多聚焦图像融合中最佳小波分解层数的选取

小波变换作为一种十分有效的信号处理方法已被应用于多聚焦图像融合系统中。但其中需要解决的一个问题是最佳小波分解层数的选取。通过大量的实验 ,总结出基于小波分解子图像面积比和低频带均方根误差两个标准来确定最佳小波分解层数的方法。实验结果表明 ,无论是依据均方根误差、灰度平均误差、熵、空间频率等客观评价标准 ,还是视觉的主观评价 ,所提出的选取方法都是有效可行的

Selection of Optimal Decomposition Level of Wavelet for Multifocus Image Fusion

Wavelet transform, a very efficient signal processing method, has been applied in multifocus image fusion systems. A critical problem is how to select the optimal decomposition level. A selection method is obtained through extensive experiments, which is based on area ratio of decomposition subimage and root square mean error of low low bands. The experimental results show that the proposed method is efficient and feasible in terms of both objective evaluation criteria such as root mean square error, mean error, entropy, spatial frequency and visual subjective evaluation criteria.