Home > Published Issues > 2017 > Volume 12, No. 5, May 2017 >

Image Compression Using Advanced Optimization Algorithms

Mohamed E. Emara, Rehab F. Abdel-Kader, and Mohamed S. Yasein
Electrical Engineering Department, Faculty of Engineering, Port-Said University, Port-Said, Egypt

Abstract—In this paper, a new image compression technique that uses three-dimensional Discrete Cosine Transform and relies on Two-Dimensional Discrete Wavelet Transform, for image classification, is proposed. The proposed technique utilizes a modified quantization table and a method for converting a three-dimensional image cube into a one-dimensional array, which provides better coding efficiency in the run length coding step. To ensure faster performance, the proposed technique uses parallel computation-based algorithms. The first one utilizes computations parallelization process using SPMD (Single Program Multiple Data) and the other method utilizes Graphics Processor Unit (GPU) programming with CUDA language. Several images have been used to test the proposed algorithm. Experimental results demonstrate that the proposed algorithm outperforms previous compression methods in terms of Peak-Signal-to-Noise Ratio with a lower compression bit rate.

Index Terms—DCT, JPEG, DWT, parallel computation

Cite: Mohamed E. Emara, Rehab F. Abdel-Kader, and Mohamed S. Yasein, "Image Compression Using Advanced Optimization Algorithms," Journal of Communications, vol. 12, no. 5, pp. 271-278, 2017. Doi: 10.12720/jcm.12.5.271-278