High-Resolution Remote Sensing Image Segmentation method based on Srelu

Chenming Li, Xiaoyu Qu, Yao Yang, Hongmin Gao, Yongchang Wang, Dan Yao, Wenjing Yuan

Abstract: This study proposes a new activation function, namely, S-type rectified linear unit activation function (SReLU), to alleviate the gradientdispersion of neural network model and improve the segmentationprecision of high-resolution remote sensing images (HRIs).