自适应预处理的AMBTC图像压缩

AMBTC Image Compression Using Adaptive Preprocessing

椒盐噪声对于图像压缩有很大的干扰作用,为了提高压缩图像的显示效果,降低计算开销,本文提出了预处理的AMBTC图像压缩算法。首先统计原始图像中椒盐噪声的像素个数,预估出噪声强度,然后对被噪声污染的像素进行自适应的修复,最后利用AMBTC压缩算法对图像进行压缩。修复后的图像仍然可以正常显示,压缩后的AMBTC图像的PSNR在27dB以上,相对于没有修复的AMBTC图像算法PSNR平均有10dB的提升,同时整个处理过程小于1s。实验结果表明,本文算法可以很好地剔除噪声的影响,得到较高的图像质量,同时能够做到实时处理。

Salt and pepper noise has a serious interference effect on image compression.In order to improve the visual quality of compressed images and reduce the computational cost,this paper proposes a preprocessing algorithm of AMBTC image compression.Firstly,the number of pixels pepper and salt noise in the original images is counted,and the noise intensity is estimated.Then the pixels polluted by noise are repaired adaptively.Finally,the images are compressed by using the AMBTC compression algorithm.The repaired image can still be displayed normally and PSNR of the compressed AMBTC images are more than 27dB in general.Compared with the unrepaired AMBTC image algorithm,the PSNR has an improvement of 10dB,and the whole processing process is less than 1s.The experimental results show that the algorithm in this paper can eliminate the influence of noise,obtain the high image quality,and achieve the real-time image processing.