HResNetAM: Hierarchical Residual Network With Attention Mechanism for Hyperspectral Image Classification

This article proposes a novel hierarchical residual network with attention mechanism (HResNetAM) for hyperspectral image (HSI) spectral-spatial classification to improve the performance of conventional deep learning networks. The straightforward convolutional neural network-based models have limitations in exploiting multiscale spatial and spectral features, this is key factor dealing high-dimensional nonlinear characteristics present HSIs. proposed can extract features at granular level, so receptive fields range will be increased, which enhance feature representation ability model. Besides, we utilize set adaptive weights different scales, further discriminative extracted features. Furthermore, double branch structure also exploited corresponding convolution kernels parallel, multiple scales are fused classification. Four benchmark datasets collected by sensors acquisition time employed experiments, comparative results reveal that method has competitive advantages terms when compared other state-of-the-art models.