在利用主动学习方法进行高光谱图像分类时，往往存在空-谱特征不能得到有效利用和样本需要进行手动标注的问题。针对这些问题，提出一种结合卷积神经网络的主动学习方法进行高光谱图像分类。该方法首先提取像素的空间邻域组成训练样本，通过卷积神经网络对样本的空间特征和光谱特征进行学习并对数据进行初步分类；然后，基于高光谱图像的空间相似性和光谱相似性，对无标注样本进行标注，并将其加入标注训练集以提高分类器的分类精度。在Salinas、PaviaU和Indian Pines这3个高光谱数据上的实验结果表明，该方法能在较少标注样本的情况下，有效提高高光谱图像的分类精度。
When we apply the active learning method to the hyperspectral image classification, there always exist problems that the spectral-spatial features cannot be effectively utilized and the samples need to be manually labeled. In this work, we propose an active learning method combined with the convolutional neural networks to solve the above problems. Firstly, the neighborhood pixels of each pixel are extracted to form training samples. Then the spatial and spectral features of the samples are trained by the convolutional neural network and the data are preliminarily classified. Then, based on the spatial similarity and spectral similarity of the hyperspectral image, some of the unlabeled samples are labeled and added to the training set, which will further increase the accuracy of the classifier. Validated by Salinas image, Pavia University image, and Indian Pines image, the proposed method effectively improves the classification accuracy with less labeled samples.
[1] 布和, 贾艳茹. 基于遥感图像的土地分类[J]. 内蒙古水利, 2015(2):177-178.
[2] Zhang H, Li Y, Zhang Y, et al. Spectral-spatial classification of hyperspectral imagery using a dual-channel convolutional neural network[J]. Remote Sensing Letters, 2017, 8(5):438-447.
[3] 陶超, 田彦平, 邹峥嵘. 主动学习与图的半监督相结合的高光谱影像分类[J]. 测绘学报, 2015, 44(8):919-926.
[4] Dai X, Guo S, Ren Y, et al. Hyperspectral remote sensing image classification using the stacked sparse autoencoder[J]. Journal of University of Electronic Science and Technology of China, 2016, 45(3), 382-386.
[5] 杜培军, 夏俊士, 薛朝辉, 等. 高光谱遥感影像分类研究进展[J]. 遥感学报, 2016, 20(2):236-256.
[6] 刘康. 基于主动学习的高光谱图像分类技术研究[D]. 北京:中国矿业大学(北京), 2014.
[7] 程圆娥, 周绍光, 袁春琦. 基于主动深度学习的高光谱影像分类[J]. 计算机工程与应用, 2017, 53(17):192-196.
[8] 王立国, 李阳. 融合主动学习的高光谱图像半监督分类[J]. 哈尔滨工程大学学报, 2017, 38(8):1322-1327.
[9] Zhong Z, Li J, Luo Z, et al. Spectral-spatial residual network for hyperspectral image classification:a 3-D deep learning framework[J]. IEEE Transactions on Geoscience and Remote Sensing, 2018, 56(2):847-858.
[10] Jacopo A, Elena M, Lutgarde B, et al. Spectral-spatial classification of hyperspectral images:three tricks and a new learning setting[J]. Remote Sensing, 2018, 10(7):1156-1174.
[11] Lee H, Kwon H. Going deeper with contextual CNN for hyperspectral image classification[J]. IEEE Trans Image Process, 2017, 26(10):4843-4855.
[12] Angluin D. Queries and concept learning[J]. Machine Learning, 1988, 2(4):319-342.
[13] 陈荣, 曹永锋, 孙洪. 基于主动学习和半监督学习的多类图像分类[J]. 自动化学报, 2011, 37(8):954-962.
[14] Joshi A, Porikli F, Papanikolopoulos N. Multi-class active learning for image classification[C]//IEEE Conference on Computer Vision and Pattern Recognition. IEEE Computer Seciety, 2009:2372-2379.
[15] Kruse F A, Lefkoff A B, Boardman J W, et al. The spectral image processing system (SIPS):interactive visualization and analysis of imaging spectrometer data[J]. Remote Sensing of Environment, 1993, 44:145-163.
[16] Tuia D, Ratle F, Pacifici F, et al. Active learning methods for remote sensing image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2009, 47(7):2218-2232.
[17] Bengio Y, Lamblin P, Popovici D, et al. Greedy layer-wise training of deep networks[J]. Advances in Neural Information Processing Systems, 2007, 19:153-160.
[18] Xing C, Ma L, Yang X. Stacked denoise autoencoder based feature extraction and classification for hyperspectral images[J]. Journal of Sensors, 2016, 2016:1-10.
[19] Han X, Zhong Y, Zhang L. Spatial-spectral classification based on the unsupervised convolutional sparse auto-encoder for hyperspectral remote sensing imagery[J]. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2016, 3(7):25-31.
