An image invariant orthogonal moment based on a Walsh transform

被引：0
|
作者：
Li J. [1 ]
Chen H. [2 ]
Duan X. [2 ]
Liu H. [3 ]
Ren Y. [1 ]
机构：
[1] School of Mechanical Engineering, Northwestern Polytechnical University, Xi'an
[2] National Key Laboratory of Special Technology on UAV, Northwestern Polytechnical University, Xi'an
[3] School of Automation, Northwestern Polytechnical University, Xi'an
来源：
Harbin Gongcheng Daxue Xuebao/Journal of Harbin Engineering University | 2019年 / 40卷 / 10期
关键词：
Image characteristic; Image orthogonal moment; Image reconstruction; Sequency; Target recognition; Walsh image moment; Walsh transform;
D O I：
10.11990/jheu.201812082
中图分类号：
学科分类号：
摘要：
Aiming at the complex process of computer modeling of human visual system and the inefficiency of image engineering caused by the process of strong image feature extraction, in the application of machine vision, an image moment-Walsh image moment is proposed and deduced from the fast speed of image feature extraction and the stability of image feature itself, combined with invariant moment, orthogonal moment and image moment. The Walsh image moment is used as the transformation core. Based on Walsh transform, a fast method for calculating the moment of Walsh image is presented. The experimental results show that Walsh image moments have certain descriptive ability to images, and are invariant in translation, rotation and scaling. Their computational efficiency and accuracy are superior to those of orthogonal moments based on sine-cosine functions, which have obvious advantages in fast feature extraction of small images. At the same time, from the diversity of the permutation matrix in the construction process of Walsh image moments, the Walsh image moment has certain concealment and non-uniqueness for the calculation of the same image. © 2019, Editorial Department of Journal of HEU. All right reserved.
引用
收藏
页码：1784 / 1789
页数：5
相关论文
共 16 条
  • [1] Hu M.K., Visual pattern recognition by moment invariants, IRE Transactions on Information Theory, 8, 2, pp. 179-187, (1962)
  • [2] Prokop R.J., Reeves A.P., A survey of moment-based techniques for unoccluded object Representation and recognition, CVGIP: Graphical Models and Image Processing, 54, 5, pp. 438-460, (1992)
  • [3] Karakasis E.G., Amanatiadis A., Gasteratos A., Et al., Image moment invariants as local features for content based image retrieval using the Bag-of-Visual-Words model, Pattern Recognition Letters, 55, pp. 22-27, (2015)
  • [4] Wu H., Yan S., Computing invariants of Tchebichef moments for shape based image retrieval, Neurocomputing, 215, pp. 110-117, (2016)
  • [5] Upneja R., Accurate and fast Jacobi-Fourier moments for invariant image recognition, Optik, 127, 19, pp. 7925-7940, (2016)
  • [6] Wang X., Li B., Xu M., Et al., Quality assessment of Traditional Chinese Medicine using HPLC-PAD combined with Tchebichef image moments, Journal of Chromatography B, 1040, pp. 8-13, (2017)
  • [7] Gilmore C., Peters S., Lovetri J., Et al., GPR target detection using a neural network classifier of image moments as invariant features, Proceedings of 2004 International Symposium on Antenna Technology and Applied Electromagnetics and URSI Conference, pp. 1-4, (2004)
  • [8] Kushwaha A., Singh A., Shrivastav S.K., Signature classification using image moments, Progress in Intelligent Computing Techniques: Theory, Practice, and Applications, pp. 235-244, (2018)
  • [9] Guo L., Zhang X., Lin J., Et al., Spectrum characteristic analysis of Walsh code, Journal of Harbin Engineering University, 24, 5, (2003)
  • [10] Xiao B., Cui J., Qin H., Et al., Moments and moment invariants in the Radon space, Pattern Recognition, 48, 9, pp. 2772-2784, (2015)
12