基于边缘特征的多传感器图像融合算法

针对传统像素级图像融合方法在低频系数融合中，采用偏袒法和平均法容易导致融合图像出现模糊、对比度下降的问题，结合像素级和特征级融合的优势，提出一种基于边缘特征的图像融合算法。算法对于低频系数，采用区域能量自适应加权的方法；对于高频系数，通过对低频边缘特征的融合以指导其融合。分别对红外与可见光图像和多聚焦图像进行实验，并对融合图像进行主客观评价，实验表明，该算法得到的融合图像具有较好的主观视觉效果和客观量化指标，融合性能优于传统的融合方法。

Multi-sensor image fusion algorithm based on edge feature

Specific to the drawback that favoritism and average methods for low frequency coefficient fusion are weaken in maintaining the contrast of fusion image in traditional signal level image fusion, combining with the superiorities of signal level and feature level fusion, a novel fusion algorithm based on edge feature was proposed. Firstly, the registered multi-sensor images from the same scene were transformed by wavelet transforms. Secondly, the high and low frequency coefficients were fused separately by using different fusion strategies: the low frequency coefficient was fused by adaptive regional energy, while the high frequency coefficient fusion was conducted by using the edge feature fusion of low frequency coefficient. Finally, the target image was obtained by performing inverse wavelet transforms. The algorithm has been used to fuse infrared and visible images, and multi-focus images. The experimental results indicate that the fused image obtained by the proposed method has a better subjective visual effect and objective evaluation criteria, it performs dramatically better than traditional fusion methods.