基于Contourlet变换的区域特征自适应图像融合算法

Regional Feature Self-Adaptive Image Fusion Algorithm Based on Contourlet Transform

Contourlet变换克服了小波变换在处理高维信号时的不足,比小波变换具有更好的方向性、较高的逼近精度和更好的稀疏表达性能。因此将Contourlet变换应用于图像融合领域,能更好的提取图像边缘特征,为融合提取更多的特征信息。基于Contourlet变换的区域特征自适应图像融合算法是将图像进行Contourlet变换分解后,针对不同的频率域特点选择不同的融合规则,针对高频系数特性选用了区域特征自适应的融合规则,最后通过重构得到融合图像。将基于小波变换的融合算法和本文所提算法进行了主观和客观的对比,结果表明,基于Contourlet变换区域特征自适应的图像融合算法是一种有效可行的图像融合算法。

Contourlet transform overcomes the weakness of wavelet transform in dealing with high-dimensional signals, it provides a flexible multiresolution, local and directional image expansion and a sparse representation for two-dimensional piecewise smooth signal resembling images. It can satisfy the anisotropy scaling relation for curves, and thus offers a fast and structured curvelet-like decomposition. When contourlet transform is applied to image fusion, the characteristic of original images can be effectively extracted and more important information is preserved. The fusion algorithm based on contourlet tranform can be divided into three steps. Firstly, the original images are decomposed with contourlet transform. Secondly, because different fusion rules fit different frequency bands, we designed the regional feature self-adaptive fusion rule is used in high-frequency domain. Finally the fused coefficients are reconstructed to obtain fusion results. Two sets images are taken as experimental data, subjective and objective standards are used to evaluate the results, and comparison with results based on wavelet transform is also carried out. The results show that this method gets better fusion results than wavelet tranform. And the regional feature selfadaptive image fusion algorithm based on contourlet transform is an effective and feasible algorithm.