In the multi-focus image fusion algorithm based on wavelet transform, the quality of fused images is usually reduced due to the discrepancy between the adjacent pixels.According to the transform domain of the fusion rule, the eigenvalues of region energy were extracted by using the coefficients of the wavelet transform, and the global match degree was regarded as a way to determine the fusion rules.A new fusion algorithm based on region energy framework was proposed.To improve the quality of fused images, the consistency of the source of each wavelet component was detected.Finally, the algorithm was simulated in experiments.The results verify that this algorithm has better fusion effects than those based on weighted average, selecting the largest gray value and region energy.%在基于小波变换多聚焦图像融合算法中,由于融合图像中相邻像素点之间的不一致使融合图像质量大为降低.应用变换域的融合规则,以小波变换系数提取区域能量为特征值,并根据全局匹配度来决策融合规则,提出了能量框架融合算法,对小波各分量的来源进行一致性检测,使融合图像的小波分解系数更精确地应用于融合图像重构,并对此算法进行了仿真实验,通过对评价指标的分析比较,结果显示此算法图像融合效果优于加权平均、灰度值取大和区域能量算法.
展开▼
