Journal
IEEE TRANSACTIONS ON IMAGE PROCESSING
Volume 31, Issue -, Pages 5227-5241Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TIP.2022.3193747
Keywords
Feature extraction; Image reconstruction; Training; Hyperspectral imaging; Calibration; Convolution; Unsupervised learning; Deep neural network; hyperspectral image classification; latent reconstruction; open-set classification; spectral feature reconstruction; open-set environment
Funding
- National Natural Science Foundation of China [61922029, 62101072]
- Hunan Provincial Natural Science Foundation of China [2021JJ30003, 2021JJ40570]
- Science and Technology Plan Project Fund of Hunan Province [2019RS2016]
- Scientific Research Foundation of Hunan Education Department [20B022, 20B157]
Ask authors/readers for more resources
Deep learning methods have shown significant progress in hyperspectral image classification, but robustness in handling unknown objects still needs improvement. To enhance classification accuracy and maintain robustness in open-set environments, a spectral-spatial latent reconstruction framework is proposed for reconstructing spectral and spatial features.
Deep learning-based methods have produced significant gains for hyperspectral image (HSI) classification in recent years, leading to high impact academic achievements and industrial applications. Despite the success of deep learning-based methods in HSI classification, they still lack the robustness of handling unknown object in open-set environment (OSE). Open-set classification is to deal with the problem of unknown classes that are not included in the training set, while in closed-set environment (CSE), unknown classes will not appear in the test set. The existing open-set classifiers almost entirely rely on the supervision information given by the known classes in the training set, which leads to the specialization of the learned representations into known classes, and makes it easy to classify unknown classes as known classes. To improve the robustness of HSI classification methods in OSE and meanwhile maintain the classification accuracy of known classes, a spectral-spatial latent reconstruction framework which simultaneously conducts spectral feature reconstruction, spatial feature reconstruction and pixel-wise classification in OSE is proposed. By reconstructing the spectral and spatial features of HSI, the learned feature representation is enhanced, so as to retain the spectral-spatial information useful for rejecting unknown classes and distinguishing known classes. The proposed method uses latent representations for spectral-spatial reconstruction, and achieves robust unknown detection without compromising the accuracy of known classes. Experimental results show that the performance of the proposed method outperforms the existing state-of-the-art methods in OSE.
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