期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2017
卷号：17
期号：1
页码：200-208
出版社：International Journal of Computer Science and Network Security
摘要：JPEG has been widely used for lossy image compression. In this paper, we introduce an improved JPEG compression algorithm by modifying the luminance quantization table for color image. Quantization step plays an important role in JPEG compression process. The default quantization table (8x8) depends upon two components named luminance and chrominance of image. Human visual system cannot recognize the slight modification in luminance table slightly. We modify the 8x8 default luminance quantization table. In order to evaluate the performance of modified JPEG algorithm, we have used different color images such as Lena, Baboon, Brick-house and Pepper. The performance of modified JPEG is assessed in terms of compression ratio (CR), mean square error (MSE) and peak signal to noise ratio (PSNR). Experimental results indicate that the performance of modified quantization outperforms the standard quantization in terms of CR, MSE, and PSNR. We observed that CR, MSE and PSNR values of proposed JPEG for different image range from 9.9 to 47, 9.47 to 452, and 21.6 to 38.4 respectively, in contrast to conventional JPEG whose values range from 7 to 25, 18.3 to 927, and 18.5 to 35.5 respectively. This is because the quantization table is modified in a way so that only the lowest frequency DCT coefficients have significant amplitude values.
关键词：JPEG image compression luminance quantization table compression ratio mean square error peak signal-to-noise ratio
