基于改进对比度的有限离散剪切波图像融合

为了提高多聚焦图像的融合精度，结合有限离散剪切波变换(FDST)良好的局部化特性及平移不变性，提出了一种基于有限离散剪切波变换与改进对比度相结合的图像融合新算法。对经过严格配准后的多聚焦图像进行FDST分解，得到低频子带系数和不同尺度不同方向的高频子带系数；对低频子带系数采用区域平均能量匹配度自适应融合算法，高频子带系数的选取则根据低频与高频系数关联得到的对比度进行融合；应用有限离散剪切波逆变换重构得到融合图像，并对融合结果进行主观视觉和客观评价。通过仿真实验，算法在主观视觉效果上有着明显的优越性。在不同融合算法比较的融合结果中，熵值、互信息量和边缘相似度分别平均提高了1.4%、34.6%和8.0%，各项客观评价指标优于其他算法。

Image fusion based on improved contrast in finite discrete shearlet domain

In order to improve the accuracy of multi focus image fusion, combining with good localization and shift invariance of finite discrete shearlet transform(FDST), a new image fusion algorithm based on FDST and improved contrast was proposed. Firstly, the registration multi focus images were decomposed by FDST, and the low frequency sub band coefficients and high frequency sub band coefficients of different scales and directions were obtained. The fusion principle of low frequency sub band coefficients was based on the method of regional average energy matching degree. As for high frequency sub band coefficients,the sum of contrast was adopted as the fusion rule, which combined the low frequency coefficients with the high frequency coefficients. Finally，low frequency information and high frequency information were reconstructed to image by finite discrete shearlet inverse transform, and both subjective visual evaluation and objective performance assessments of the fusion results were implemented. Simulation results indicate that the proposed algorithm is obviously superior on subjective visual effects. The values of entropy, mutual information quantity and edge similarity increase averagely by 1.4%,3.4% and 0.8%, respectively, compared with other fusion algorithms. It is superior to other fusion algorithms on objective evaluation.