SEGMENTATION METHOD OF HIGH RESOLUTION REMOTE SENSING IMAGE FOR FAST TARGET RECOGNITION

Chenming Li, Hongmin Gao, Yao Yang, Xiaoyu Qu, Wenjing Yuan

Keywords

High-resolution remote sensing Image; Image Segmentation; Pulse Coupled Neural Network; Threshold Function

Abstract

Pulse-Coupled Neural Network (PCNN) is a simplified neural network model based on the principle of cat vision. Compared with traditional neural networks, PCNN can extract effective information from complex backgrounds without learning or training, and is widely used in image segmentation, edge detection, and so on. Nevertheless, the threshold function of the traditional PCNN has a slow decline rate resulting in slow segmentation of the remote-sensing image in the network; thus, a new threshold function is studied. The improved threshold function replaces the traditional exponentially decreasing threshold with a linearly decreasing threshold, which increases the speed of the threshold drop and reduces the number of network iterations. Experiment results show that faster segmentation speed is obtained by using the improved threshold function to construct the PCNN for remote-sensing image segmentation, and better applied in fast target recognition.

Important Links:

Go Back