The demand for handling images in digital form has increased dramatically in recent years. Image compression is an effective technique of reducing or less the amount of image information that are required to represent/show an image in better format, after compression image size get reduced. The objective of image compression is to reduce the amount of digital images information and therefore reduce the price, storage capacity as well as transmission cost. Image compression performs a key role in various important applications, like image database, image digitization, security industry, health industry etc. This project presents a procedure of employing both methods of compression in brilliant manner to achieve effective compression ratio and less error rate. In this proposed method we are merging the Huffman encoding technique along with LPC for the enhancement of compression ratio. First of all, an medical (MRI) image is separated into two parts, the ROI (Region of interest) and the NROI (Non ROI); then, the two sections are coded individually based on Huffman and LPC. Here, Huffman will provide the tree based encoding scheme for lossless compressionof imageand LPC method is used for lossy compression The experimental results shows thatbetter Signal to Noise Ratio (SNR) with acceptable Compression Ratio (CR) has been achieved using hybrid scheme based on Huffman and LPC, the algorithm also has better robustness.