Modelling of Efficient Medical Image Compression System Based on Joint Operation of DWT and Neural Network

The inherent issue associated with any digital images is its underlying redundancy and large dimension size, which requires a huge amount of storage space as well as higher bandwidth for transmission over a wireless channel. This factor has inspired the researchers to arrive with an optimal solution that compresses digital images with considerable visual quality. However, the researches in the area of images compression are still very active and propose different solutions based on the modern technologies such as artificial intelligence and machine learning algorithms to establish an optimal mechanism to reduce data communication cost that requires high computing process, high storage cost, and parallel architecture. In this research, modelling of the joint approach based on discrete wavelet transform as well as Back propagation neural network for image compression was carried out to achieve optimal balance among compression ratio and visual image quality in terms of PSNR. Experiments had been achieved, the results obtained based on comparative analysis with three different techniques DWT, BPNN, and joint DWT-BPNN are discussed. The study outcome shows better performance achieved by the proposed joint DWT-BPNN in terms of compression ratio, Computational cost, and peak signal to noise ratio (PSNR) with 10 input image samples. However, the performance of the proposed system for the image compression system can be further enhanced in future research work by modifying the different network configurations and parameters. Based on the experimental observations practically, it can be realized that the joint operation of both DWT and BPNN can compress images without much degrading the visual quality of images.