基于融合视觉词袋的高分遥感水体提取算法

doi:10.16182/j.issn1004731x.joss.20-0977

摘要/Abstract

摘要：

针对高分遥感影像中水体目标提取易受植被阴影及光照条件等影响问题，提出基于融合视觉词袋的高分遥感水体目标提取算法。在分析遥感水体特性基础上，提出一种遥感水体目标频谱特征提取方法。为增强水体目标特征表达能力，设计一种基于局部二值模式和频谱特征的融合视觉词袋模型。基于提出的模型，设计水体目标分类器。为进一步细化水体和非水体边界，提出对水陆交界区域进行二次识别，得到优化后的水体目标提取结果。实验结果表明：提出算法能有效检测高分遥感影像中水体目标，在准确率和Kappa系数上表现良好。

关键词: 高分, 遥感影像, 水体目标提取, 频谱特征, 融合视觉词袋模型

Abstract:

Aiming at the problem that water body extraction is easily influenced by shadow or light in high resolution remote sensing images, an improved algorithm based on fusion of visual word bags is proposed. Based on the deep analysis of the characteristics of remote sensing water body targets, a spectral feature extraction approach is designed. To enhance the description ability of water body targets, a novel visual word bag fusion model based on local binary pattern and spectral feature is constructed. Based on the proposed visual word bag fusion model, a water body target classifier is presented. To further classify the boundaries of the water body targets and the non-water body objects, an optimization algorithm is proposed and the final water body targets extraction results can be obtained. Experimental results show that the proposed algorithm can extract the water body targets exactly, and performs well in accuracy and Kappa coefficient.

Key words: high resolution, remote sensing image, water body extraction, spectral feature, visual word bag fusion model

中图分类号:

TP391

王鑫, 徐明君, 肖坚, 徐立中. 基于融合视觉词袋的高分遥感水体提取算法[J]. 系统仿真学报, 2022, 34(5): 1033-1043.

Xin Wang, Mingjun Xu, Jian Xiao, Lizhong Xu. Water Body Extraction from High Resolution Remote Sensing Images Based on Fused Visual Word Bags[J]. Journal of System Simulation, 2022, 34(5): 1033-1043.

