A hyperspectral image classification algorithm based on atrous convolution
| Autor: | Weike Liu, Yongguo Zheng, Xiaoqing Zhang, Zhiyong Wang |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Deep Convolutional Neural Networks
Computer Networks and Communications Computer science 0211 other engineering and technologies lcsh:TK7800-8360 02 engineering and technology Gridding problem Convolutional neural network lcsh:Telecommunication 030218 nuclear medicine & medical imaging Convolution 03 medical and health sciences 0302 clinical medicine lcsh:TK5101-6720 Pyramid Hyperspectral image classification Pyramid (image processing) 021101 geological & geomatics engineering Atrous Convolution Pixel lcsh:Electronics Hyperspectral imaging Construct (python library) Computer Science Applications Computer Science::Computer Vision and Pattern Recognition Signal Processing Algorithm |
| Zdroj: | EURASIP Journal on Wireless Communications and Networking, Vol 2019, Iss 1, Pp 1-12 (2019) |
| ISSN: | 1687-1499 |
| DOI: | 10.1186/s13638-019-1594-y |
| Popis: | 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. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink