A hyperspectral image classification algorithm based on atrous convolution

Xiaoqing Zhang,Zhiyong Wang,Weike Liu,Yongguo Zheng

doi:10.1186/s13638-019-1594-y

Xiaoqing Zhang, Zhiyong Wang + Show 2 more

Open Access

https://doi.org/10.1186/s13638-019-1594-y

Copy DOI

Abstract

Hyperspectral images not only have high spectral dimension, but the spatial size of datasets containing such kind of images is also small. Aiming at this problem, we design the NG-APC (non-gridding multi-level concatenated Atrous Pyramid Convolution) module based on the combined atrous convolution. By expanding the receptive field of three layers convolution from 7 to 45, the module can obtain a distanced combination of the spectral features of hyperspectral pixels and solve the gridding problem of atrous convolution. In NG-APC module, we construct a 15-layer Deep Convolutional Neural Networks (DCNN) model to classify each hyperspectral pixel. Through the experiments on the Pavia University dataset, the model reaches 97.9% accuracy while the parameter amount is only 0.25 M. Compared with other CNN algorithms, our method gets the best OA (Over All Accuracy) and Kappa metrics, at the same time, NG-APC module keeps good performance and high efficiency with smaller number of parameters.

Highlights

Hyperspectral remote sensing is a novel technology that can simultaneously acquire spectral and spatial information at the nanometer scale while maintaining the advantages of the previous wide-band remote sensing technology
Since our algorithm aims at small target recognition of different features, we select the Pavia University dataset to verify the effectiveness of the algorithm
Even though non-gridding multi-level concatenated atrous pyramid convolution module (NG-APC) algorithm has less accuracy than 3D Convolutional Neural Networks (CNN) [32] classification results in bitumen and bricks features, it has much higher than the average level (0.821 and 0.900)

Summary

Introduction

Hyperspectral remote sensing is a novel technology that can simultaneously acquire spectral and spatial information at the nanometer scale while maintaining the advantages of the previous wide-band remote sensing technology. In 2017, Chen proposed the atrous spatial pyramid pooling [24] [25], which was applied to the up-sampling part of the encode-decode architecture to expand the convolution receptive field to acquire long-distance features effectively. The researchers have proposed two solutions: First is the parallel method, in which we use different rates in the same layer of convolution and concatenate the results of the convolution.

Results

Conclusion