图/表 16

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

图11

图12

图13

表1

表2

表3

参考文献 20

[1]	石祥滨, 钟健, 刘翠微, 等. 小样本高光谱遥感图像深度学习方法[J]. 系统仿真学报, 2018, 30(7): 2744-2752.
	Shi Xiangbin, Zhong Jian, Liu Cuiwei, et al. Deep Learning Method for Hyperspectral Remote Sensing Images with Small Samples[J]. Journal of System Simulation, 2018, 30(7): 2744-2752.
[2]	杨树文, 薛重生, 刘涛, 等. 一种利用TM影像自动提取细小水体的方法[J]. 测绘学报, 2010, 39(6): 611-617.
	Yang Shuwen, Xue Chongsheng, Liu Tao, et al. A Method of Small Water Information Automatic Extraction from TM Remote Sensing Images[J]. Acta Geodaetica et Cartographica Sinica, 2010, 39(6): 611-617.
[3]	McFeeters, S K. The Use of the Normalized Difference Water Index (NDWI) in the Delineation of Open Water Features[J]. International Journal of Remote Sensing (S0143-1161), 1996, 17(7): 1425-1432.
[4]	崔佳玲. 基于纹理的高分辨率遥感图像水陆分离算法[D]. 武汉: 华中科技大学, 2016.
	Cui Jialing. High-resolution Remote Sensing Images Water-land Separation Algorithm Based on Texture Feature[D]. Wuhan: Huazhong University of Science & Technology, 2016.
[5]	王宁, 程家骅, 张寒野, 等. U-net模型在高分辨率遥感影像水体提取中的应用[J]. 国土资源遥感, 2020, 32(1): 35-42.
	Wang Ning, Cheng Jiahua, Zhang Hanye, et al. Application of U-net Model to Water Extraction with High Resolution Remote Sensing Data[J]. Remote Sensing for Land & Resources, 2020, 32(1): 35-42.
[6]	陈前, 郑利娟, 李小娟, 等. 基于深度学习的高分遥感影像水体提取模型研究[J]. 地理与地理信息科学, 2019, 35(4): 43-49.
	Chen Qian, Zheng Lijuan, Li Xiaojuan, et al. Water Body Extraction from High-Resolution Satellite Remote Sensing Images Based on Deep Learning[J]. Geography and Geo-Information Science, 2019, 35(4): 43-49.
[7]	Zhang Y, Jin R, Zhou Z, et al. Understanding Bag-of-Words Model: a Statistical Framework[J]. International Journal of Machine Learning and Cybernetics (S1868-8071), 2010, 1(1): 43-52.
[8]	Zhao L J, Tang P, Huo L Z. A 2-D Wavelet Decomposition-Based Bag-of-Visual-Words Model for Land-Use Scene Classification[J]. International Journal of Remote Sensing (S0143-1161), 2014, 35(6): 2296-2310.
[9]	王鑫, 徐明君, 李可, 等. 一种有效的高分辨率遥感影像水体提取方法[J]. 计算机工程与应用, 2019, 55(20): 145-151.
	Wang Xin, Xu Mingjun, Li Ke, et al. Effective Water Body Extraction Method for High Resolution Remote Sensing Images[J]. Computer Engineering and Applications, 2019, 55(20): 145-151.
[10]	王小鹏, 文昊天, 王伟, 等. 形态学边缘检测和区域生长相结合的遥感图像水体分割[J]. 测绘科学技术学报, 2019, 36(2): 149-154.
	Wang Xiaopeng, Wen Haotian, Wang Wei, et al. Water Segmentation of Remote Sensing Image Using Morphological Edge Detection and Region Growing[J]. Journal of Geomatics Science and Technology, 2019, 36(2): 149-154.
[11]	邓欣, 龙灿, 米建勋, 等. 基于多域特征与随机子空间集成的脑电分类[J]. 系统仿真学报, 2020, 32(9): 1787-1798.
	Deng Xin, Long Can, Mi Jianxun, et al. EEG Classification Based on Multi-domain Features and Random Subspace Ensemble[J]. Journal of System Simulation, 2020, 32(9): 1787-1798.
[12]	谢士琴, 赵天忠, 王威, 等. 高分二号卫星影像融合方法比较研究[J]. 系统仿真学报, 2017, 29(11): 2742-2746.
	Xie Shiqin, Zhao Tianzhong, Wang Wei, et al. Study on Fusion Algorithms of GF-2 Satellite Image[J]. Journal of System Simulation, 2017, 29(11): 2742-2746.
[13]	张永, 杨浩. 基于优化视觉词袋模型的图像分类方法[J]. 计算机应用, 2017, 37(8): 2244-2247.
	Zhang Yong, Yang Hao. Image Classification Method Based on Optimized Bag-of-Visual Words Model[J]. Journal of Computer Applications, 2017, 37(8): 2244-2247.
[14]	Liang X, Hu W Y. A Coupled Co-Occurrence Matrix/Multi-Scale Segmentation Method to Extract Water from High Resolution Remote Sensing Image[C]//GEOINFORMATICS 2018: 26th International Conference on Geoinformatics. Kunming, China: International Conference on Geoinformatics, 2018: 1-4.
[15]	Boell M G, Alves H R, Volpato M, et al. Exploiting Feature Extraction Techniques for Remote Sensing Image Classification[J]. IEEE Latin America Transactions (S1548-0992), 2018, 16(10): 2657-2664.
[16]	Li X, Lyu X, Tong Y, et al. An Object-Based River Extraction Method via Optimized Transductive Support Vector Machine for Multi-Spectral Remote-Sensing Images[J]. IEEE Access (S2169-3536), 2019, 7: 46165-46175.
[17]	Chen Y, Fan R, Yang X, et al. Extraction of Urban Water Bodies from High-Resolution Remote-Sensing Imagery Using Deep Learning[J]. Water (S2073-4441), 2018, 10(5): 585.
[18]	Li L, Yan Z, Shen Q, et al. Water Body Extraction from Very High Spatial Resolution Remote Sensing Data Based on Fully Convolutional Networks[J]. Remote Sensing (S2072-4292), 2019, 11(10): 1162.
[19]	Zhang F, Li J, Zhang B, et al. A Simple Automated Dynamic Threshold Extraction Method for the Classification of Large Water Bodies From Landsat-8 OLI Water Index Images[J]. International Journal of Remote Sensing (S0143-1161), 2018, 39(11): 3429-3451.
[20]	Feng W, Sui H, Huang W, et al. Water Body Extraction From Very High-Resolution Remote Sensing Imagery Using Deep U-Net and a Superpixel-Based Conditional Random Field Model[J]. IEEE Geoscience and Remote Sensing Letters (S1545-598X), 2019, 16(4): 618-622.

序号	Acc/%	Los/%	Fal/%	Kappa
1	98.50	0.96	0.54	0.966 5
2	98.32	1.14	0.55	0.901 0
3	98.40	0.97	0.63	0.944 2
4	99.26	0.30	0.44	0.966 5
5	98.60	0.74	0.66	0.971 9
6	97.78	1.52	0.70	0.941 9
7	98.58	0.58	0.85	0.966 9
8	98.23	0.65	1.11	0.952 0
9	98.85	0.90	0.24	0.967 7
10	98.34	0.71	0.95	0.957 7
平均	98.49	0.85	0.67	0.953 6

序号	Before Acc/%	Before Kappa	After Acc/%	After Kappa
1	97.06	0.933 5	98.50	0.966 5
2	96.67	0.784 7	98.32	0.901 0
3	97.01	0.891 1	98.40	0.944 2
4	99.03	0.954 9	99.26	0.966 5
5	97.86	0.957 1	98.60	0.971 9
6	96.43	0.908 5	97.78	0.941 9
7	97.76	0.948 7	98.58	0.966 9
8	97.49	0.930 1	98.23	0.952 0
9	97.61	0.931 3	98.85	0.967 7
10	97.27	0.928 8	98.34	0.957 7
平均	97.42	0.916 9	98.49	0.953 6

算法	Acc/%	Los/%	Fal/%	Kappa
1	97.57	1.10	1.33	0.925 8
2	93.13	5.91	0.96	0.843 1
本文	98.4	0.85	0.67	0.953 6