A Review of Image Compression Using Fractal Image Compression with Neural Network

Abstract

Generally the fractal image compression is a  new process in the images compression. It is a block based image compression technique, which detects and  decodes the existing similarities between different  regions in the image. The main disadvantage of FIC is  that the encoding time is comparatively very high, w 

here as the decoding time is very short. An artificial  intelligence technique like neural network is used to  reduce the search space and encoding time for the MRI  images with an algorithm called as “back propagation”  neural network algorithm. Initially, MRI image is  divided into ranges and domains of fixed size. The best  matched domain is selected for each range block and its  range index and best matched domain index are  produced, which acts as input to the expert system and  which results reduced the sets of matched domain  blocks. The neural network is then trained with these  resultant values. This trained net is now used to  compress other MRI images which lead to a very less  encoding time. During the decoding phase, the  transformation parameters are recursively applied to  any random original image, which then converges to the  fractal image after some changes. The simulation results  show that the performance of this Neural Network  based FIC is really. This paper shows the neural  network based FIC which produces high development in  encoding time without corrupting the image quality  when compared to normal FIC.