用视觉特性和自适应优化搜索实现多聚焦彩色图像重构

A Multi-focus Color Image Reconstruction Approach-Based Human Vision Featureand Adaptive Optimal Search

鉴于来自同一景物的两幅变焦点彩色图像,每幅图像中都有聚焦不同的目标,针对这种图像的清晰恢复提出了一种新的图像融合算法,其基本思想是先对源图像进行图像块分解;其次,利用人眼视觉特性来分析图像子块的清晰度,形成融合图像;最后通过自适应优化搜索算法得出重构图像。采用平均误差、均方误差、标准色差等量化判据对融合图像序列进行分析及效果评价,量化评价结果和视觉判断均说明,该方法性能优化,完成了多聚焦彩色图像对源图像的重构。

For two color images from the same scene, each of them has different focus. A novel image fusion scheme for clear restoration of the image is presented. The basic idea is that: first to perform block decomposition of each source image, and then image definition is analyzed based on human vision feature to form the fused image, finally the reconstructed image is obtained by adaptive optimal search algorithm. With the use of the parameters such as mean absolute error, mean square error and normalized color difference, the performance of the fused image sequences are evaluated and analyzed. The quantitative results and the human vision both show that the fusion scheme is optimal to achieve color reconstruction of multi-focus image to original image.