Adaptive Thresholding Based Medical Image Compression Technique Using Haar Wavelet Based Listless SPECK Encoder and Artificial Neural Network

摘要

Medical image compression leads a wide attention now days. Image compression plays an important role to reduce the bandwidth and ensures good quality of service in multimedia application. While transmitting the images, only some of the regions (Regions of Interest-ROI) are considered to be more significant than the others, which can be coded with high spatial resolution than the background (Non-ROI) in order to attain high compression rate with high quality ROI. In this research, we are intended to propose the medical Image Compression technique based on ROI and non ROI of the image. The ROI is segmented by K-means clustering based algorithm. This segmentation algorithm will pop out the important regions with the main objective of highlighting any pixels that vary from the rest of the background and catch the human attention. For the ROI regions, the image coding is done using wavelet transform based on modified Listless SPECK (LSK). Here the efficiency of the proposed methodology will be proved by using various medical images. The performance measure can be analyzed by using PSNR and CR. The execution time of the proposed method will be reduced when compare to the other existing methods. The experimental result shows that the application of ROI coding using LSK brings about high compression rate and quality ROI.