WAVELET TRANSFORM AND NEURAL NETWORK COMBINED METHOD FOR MEDICAL IMAGES COMPRESSION AND IMPLEMENTATION ON FPGA

摘要

Medical images present specific characteristics which require to be exploited by an explicit and efficient compression algorithm. The Vector Quantization (VQ) constitutes a crucial stage in lossless digital images compression where it allows to create a dictionary on the level "block" by a neural network approach that of Kohonen's Self Organizing Map (K-SOM) which is widely used in implementation of FPGA circuits for digital image compression applications. This paper is devoted to the implementation of a new combined method based on wavelet transform and neural networks (WT-SOM) that is designed for the implementation on FPGA, for medical images compression applications. Simulation results show the effectiveness of our proposed method.