An Advance Facial Biometric Based System for Lossless Image Compression Algorithm

Abstract

Recent interest has been generated in multimodal biometrics technology due to its potential to increase recognition rates by over- coming some of the fundamental limitations of single biometric modalities. The importance of different data compression algo- rithms is highlighted in the process of storing big data in data warehouses and archives and in reducing the volume of big data during its transmission through communication channels, so many compression algorithms have emerged that are designed to process different data. Images are used in different fields in our daily lives, such as social networks, medical diagnoses, remote sensing, and other fields, and since the size and accuracy of images are constantly evolving, the issue of storing and transferring images requires a compressing process to reduce their size. Although many approaches in data compression have been developed to alleviate these challenges, more efforts are still needed, especially for lossless image compression, which is a promising technique when critical information loss is not allowed. In this paper, we propose a new algorithm called the Lossless Image Compression Algorithm using a Column Subtraction model (LICA-CS). LICA-CS is efficient, low in complexity, decreases the image bit-depth, and enhances state-of-the-art performance. LICA-CS first implements a color transformation method as a pre-processing phase, which strategically minimizes inter-channel correlations to optimize compression outcomes. After that, a novel subtraction method was developed to compress the image data column-wise. We tackle the similarity and proximity of pixel values within adjacent columns, which offers a distinct advantage in reducing image size observing a significant size reduction of 71%. This is achieved through the subtraction of neighboring columns. The conducted experiments on colored images show that LICA-CS outperforms existing algorithms in terms of compression rate. Moreover, our method exhibited remarkable enhancements in execution time, with compression and decompression processes averaging 1.93 seconds. LICA-CS advances the state-of-the-art in lossless image compression, promising enhanced efficiency and effectiveness in image compression technologies