Authors : Abdulrahman Alturki and Abdulrahman A. Alrobaian
Abstract: Image compression still remains a hot research topic due to the generation of massive amount data which needs to be stored or transmitted. Numerous approaches have been presented for image compression to represent the images in a compacted form with no repeated or unrelated pixels. Presently, evolutionary algorithms become more popular to solve the real world problems in an efficient manner. In this study, Firefly (FF) optimization algorithm based on Discrete Cosine Transformation (DCT) is introduced to determine the best fitness value for all DCT block. When the fitness values are computed for DCT blocks, compression process takes place. To enhance the overall compression performance, image warping process is also used as a preprocessing step. However, Space Invariant Feature Transform (SIFT) matching procedure is employed to validate the difference between reference and reconstructed image. A detailed comparison study is performed between the proposed Firefly (FF) algorithm and existing Pollination Based Optimization (PBO) using a set of benchmark images. The proposed method is successfully applied and the experimental analysis prove that the presented FF method is found to be better than previous methods in terms of various performance measures like Compression Ratio (CR), Compression Time (CT), Peak Signal to Noise Ratio (PSNR), Mean Square Error (MSE) and Structural Similarity Index (SSIM).
Abdulrahman Alturki and Abdulrahman A. Alrobaian, 2019. A Novel Lossless Image Compression Technique Based on Firefly Optimization Algorithm. Journal of Engineering and Applied Sciences, 14: 2642-2647.