Abstract
Medical images of MRI, CT and X-ray are the digital form of biological features of human organs. The bulk data of medical images generated by medical imaging processes requires different image compression techniques to reduce the storage space with considerable image quality. This paper presents the compression technique of Feed Forward Back Propagation Neural Network (FFBPNN) for compression and decompression grayscale medical images i.e. MRI of brain image. The FFBPNN with back propagation training algorithms is used to increase the performance by giving better image quality for different compression ratios. This compressed output is encoded with Huffman encoding techniques for much better compression of medical images. From the results, it can be concluded that FFBPNN technique using Levenberg Marquardt (LM) algorithm provides low Mean Squared Error (MSE), high Peak Signal to Noise Ratio (PSNR), good Structural Similarity Index Measurement (SSIM) compared to the existing algorithms.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have