期刊文献+

图像区域不变矩的快速计算方法 被引量:3

Fast method for computing invariant moments within image regions
原文传递
导出
摘要 为了实时提取图像中任意尺寸窗口区域的不变矩,提出了一种快速计算方法。该算法通过构造一组积分图像,避免了在直接计算中出现的大量重复运算,使得图像中任一子窗口区域的不变矩都可以通过组合几个查找表运算得到。由于查找表运算的加法和乘法次数是恒定的,从而使得不同尺寸窗口区域的不变矩的计算均能在相同时间内完成。实验结果表明,该方法不会引起计算精度的损失,极大地降低了计算复杂度。对于从300×300pixels图像中提取的所有81×81pixels的窗口区域的不变矩这一任务而言,快速算法仅需31ms,比直接计算的速度提高了324倍。 To compute invariant moments within all arbitrary size window regions in an image in real-time, a fast method is proposed. By constructing a set of integral images, the method removes the large number of repeated computations presented in the direct method, and evaluates the moments of any window region conveniently through several table lookup operations. Since the number of table lookup operations is constant, then the moments for different window region sizes can he computed in constant time. Experimental results show that the fast method can provide the same accuracy as the direct one while decreases the computational cost significantly. Specifically, in an image with a resolution of 300× 300 pixels, the fast method requires only 31ms to extract all the moments for 81 × 81 pixels window regions, which is 324 times faster compared to the direct method.
作者 尚海林
出处 《光学技术》 CAS CSCD 北大核心 2012年第6期756-760,共5页 Optical Technique
关键词 不变矩 积分图像 直接算法 快速算法 invariant moments integral image direct method fast method
  • 相关文献

参考文献7

  • 1Hu M K. Visual pattern recognition by moment invariants [J]. IEEE Transactions on Information Theory, 1962,8: 179-187.
  • 2Collins R T, Lipton A J, Kanade T, et al. A system for video surveillance and monitoring [R]. Technical Report, Carnegie Mellon University, CMU-RI-TR-00-12, 2000.
  • 3孔刚. 复杂背景下扩展目标跟踪技术研究[D]. 北京: 中国科学院博士学位论文, 2005.
  • 4Ferrando S, Gera G, Massa M, et al. A new method for real time abandoned object detection and owner tracking [C]∥ Proceedings of IEEE Int Conf Image Processing. 2006: 3329-3332.
  • 5Viola P, Jones M J. Robust real-time face detection [J]. International Journal of Computer Vision,2004, 57(2): 137-154.
  • 6Porikli F. Integral histogram: a fast way to extract histograms in Cartesian spaces [C]∥ Proceedings of IEEE Int Conf Computer Vision and Pattern Recognition. 2005.
  • 7Hussein M, Porikli F, Davis L. Kernel integral image: a framework for fast non-uniform filtering [C]∥ Proceedings of IEEE Int Conf Computer Vision and Pattern Recognition. 2008.

同被引文献21

  • 1Manjunath BS,Ohm JR,Vasudevan V,et al.Color and texture descriptors[J].IEEE Transaction on Circuits and Systems for Video Technology,2001,11(6):703-715.
  • 2Ferrando S,Gera G,Massa M.A new method for real time abandoned object detection and owner tracking[C]//Proceedings of IEEE International Conference on Image Processing,2006:3329-3332.
  • 3Rizon M,Yazid H,Saad P,et al.Object detection using geometric invariant moment[J].American Journal of Applied Sciences,2006,2(6):1876-1878.
  • 4Das D,Ghosh M,Chakraborty C,et al.Invariant moment based feature analysis for abnormal erythrocyte recognition[C]//International Conference on Systems in Medicine and Biology,2010,242-247.
  • 5Diego M,Arturo A,Gegundez-Arias ME,et al.A new supervised method for blood vessel segmentation in retinal images by using gray-level and moment invariants-based features[J].IEEE Transactions on Medical Imaging,2011,30(1):146-158.
  • 6Shen D,Horace HS.Discriminative wavelet shape descriptors for recognition of 2-D patterns[J].Pattern Recognition,1999,32(2):151-165.
  • 7Hu MK.Visual pattern recognition by moment invariants[J].IRE Transactions on Information Theoty,1962,8(2):179-187.
  • 8周振环,赵明.基于不变矩景象匹配算法[J].计算机工程,2007,33(22):203-205. 被引量:6
  • 9HU. M. K Visual Pattern Recognition by Momenl Invariants[J]. IRF Transactions on Information Theory, 1962 ( 8 ):179 -187.
  • 10苗静,杨勇,谷欣超,孙爽滋,杨兆龙,张子锐.不变矩及其在基于形状特征图像检索中的应用[J].长春理工大学学报(自然科学版),2009,32(1):126-128. 被引量:5

引证文献3

二级引证文献9

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部