| Related citation: | Shanshan Zheng,Wen Liu,Rui Shan,Jingyi Zhao,Guoqian Jiang,Zhi Zhang.Multi-Scale Dilated Convolutional Neural Network for Hyperspectral Image Classification[J].Journal of Harbin Institute Of Technology(New Series),2021,28(4):25-32.DOI:10.11916/j.issn.1005-9113. 2020017. |
|
| Author Name | Affiliation | | Shanshan Zheng | School of Science, Yanshan University, Qinhuangdao 066004, Hebei, China | | Wen Liu | School of Science, Yanshan University, Qinhuangdao 066004, Hebei, China | | Rui Shan | School of Science, Yanshan University, Qinhuangdao 066004, Hebei, China | | Jingyi Zhao | College of Mechanical Engineering, Yanshan University, Qinhuangdao 066004, Hebei, China | | Guoqian Jiang | School of Electrical Engineering, Yanshan University, Qinhuangdao 066004, Hebei, China | | Zhi Zhang | Beijing Institute of Space Mechanic & Electricity, Beijing 100094, China |
|
| Abstract: |
| Aiming at the problem of image information loss, dilated convolution is introduced and a novel multi-scale dilated convolutional neural network (MDCNN) is proposed. Dilated convolution can polymerize image multi-scale information without reducing the resolution. The first layer of the network used spectral convolutional step to reduce dimensionality. Then the multi-scale aggregation extracted multi-scale features through applying dilated convolution and shortcut connection. The extracted features which represent properties of data were fed through Softmax to predict the samples. MDCNN achieved the overall accuracy of 99.58% and 99.92% on two public datasets, Indian Pines and Pavia University. Compared with four other existing models, the results illustrate that MDCNN can extract better discriminative features and achieve higher classification performance. |
| Key words: multi-scale aggregation dilated convolution hyperspectral image classification (HSIC) shortcut connection |
| DOI：10.11916/j.issn.1005-9113. 2020017 |
| Clc Number:TP391 |
| Fund: |
|
| Descriptions in Chinese: |
| 基于多尺度空洞卷积神经网络的高光谱图像分类 郑姗姗1，刘文1，单锐1，赵静一2，江国乾3，张智4 （1. 燕山大学理学院，河北 秦皇岛 066004；2. 燕山大学机械工程学院，河北 秦皇岛 066004；3. 燕山大学电气工程学院，河北 秦皇岛 066004；4. 北京航天研究所，北京 100094） 创新点说明：1）将图像分割方法——空洞卷积用于卷积神经网络进行高光谱图像分类，提取更加广泛、抽象的图像特征。 2）构建基于多尺度空洞卷积神经网络的高光谱图像分类方法。搭建多尺度聚合结构，在每个通道中使用快捷连接和空洞卷积结构，有效提取图像特征，避免信息丢失。 研究目的： 针对图像信息丢失问题，得到高精度的高光谱图像分类方法。 研究方法： 在Indian Pines和Pavia University数据集上，与4个已有的高光谱图像分类方法进行对比实验，比较OA, AA和Kappa值。 研究结果： 多尺度空洞卷积神经网络在Indian Pines和Pavia University数据集上OA值分别达到了99.58%，99.92%。AA值分别达到了99.57%，99.90%。Kappa分别达到了99.52%，99.89%。 结论： 1）在卷积神经网络中引入空洞卷积，可以有效避免图像信息丢失。 2）多尺度空洞卷积神经网络能提取更佳的判别性特征，实现高分类性能。 关键词：多尺度聚合；空洞卷积；高光谱图像分类；快捷连接 |
