Study On Algorithms Of Image Compression Based On BP Neural Network

With the development of digital image communication, the storage and transmission of mass data of digital image becomes the basic technique of multimedia. There are many methods of traditional image compression, but they all have their own limits,lower compression rate and lower compression quality. Artificial neural networks technology research has gone a long way in the past decade. The specialists at home and abroad in the image processing field have paid high attention and been engaged in the advantages of Neural Network techniques such as the abilities of parallel computing,nonlinear mapping and self-adaptiveness,and applied a variety of neural network models into the image processing field. The algorithms of image compression based on BP neural network are studied in this paper. The main contents are as follows: Firstly, the author gives an overview on the characteristics and history of the image compression and introduces the advantages of some basic compressing algorithms in detail. The author also discusses on some problems on application,which will be partly settled in this thesis. And then,the author analysis the principles of BP algorithm and two improved algorithms, simulates with the image compression, illustrates the advantage and disadvantage of BP neural network in the field of image compression. The author proposes a joint-optimized algorithm based on Cauchy error estimator, shape factor is adopted in Sigmoid function. Simulation results show that this algorithm has better compressing time and PSNR compare to BP neural network algorithm. What is more,the author proposes a joint image compression algorithm with Wavelet transform and BP neural network, through simulation, it has a higher compressing rate. Finally, the author makes a conclusion and proposes the future research directions.