LIF-SSEM:A Spiking Neural Network for Hyperspectral Image Classification
This paper proposes a spiking neural network(LIF-SSEM)based on LIF neurons and attention mechanism to address the challenges of deep learning models with multiple parameters,complex algorithms,and high computational energy consumption,which are difficult to apply to hyperspectral image classification based on mobile computing.The model utilizes the SNN architecture and attention mechanism to integrate the spatiotemporal information of hyperspectral images,which can eliminate the uncertainty caused by spectral heterogeneity in image classification.The model can effectively extract spatial spectral features by simultaneously utilizing LIF neurons and approximate derivative algorithms.The experimental results on IP,PU,SA,WHHC,and WHLK datasets show that the LIF-SSEM model has good performance advantages.
NETL
NSTL
万方数据
