Analysis of the functions realizing the correlation processing for the range measuring from digital images

Authors

  • Vladimir L. Kozlov Belarusian State University, Niezaliežnasci Avenue, 4, 220030, Minsk, Belarus

Abstract

Correlation processing of images is one of the most widely used and promising tools for searching, identifying, localizing, and tracking objects of relatively complex shapes. In the literary sources we have found no information about the possibilities to use various functions implementing image correlation processing to solve problems of measuring distances, dimensional parameters of objects, and constructing a depth map when one of the stereopair images is subjected to noise or monotonic amplitude shift. In this paper we solve the problem of analyzing the possibilities and applications of the functions realizing the correlation processing of digital optical images in the case of image distortions. The stu dies have been performed for the following functions: normalized cross-correlation function (NCC); the sum of absolute differences (SAD); the sum of squared differences (SSD); the normalized sum of squared difference (NSSD); RANK-conversion. An algorithm is proposed to reduce computational time of the normalized cross-correlation function NCC. It provides an analysis time comparable to that of SSD and SAD functions, where normalization is not used.

Author Biography

  • Vladimir L. Kozlov, Belarusian State University, Niezaliežnasci Avenue, 4, 220030, Minsk, Belarus

    doctor of science (technical), docent; professor at the department of quantum radiophysics and optoelectronics, faculty of radiophysics and computer technologies

References

  1. Hartley R., Zisserman A. Multiple view geometry in computer vision. Cambridge, 2004.
  2. Mamaev N., Lukin A., Yurin D. HeNLM-3D: A 3D computer tomography image denoising algorithm. Pattern Recognition and Information Processing (PRIPʼ2014) : proc. of the XII Intern. conf. (Minsk, 28–30 May, 2014) / ed. by A. Tuzikov, V. Kovalev. Minsk, 2014. P. 176–180.
  3. Kurgalin S. D., Krylovetsky A. A., Bozhok A. O. [Methods and algorithms for constructing three-dimensional models in computer vision systems]. Telematikaʼ2014 : proc. of the XXI All-Russian sci. and method. conf. (Saint Petersburg, 23–26 June, 2014). Saint Petersburg, 2014. P. 123–124 (in Russ.).
  4. Szeliski R. Computer vision: algorithms and applications. Berlin, 2011.
  5. Brown M. Z. Advances in computational stereo. IEEE trans. PAMI. 2003. Vol. 25, No. 8 P. 993–1008.
  6. Chumachenko A. V. [Optimization of SAD calculation for the problem of relief reconstruction from stereopairs images in high-performance stereovision systems]. Izv. Yuzhnogo federalʼnogo univ. Tekh. nauki. 2013. No. 3 (140). P. 89–96 (in Russ.).
  7. Guzik V. F., Chumachenko A. V. [Statistical method of optimization of local algorithms for establishing pixel correspondences on stereopairs]. Izv. vyssh. uchebn. zaved. Sev.-Kavk. reg. Ser.: Tekh. nauki. 2011. No. 4. P. 20–25 (in Russ.).
  8. Zabih R., Woodfill J. Non-parametric local transforms for computing visual correspondence. Computer Vision – ECCVʼ94 : proc. of Third Eur. conf. on computer vision (Stockholm, 2–6 May, 1994). Stockholm, 1994. P. 150–158.
  9. Middlebury stereo datasets [Electronic resource]. 2011. URL: http://vision.middlebury.edu/stereo/data (date of access: 14.09.2013).
  10. [Method for determining the distance to the object through a digital camera] : pat. 17809 Rep. Belarus, MKI G 01 C 3/00 / V. L. Kozlov, A. S. Vasilchuk ; declarer Belarus. state univ. No. а 20111773 ; declar. 20.12.11 ; publ. 09.09.13. Off. newsl. 2013. No. 5. P. 143 (in Russ.).

Downloads

Published

2017-09-29

Issue

No. 3 (2017): Journal of the Belarusian State University. Physics

Section

Ranging Physics

License

The authors who are published in this journal agree to the following:

  1. The authors retain copyright on the work and provide the journal with the right of first publication of the work on condition of license Creative Commons Attribution-NonCommercial. 4.0 International (CC BY-NC 4.0).
  2. The authors retain the right to enter into certain contractual agreements relating to the non-exclusive distribution of the published version of the work (e.g. post it on the institutional repository, publication in the book), with the reference to its original publication in this journal.
  3. The authors have the right to post their work on the Internet (e.g. on the institutional store or personal website) prior to and during the review process, conducted by the journal, as this may lead to a productive discussion and a large number of references to this work. (See The Effect of Open Access.)

How to Cite

(1)
Kozlov, V. L. Analysis of the Functions Realizing the Correlation Processing for the Range Measuring from Digital Images. Журнал Белорусского государственного университета. Физика 2017, No. 3, 102-110.