Research and Implementation of Satellite Image Target Recognition Based on Deep Learning

Abstract

Space remote sensing technology is one of the most important means to acquire ground information at present. In recent years, due to the progress of remote sensing satellite level, the secondary development technology of image recognition, target classification, information extraction and surveillance for ships on the sea has become the focus of research. the convolution neural network algorithm is used to construct of the ship satellite image target recognition network. The convolution neural network model is built by synthesizing the parameter characteristics of the data set. The standard data set and encapsulated ship satellite image target data set are used for training and testing. According to the test results, the convolution neural network has a good ability to recognize ship satellite image target, which proves that the convolution neural network structure is feasible and has a good research and application value.