基于提升小波变换的图像清晰度评价算法

Algorithm of image definition evaluation based on lifting wavelet transform

针对图像采集系统存在噪声的情况,提出了一种新的基于提升小波变换的清晰度评价函数.该算法利用第二层提升小波分解得到的高、低频子带的能量值构造一种新的清晰度评价函数.比较分析了新提出的评价函数和经典的方差函数、拉普拉斯函数及另外一种基于小波变换评价函数的清晰度评价性能.大量的仿真实验和实际测试结果表明:新提出的清晰度评价算子不仅在保持最高灵敏度的同时具有最好的抗噪性能,而且计算复杂度低,评价结果更为准确、稳定、可靠.

A new measure based on lifting wavelet transform was proposed for the image noise by the energy of high frequency and low-frequency subbands after the second-level decomposition. The performance of this new measure was compared with the classic and popular focus measures variance,laplace and another wavelet based high frequency focus measure. Detailed simulations as well as real data show that the new algorithm can measure image definition correctly with the highest sensitivity in noisy condition while enhancing computation efficiency and are more accurate, stable and reliable.