基于Gabor能量空间的尺度不变兴趣点检测方法

曹政才; 马逢乐; 付宜利; 张剑

doi:10.3724/SP.J.1004.2014.02356

基于Gabor能量空间的尺度不变兴趣点检测方法

doi: 10.3724/SP.J.1004.2014.02356

曹政才^1,2,,
马逢乐^1,2,
付宜利²,
张剑³

1.
北京化工大学信息科学与技术学院北京 100029;
2.
哈尔滨工业大学机器人技术与系统国家重点实验室哈尔滨 150006;
3.
同济大学机械与能源工程学院上海 200092

基金项目:

Supported by National Natural Science Foundation of China (61105089), State Key Laboratory of Robotics and System (SKLRS-2013-ZD-03), and Open Foundation of the State Key Laboratory of Fluid Power Transmission and Control (GZKF-201212)

计量
- 文章访问数: 1472
- HTML全文浏览量: 68
- PDF下载量: 1248
- 被引次数: 0
出版历程
- 收稿日期: 2013-09-09
- 修回日期: 2014-04-23
- 刊出日期: 2014-10-20

A Scale Invariant Interest Point Detector in Gabor Based Energy Space

1.
College of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, China;
2.
State Key Laboratory of Robotics and Systems, Harbin Institute of Technology, Harbin 150001, China;
3.
School of Mechanical Engineering, Tongji University, Shanghai 200092, China

Funds:

More Information

Corresponding author: CAO Zheng-Cai Professor at the College of Information Science and Technology, Beijing University of Chemical Technology. He received his Ph. D. degree from Harbin Institute of Technology in 2005. His research interest covers sensor technology and intelligent control of robot. Corresponding author of this paper. E-mail: giftczc@163.com

摘要

摘要: 兴趣点检测是中层视觉感知过程的关键步骤,也是众多机器视觉系统的重要组成部分.此前的大多数兴趣点检测子都是针对特殊的二维图像结构设计的,比如角点、交叉点、端点等,所以对与其差别较大的特征不能检测.采用在Gabor能量空间中迭代搜索的方法,本文提出了一种尺度不变兴趣点检测子.基于结构不同的二维图像特征在相频域中表现相似的特点,该检测子能检测大多数特征.首先,基于Gabor滤波器响应获得一系列能量图像,通过极值点检测得到候选兴趣点;其次,使用一种迭代方法同时选择特征尺度与精确定位特征点位置;最后为了提高算法的实时性,采用了一种递推方法加速能量图像的计算过程.实验结果表明相对于其它检测子,本文提出的方法具有更广泛的适应性,并且在旋转、尺度、光照等变化下具有良好的稳定性.
- 兴趣点检测 /
- Gabor滤波器 /
- 能量图像 /
- 尺度不变性
Abstract: Interest point detection is a fundamental issue in many intermediate level vision problems and plays a significant role in vision systems. The previous interest point detectors are designed to detect some special image structures such as corners, junctions, line terminations and so on. These detectors based on some simplified 2D feature models will not work for image features that differ significantly from the models. In this paper, a scale invariant interest point detector, which is appropriate for most types of image features, is proposed based on an iterative method in the Gabor based energy space. It detects interest points by noting that there are some similarities in the phase domain for all types of image features, which are obtained by different detectors respectively. Firstly, this approach obtains the positions of candidate points by detecting the local maxima of a series of energy maps constructed by Gabor filter responses. Secondly, an iterative algorithm is adopted to select the corresponding characteristic scales and accurately locate the interest points simultaneously in the Gabor based energy space. Finally, in order to improve the real-time performance of the approach, a fast implementation of Gabor function is used to accelerate the process of energy space construction. Experiments show that this approach has a broader applicability than the other detectors and has a good performance under rotation and some other image changes.
- Interest point detection /
- Gabor filter /
- energy map /
- scale invariant

HTML全文

参考文献(13)

[1]	Schmid C, Mohr R, Bauckhage C. Evaluation of interest point detectors. International Journal of Computer Vision, 2000,37(2): 151-172
[2]	[2] Mikolajczyk K, Schmid C. An affine invariant interest point detector. In: Proceedings of the 7th European Conference on Computer Vision. Copenhagen, Denmark: 2002. 128-142
[3]	[3] Mikolajczyk K, Schmid C. Scale and affine invariant interest point detectors. International Journal of Computer Vision, 2004,1(60): 63-86
[4]	[4] Lowe D G. Distinctive image features from scale-invariant keypoint. International Journal of Computer Vision, 2004,60(2): 91-110
[5]	[5] Bay H, Ess A, Tuytelaars T, Van G L. Speeded-up robust features (Surf). Computer Vision and Image Understanding, 2008,110(3): 346-359
[6]	[6] Tuytelaars T, Van G L. Matching widely separated views based on affine invariant regions. International Journal of Computer Vision, 2004,59(1): 61-85
[7]	[7] Peter Kovesi. Phase congruency: a low level image invariant. Psychological Research, 2000,64(2): 136-148
[8]	[8] Zaafouri A, Sayadi M, Fnaiech F. On the choice of filters bank for phase congruency computing in noisy images. In: Proceedings of the 16th IEEE Mediterranean Electrotechnical Conference (MELECON). Medine Yasmine Hammamet, Tunisia: IEEE, 2012. 129-132
[9]	[9] Wang K, Xiao P, Feng X, Wu G. Image feature detection from phase congruency based on two-dimensional Hilbert transform. Pattern Recognition Letters, 2011,32(15): 2015 -2024
[10]	Malik J, Sainarayanan G. Corner detection using phase congruency features. In: Proceedings of the 17th IEEE International Conference on Signal and Image Processing (ICSIP). Hong Kong, China: IEEE, 2010. 217-221
[11]	Du P M, Shi X L, Wang N N, Deng R J. Iris recognition based on principal phase congruency. In: Proceedings of the 24th Chinese Control and Decision Conference (CCDC). Shenyang, China: 2012. 1159-1162
[12]	Shojaeilangari S, Yan W Y, Li J, Eamkhwang T. Feature extraction through Binary Pattern of Phase Congruency for facial expression recognition. In: Proceeding of the 12th International Conference on Control, Automation, Robotics and Vision (ICARCV). Shenzhen, China: 2012. 166-170
[13]	Young I T, Vliet L J. Recursive implementation of the Gausian filter. Signal Processing, 1995,44: 139-151