用于图像匹配的改进Harris特征点检测算法
|
摘要
原始Harris特征点检测算法采用高斯滤波进行平滑处理,增强了其鲁棒性,但是也提高了该算法的复杂度,导致其不能应用到许多图像匹配系统中,还存在对T型和斜T型特征点定位不准确的问题。为此,提出一种新的特征点检测算法。使用加速分割测试特征的特征点检测原理排除大量的非特征点,利用邻域像素比较法消除部分强干扰点,采用改进的高效非极大值抑制算法获得结果特征点。实验结果表明,该算法具有较好的匹配精度和较快的检测速度,检测时间仅为原始Harris算法的13.9%,适用于实时图像匹配系统。
By using Gaussian filtering for smooth processing,the original Harris feature point detection algorithm enhances its robustness.But it also increases the complexity of the algorithm which can not be applied to many image matching systems.Its positioning accuracy of T-type and diagonal T-type feature points is low.In order to solve the above problems,a new feature point detection algorithm is proposed.Amounts of non-feature points are excluded by using the principle of Features from Accelerated Segment Test(FAST)feature point detection.Some strong interference points are ruled out by using neighborhood pixels comparison method.The resulting feature points are obtained by using the improved efficient non-maximum suppression algorithm.Experimental results demonstrate that the improved algorithm has better matching accuracy and higher detection speed,its detection time is only approximately13.9%that of the original Harris algorithm and it is quite suitable for real-time image matching systems.
By using Gaussian filtering for smooth processing,the original Harris feature point detection algorithm enhances its robustness.But it also increases the complexity of the algorithm which can not be applied to many image matching systems.Its positioning accuracy of T-type and diagonal T-type feature points is low.In order to solve the above problems,a new feature point detection algorithm is proposed.Amounts of non-feature points are excluded by using the principle of Features from Accelerated Segment Test(FAST)feature point detection.Some strong interference points are ruled out by using neighborhood pixels comparison method.The resulting feature points are obtained by using the improved efficient non-maximum suppression algorithm.Experimental results demonstrate that the improved algorithm has better matching accuracy and higher detection speed,its detection time is only approximately13.9%that of the original Harris algorithm and it is quite suitable for real-time image matching systems.
引文
[1]吴振国,杨红乔.一种具有仿射不变性的图像匹配算法[J].计算机工程,2013,39(8):215-218.
[2]曾峦,王元钦,谭久彬.改进的SIFT特征提取和匹配算法[J].光学精密工程,2011,19(6):1391-1397.
[3]Moravec H P.Towards Automatic Visual Obstacle Avoidance[C]//Proceedings of the 5th International Joint Conference on Artificial Intelligence.Cambridge,USA:MIT Press,1977:584-590.
[4]Harris C,Stephens M.A Combined Corner and Edge Detector[C]//Proceedings of the 4th Alvey Vision Conference.Manchester,UK:[s.n.],1988:147-151.
[5]Rosten E,Porter R,Drummond T.Faster and Better:A Machine Learning Approach to Corner Detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(1):105-119.
[6]Mokhtarian F,Suomela R.Robust Image Corner Detec-tion Through Curvature Scale Space[J].IEEE Transac-tions on Pattern Analysis and Machine Intelligence,1998,20(12):1376-1381.
[7]He Xiaochen.Corner Detector Based on Global and Local Curvature Properties[J].Optical Engineering,2008,47(5).
[8]Awrangjeb M,Lu Guojun.Robust Image Corner Detection Based on the Chord-to-point Distance Accumulation Technique[J].IEEE Transactions on Multimedia,2008,10(6):1059-1072.
[9]王崴,唐一平.一种改进的Harris特征点提取算法[J].光学精密工程,2008,16(10):1995-2001.
[10]Neubeck A,Gool L V.Efficient Non-maximum Suppres-sion[C]//Proceedings of the 18th International Con-ference on Pattern Recognition.Washington D.C.,USA:IEEE Press,2006:850-855.
[11]Mokhtarian F,Mohanna F.Performance Evaluation of Corner Detectors Using Consistency and Accuracy Measures[J].Computer Vision and Image Understand-ing,2006,102(1):81-94.
[12]Tsai Duming,Lin Chienta.Fast Normalized Cross Correlation for Defect Detection[J].Pattern Recognition Letters,2003,24(15):2625-2631.
[2]曾峦,王元钦,谭久彬.改进的SIFT特征提取和匹配算法[J].光学精密工程,2011,19(6):1391-1397.
[3]Moravec H P.Towards Automatic Visual Obstacle Avoidance[C]//Proceedings of the 5th International Joint Conference on Artificial Intelligence.Cambridge,USA:MIT Press,1977:584-590.
[4]Harris C,Stephens M.A Combined Corner and Edge Detector[C]//Proceedings of the 4th Alvey Vision Conference.Manchester,UK:[s.n.],1988:147-151.
[5]Rosten E,Porter R,Drummond T.Faster and Better:A Machine Learning Approach to Corner Detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(1):105-119.
[6]Mokhtarian F,Suomela R.Robust Image Corner Detec-tion Through Curvature Scale Space[J].IEEE Transac-tions on Pattern Analysis and Machine Intelligence,1998,20(12):1376-1381.
[7]He Xiaochen.Corner Detector Based on Global and Local Curvature Properties[J].Optical Engineering,2008,47(5).
[8]Awrangjeb M,Lu Guojun.Robust Image Corner Detection Based on the Chord-to-point Distance Accumulation Technique[J].IEEE Transactions on Multimedia,2008,10(6):1059-1072.
[9]王崴,唐一平.一种改进的Harris特征点提取算法[J].光学精密工程,2008,16(10):1995-2001.
[10]Neubeck A,Gool L V.Efficient Non-maximum Suppres-sion[C]//Proceedings of the 18th International Con-ference on Pattern Recognition.Washington D.C.,USA:IEEE Press,2006:850-855.
[11]Mokhtarian F,Mohanna F.Performance Evaluation of Corner Detectors Using Consistency and Accuracy Measures[J].Computer Vision and Image Understand-ing,2006,102(1):81-94.
[12]Tsai Duming,Lin Chienta.Fast Normalized Cross Correlation for Defect Detection[J].Pattern Recognition Letters,2003,24(15):2625-2631.