Journal
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
Volume 18, Issue 10, Pages 1786-1790Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/LGRS.2020.3008051
Keywords
Feature extraction; Training; Task analysis; Data models; Entropy; Convolutional neural networks; Data mining; Convolutional neural network (CNN); hyperspectral image (HSI) classification; spatial constraint
Categories
Funding
- National Key Research and Development Program of China [2018YFA0702501]
- Key State Science and Technology Project [2016ZX05024001-005]
- NSFC [41974126, 41674116]
Ask authors/readers for more resources
This letter proposes a simple but innovative end-to-end deep U-net-based model for hyperspectral image classification, which directly takes the whole HSI as network input and outputs predicted classes for each pixel location. The combination of classification loss and spatial constraint loss in the training stage enhances the spatial continuity and consistency of the predicted results, showing promising performance compared to existing CNN-based methods.
Hyperspectral image (HSI) classification is of vital importance in remote sensing-related applications. Various approaches, including the recently popular convolutional neural network (CNN)-based models, are proposed to tackle the problem of exploitation of the spatial and spectral features in the HSIs for the use of training classifier. In this letter, we design a simple but innovative end-to-end deep U-net-based model for HSI classification task. Unlike the previous CNN based models that mainly use CNN for spatial feature extraction and process the HSI data locally in small patches, our model takes the whole HSI as network input directly and outputs the predicted classes corresponding to each pixel location. Classification loss in the train data set and spatial constraint loss for the predicted result are combined as the loss function in the training stage to learn the mapping from HSI data to classification map and enhance the spatial continuity and consistency of the predicted result. Benchmark HSI data sets are used to evaluate the performance of the proposed method. Experimental results show that our model can achieve promising results comparing with the existing CNN-based methods.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available