基于改进Canny特征点的SIFT算法

Improved SIFT algorithm based on Canny feature points

在分析了传统SIFT算法的基础上,提出了一种基于改进Canny特征点的SIFT算法。该方法先用SIFT算法检测出候选关键点,再利用Canny边缘检测算法检测出图像的边缘点,通过比较生成的候选关键点和边缘点的坐标是否相等,从而判断是否去除候选关键点。实验结果表明,与传统SIFT算法比较,该方法可以有效地去除DOG算子生成的边缘响应点,进一步提高SIFT特征点的抗噪声能力,增强SIFT算法的稳定性。

After the analysis of traditional SIFT algorithm, a new SIFT algorithm based on improving the Canny feature points is proposed. The algorithm firstly detects out candidate keypoints by SIFT algorithm, then finds out the edge points of image using Canny algorithm. It would remove some candidate keypoints by the analysis whether the coordinates are equal between candidate keypoints and edge points. The experimental results show that the improved SIFT algorithm can effectively remove the edge response points, further improve the ability of anti-noise for SIFT features, and display a more stability than the traditional SIFT algorithm.