Abstract
This paper proposes a new lossless compression scheme of the P2-Bit Short Method for grayscale medical images. The proposed method is divided into two parts of the compression and decompression processes. In the compression process, Fragmentation, S-S, Merge, Bit-Conversion and B-Short processes have a vital role. The original grayscale image pixels are segregated and shuffled within the image itself by fragmentation and S-S processes in two levels of iterations. These iterations are based on 4x4 (level-1) and 8x8 (level-2) subbands. The output of the second iteration image pixels are converted into 8-bit binary values. The minimum occurrences of the bit and its positions are found from the binary image in the B-Short process. The results revealed that the minimum times occurred bit and its positions were stored as a compressed image. The compressed image was sent to the receiver for the decompression process. In decompression, the minimum times that the bit occurred were replaced based on it's position in the B-Replace process. The alternative bit filled the rest of the places. After this process, the obtained binary values were converted into decimal values in the Decimal Conversion (DC) process. The retrieved decimal image was considered as an input for the inverse iteration of Fragmentation, Inverse S-S and Merge processes for reconstructing the original grayscale image. Due to the pixel swipe and bit conversion processes, the proposed algorithm provides a high complexity. In experimentation, the proposed method provides the magnificent results of near to loss-less compression with minimum execution time, meaning that nearly 55% of the size can be reduced with good reconstructed image quality and a minimum error rate when compared to existing methods.