基于量子核函数算法的图像压缩研究

Image Compression Research Based on Quantum Kernel Function

为了提高图像压缩的质量,提出量子核函数算法。首先采用自适应灰度阈值使图像归一化处理,用量子比特表示图像白色和黑色状态的出现概率;然后每个像素点的值被编码为量子角度向量;接着构造适应度函数的极小值获得高斯径向基核函数的最优核宽,在量子核影响半径以外的像素量子编码不再起引导作用;最后给出了算法流程。实验仿真显示本文算法的压缩重构效果清晰,PSNR最大,压缩时间最少。

In order to improve the quality of image compression, Quantum Kernel Function algorithm was established. First, adaptive gray threshold was adopted for normalization of image, quantum bits was represented probability of image black and white state. Second, each pixel values were encoded into quantum angle vector. Third, minimum function value was obtained Gauss radial basis function kernel optimal kernel width, quantum core was not guided quantum code of pixel outside radius. Finally, process was described. Experiments showed this algorithm was clear reconstruction effect, PSNR was maximum, compression time was minimum.