Image compression using Least Probable Coefficients Approximation technique. (3rd January 2012)

• Record Type: Journal Article
• Authors: El-said, Shaimaa A.; Hussein, Khalid F.A.; Fouad, Mohamed M.
• Is Part Of: International journal of signal and imaging systems engineering. Volume 4:Number 4(2011)
• Page Start: 197
• Page End: 206
• Publication Date: 2012-01-03
• Subjects: image compression -- adaptive quantisation -- DCT -- discrete cosine transform -- general compression technique -- Huffman encoding -- JPEG; Signal processing -- Periodicals; Imaging systems -- Periodicals; Information theory -- Periodicals; 621.3822
• Languages: English
• ISSNs: 1748-0698

Abstract

A novel Adaptive Lossy Compression technique, namely Least Probable Coefficients Approximation (LPCA) technique, is proposed to achieve better performance and simpler implementation than JPEG at low bitrate. Compression using LPCA based on reducing the number of source symbols by applying efficient processing on the Discrete Cosine Transform (DCT) coefficients of the image in addition to efficient quantisation to reduce the transmitted values. It is a general compression technique that can be applied to any digital data not just images. Experimental comparisons are carried out to compare the performance of the proposed technique with that of JPEG. The experimental results show that: The proposed compression technique achieves high Compression Ratio (CR) with higher Signal to Noise Ratio (SNR) than that of JPEG at low bitrate without the great visual degradation that appears in case of JPEG.