Analysis of Simulation Process of Infrared Image Compression Based on BP Network

The principle and model of a Back Propagation(BP) network used in infrared image compression are presented.The implementation process of image compression is described in detail. The compression of infrared images is implemented by using the Levenberg-Marquardt Algorithm.The analysis of the simulation result shows that the quality of the recovered infrared image can be improved when the quantity of neuron in the hidden layer is increased and the compression ratio is sacrificed.