改进加权融合在消除图像拼接重影中的运用

The Application of Improved Weight Fusion to Eliminate Ghosting in Image Mosaics

为消除拼接图像间重叠部分出现的重影问题,本文提出了改进加权融合算法。首先,采用SURF算法对图像进行特征匹配,得到对应的不变矩阵H;然后,采用改进加权融合算法求取加权系数;最后,通过加权系数对待拼接图像进行融合,从而减少拼接融合时两幅图像的差异,以达到消除重影的目的。实验中分别采用了加权融合、动态规划寻找拼接线法、HSI算法以及本文提出的改进加权融合法进行测试。结果表明,所提算法能自适应的消除运动目标的干扰,灰度标准差、信息熵加权、清晰度加权评价指标相比其他融合拼接算法均有所提高,融合图像的质量与清晰度得到改善。

In this paper, the improved weighted fusion algorithm is proposed to eliminate ghosting between the mosaic images.Firstly,the SURF algorithm is used to match the features of the images,and then the corresponding invariant matrix H is obtained.Secondly, the weight coefficients are obtained by adopting the improved weight fusion algorithm.Finally, the weight coefficients are used to fuse the mosaic images, so as to reduce the difference between the two images in the mosaic fusion and achieve the goal of eliminating ghosting.In the experiment,the weight fusion,the dynamic programming,the HSI algorithm and the improved weight fusion method are adopted for testing.The results show that the proposed algorithm can eliminate the interference of adaptive moving objects,and the gray standard deviation,the information entropy and the definition weight evaluation indicators are effectively enhanced,compared with other image mosaic algorithms. Additionally, the quality and the clarity of the fusion images are improved.