Attention-Based Deep Sequential Network for Polsar Image Classification

Authors:

Abstract
In this paper, we proposed an attention-based deep sequential network (ADSN) for PolSAR images classification increasing the spatial information between pixels by way of spatial sequence. Specifically, the long short-term memory (LSTM) network is introduced to convert the time sequence into spatial sequence to extract the spatial features. Then, a spatial enhanced strategy is carried out to enhance the relationship between pixel spatial information based on LSTM. Finally, to avoid feature selection procedures, the attention mechanism is introduced in LSTM network to select the important information and improve the classification performance. The experiments clearly demonstrate that compared with state-of-art methods, the proposed method can achieve a much better performance and overall Classification accuracy.

This publication has 8 references indexed in Scilit: