Attentive-Adaptive Network for Hyperspectral Images Classification With Noisy Labels

With the development of deep neural networks, hyperspectral image (HSI) classification systems have achieved a significant improvement. These systems require numerous and accurately labeled hyperspectral data to be adequately trained. However, noisy labels are inherent in real-world hyperspectral systems, resulting in unreliable decisions. To handle noisy labels in hyperspectral classification, an end-to-end attentive-adaptive network (AAN) is proposed for robust HSI classification training. The goal is to build a classifier with strong generalization capabilities that can be applied to both clean and noisy training sets without explicit noise label pretreatment. Specifically, a spectral stem network with a nonadjacent shortcut is exploited initially to redistribute the sensitive layers for noisy labels to achieve robust spectral representation. Then, a group-shuffle attention module is proposed to capture the discriminative and robust spatial-spectral features in the presence of noisy labels. Finally, an adaptive noise-robust loss (ANRL) function is developed to fight against noisy labels by learning a parameter to balance the normalized cross entropy (NCE) and reverse cross entropy (RCE). Experimental results on three HSI benchmark datasets with simulated noisy labels demonstrate the effectiveness of AAN on HSI classification.