模式图像的特征提取与配准方法

发布时间：2018-08-02 09:34

【摘要】：目的提高产品包装模式图像的匹配速度和准确性,得出算法中影响这两方面的关键因素。方法运用高斯金字塔对原始图像进行向下采样,将图像转换为尺度空间的表示方式;利用加速稳健特征(SURF)方法提取特征点,并对SURF方法提取的特征点用Brute-Force算法进行匹配;通过对每组匹配点欧式距离的排序,剔除误匹配点,仅保留一定数量的正确匹配点对。结果得出高斯金字塔、Hessian值、颜色空间对计算速度及获取特征点数量的影响。颜色空间的改变不影响特征点数量,在计算时间上的差别也可忽略。尺度空间缩小1/2时,计算速度能够提升75%,同时剔除2/3的冗余特征点。结论文中方法能够有效提升特征提取速度,并且具有匹配精度高、鲁棒性强的特点,同时对旋转角度的变化具有较强的适应性。
[Abstract]:Aim to improve the matching speed and accuracy of product packaging pattern images and to find out the key factors affecting these two aspects in the algorithm. Methods the original image was sampled downwards by Gao Si pyramid, the image was transformed into the representation of scale space, the feature points were extracted by accelerated robust feature (SURF) method, and the feature points extracted by SURF method were matched by Brute-Force algorithm. By sorting the Euclidean distance of each group of matching points, the mismatch points are eliminated and only a certain number of correct matching points are retained. The results show the influence of Gao Si pyramid Hessian value and color space on the calculation speed and the number of feature points. The change of color space does not affect the number of feature points, and the difference in computing time can be ignored. When the scale space is reduced by 1 / 2, the computation speed can be increased by 75%, and the redundant feature points of 2 / 3 can be eliminated at the same time. Conclusion this method can effectively improve the speed of feature extraction, and has the characteristics of high matching accuracy, strong robustness, and strong adaptability to the change of rotation angle.
【作者单位】：北京工商大学;
【基金】：2016年研究生科研能力提升计划
【分类号】：TP391.41

【相似文献】