提出了一种全自动稳健的图像拼接融合算法。此算法采用Harris角检测算子进行特征点提取，使提取的精度达到了亚像素级，然后以特征点邻域灰度互相关法进行特征点匹配得到了初步的伪匹配集合，并运用稳健的RANSAC算法将伪匹配点集合划分为内点和外点，在内点域上运用LM优化算法精确地估计出了图像间的点变换关系，最后采用颜色插值对交接处进行颜色过渡。整个算法自动完成，它对有较大误差或错误的特征点数据迭代过滤，并用提纯后的数据来做模型估计，因而对图像噪声和特征点提取不准确有强健的承受能力。在参数估计时，以特征点的坐标位置误差而不是亮度误差来构造优化函数，克服了以往算法对光照的敏感性，使算法更具有实用性。实验结果表明，该算法融合效果比较理想，鲁棒性强，具有较高的使用价值。

In this paper, an automatic and robust image mosaic algorithm is presented. In this algorithm, Harris corner detector is used to extract feature points, which gains sub pixel precision for features extraction. Then, a pseudo matching set is obtained by comparing local neighborhoods of features through intensity cross correlation method and these pseudo matches are divided into inliers and outliers using robust RANSAC algorithm. In the inliers sub set, LM algorithm is used to estimate the point transformation matrix between two images accurately. In the end, the image color of the overlapping band is smoothed with bilinear interpolation technique. The whole algorithm is completed automatically. It filters the noisy or wrong input data iteratively, then estimates the model parameters through pure data, so it has strong error tolerant capacity for the image noise and inaccuracy of feature extraction. When estimating the model parameters, the energy function is constructed based on the position errors of features instead of the features' intensity errors, which conquers the original methods' shortcoming of sensitivity to illuminating conditions and makes this algorithm more practical. Experimental results show the image mosaic effect is wonderful and the algorithm is stable very much. It is high valuable in practice.