎. 联合形态滤波和MRF模型的红外小目标检测[J]. 计算机工程, 2012, 38(14): 153-156.
[2]	杨勇;黄淑英; 张锋. 基于空间势函数加权的模糊C均值聚类分割算法[J]. 计算机工程, 2007, 33(13): 191-193,.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

联合显著性与MRF的SAR建筑物分割算法

SAR Building Segmentation Algorithm Combining Saliency and MRF

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献

相关文章 2

编辑推荐

Metrics

本文评价