A Hyperspectral Image Classification Model Based on Band Clustering and Multi-Scale Structure Feature Fusion
WANG Cai-ling1，ZHANG Jing1，WANG Hong-wei2*, SONG Xiao-nan1, JI Tong3
1. College of Computer Science, Xi'an Shiyou University, Xi'an 710065, China
2. School of Artificial Intelligence, Optics and Electronics, Northwestern Polytechnical University, Xi'an 710072, China
3. College of Grass Industry, Gansu Agricultural University, Lanzhou 730070, China
摘要： 高光谱图像包含丰富的地物信息，在农业、工业和军事等领域应用广泛。因此，高光谱图像的识别与分类是一项重要的研究课题。然而，高光谱图像存在光谱维度高、噪声大、标记样本有限等问题，并未取得很好的分类效果。针对以上问题，提出一种波段聚类和多尺度结构特征融合的高光谱图像分类模型 (ASPS-MRTV)。该方法主要包括以下几个步骤，首先，对高光谱数据进行归一化处理，将归一化后的三维图像按光谱维等分为n个子空间；其次，采用粗细划分策略构造自适应子空间光谱特征提取框架，将每个空间波段拉伸为一维向量后用信息散度构造波段的相似性矩阵，按照聚类的思想对n个子空间进行自适应；然后，将每个自适应子空间的光谱波段平均值进行叠加，形成光谱特征；最后，对所得到的光谱特征数据利用多尺度相对全变分技术提取结构特征。为了增强样本的线性可分性，在数据堆叠之后进行核主成分分析，最终形成空谱特征。对比实验中统一使用惩罚参数C和核参数σ都为24.5的SVM进行分类。经测试，ASPS-MRTV网络模型在Indian Pines、University of Pavia两个数据集上分别以5%，1%的训练样本达到了97.06%、98.98%的总体分类精度。实验结果表明，该模型与SVM、ASPS(ED)、ASPS(ID)、ASPS-LBP、ASPS-GlCM、ASPS-BF模型相比，在分类性能和计算效率方面都取得了更优的效果，有效提高小样本下高光谱图像的分类精度。
关键词：高光谱图像；多尺度结构特征；信息散度；核主成分分析；空谱特征
Abstract：Hyperspectral images contain abundant ground object information and are widely used in agriculture, industry, military and other fields. Therefore, its identification and classification is an important research topic. However, hyperspectral images have problems, such as high spectral dimension, large noise and limited labeled samples, so they have not achieved good classification results. This paper proposes a hyperspectral image classification model based on band clustering and multi-scale structural feature fusion (ASPS-MRTV). The method mainly includes the following steps. First, hyperspectral data is normalized, and the normalized 3D image is divided into n subspaces according to spectral dimension. Secondly, an adaptive subspace spectral feature extraction framework is constructed using the coarse and thin division strategy. Each spatial band is stretched into a one-dimensional vector, and then the similarity matrix of the band is constructed by using the information divergence. Then, the spectral band averages of each adaptive subspace were superimposed to form spectral features. Finally, the multi-scale relative total variation technique extracts structural features from the obtained spectral feature data. In order to enhance the linear separability of the samples, kernel principal component analysis was performed after data stacking to form the null spectral features. In the comparison experiment, SVM with penalty parameter C and kernel parameter of 24.5 were uniformly used for classification. After testing, the ASPS-MRTV network model achieves the overall classification accuracy of 97.06% and 98.98% on Indian Pines and University of Pavia datasets with 5% and 1% training samples, respectively. Experimental results show that compared with SVM, ASPS(ED), ASPS(ID), ASPS-LBP, ASPS-GLCM and ASPS-BF models, the proposed model achieves better classification performance and computational efficiency and effectively improves the classification accuracy of hyperspectral images under small samples.
Key words：Hyperspectral image; Multi-scale structural characteristics; Information divergence; Nuclear principal component analysis; Spatial spectrum characteristics
王彩玲，张 静，王洪伟，宋晓楠，纪 童. 一种波段聚类和多尺度结构特征融合的高光谱图像分类模型[J]. 光谱学与光谱分析, 2024, 44(01): 258-265.
WANG Cai-ling，ZHANG Jing，WANG Hong-wei, SONG Xiao-nan, JI Tong. A Hyperspectral Image Classification Model Based on Band Clustering and Multi-Scale Structure Feature Fusion. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(01): 258-265.
[1] Feng F B, Li W D, Du Q, et al. Remote Sensing, 2017, 9(4): 323.
[2] CHEN Yun-hao, WANG Si-jia, ZHAO Yi-fei, et al(陈云浩，王思佳，赵逸飞，等). Geography and Geo-information Science(地理与地理信息科学), 2019, 35(5): 1.
[3] YANG Fei-fei, LI Shi-juan, LIU Sheng-ping, et al(杨菲菲，李世娟，刘升平，等). Journal of Agricultural Science and Technololgy(中国农业科技导报), 2020, 22(4): 85.
[4] YUAN Xu-peng, ZHANG Da-wei, WANG Cheng, et al(原续鹏，张大伟，王 成，等). Optical Instruments(光学仪器), 2017, 39(1): 73.
[5] Imani M, Ghassemian H. Principal Component Discriminant Analysis for Feature Extraction and Classification of Hyperspectral Images, 2014 Iranian Conference on Intelligent Systems (ICIS). IEEE, 2014: 1.
[6] Sun W, Du Q. IEEE Geoscience and Remote Sensing Magazine, 2019, 7(2): 118.
[7] Tan J, Gao Y, Liang Z, et al. IEEE Transactions on Medical Imaging, 2019, 39(6): 2013.
[8] Ghamisi P, Benediktsson J A, Sveinsson J R. IEEE Transactions on Geoscience and Remote Sensing, 2013, 52(9): 5771.
[9] Xia J, Dalla Mura M, Chanussot J, et al. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53(9): 4768.
[10] Dalla Mura M, Benediktsson J A, Waske B, et al. IEEE Transactions on Geoscience and Remote Sensing, 2010, 48(10): 3747.
[11] Hou B, Huang T, Jiao L. IEEE Geoscience and Remote Sensing Letters, 2015, 12(12): 2364.
[12] He L, Liu C, Li J, et al. IEEE Transactions on Geoscience and Remote Sensing, 2020, 58(7): 4818.
[13] Quesada-Barriuso P, Argüello F, Heras D B. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7(4): 1177.
[14] Jia S, Wu K, Zhu J, et al. IEEE Transactions on Geoscience and Remote Sensing, 2018, 57(2): 1142.
[15] MU Cai-hong, LIU Yi, LIU Jian, et al(慕彩红，刘 逸，刘 健，等). Chinese Patent(中国专利)：CN107451614B. 2019.
