Investigation of methods for correcting geometric distortion in a video

S. E. Myasnikov, K.V. Ezhova

Abstract


The article discusses the existing methods of correcting radial distortion for video images. In the modern world, visual data plays a key role in various fields, including medicine, the automotive industry, computer vision, artificial intelligence, and many others. However, when images are obtained, distortions occur, which can seriously affect the accuracy and reliability of data analysis. One of the most common types of distortion is image distortion – distortion of the geometric shapes of objects in the image due to imperfections of optical systems or distortion during data transmission. The study of image distortion correction methods is an urgent and important task in the field of computer vision and image processing. Effective methods of distortion correction allow you to improve the quality of images, while maintaining important geometric characteristics of objects on them. The exact correction of radial distortion is important especially for many optical recognition tasks, primarily because distortion distorts straight lines, and many algorithms for selecting and analyzing objects are built on the detection of rectilinear segments. At the same time, for processing a video stream with aberrations, several strict requirements are imposed both at the hardware level and at the software implementation level. For video surveillance cameras, such requirements are their technical characteristics, which should provide a sufficiently high quality of the transmitted video stream for the tasks of subsequent recognition of objects on it. For software, this is the required speed for real–time operation and the required accuracy of algorithms. In this paper, an in-depth analysis of existing methods of distortion correction is carried out, their applicability and acceptable parameters of optical systems with parameters of the output video image are investigated. The first section discusses the concept of radial distortion and its types. The second section presents the main approaches to correcting radial distortion. The third section contains a description of the concept of the system being developed and the prospects for its further development within the framework of research work.

 


Full Text:

PDF (Russian)

References


Edge detection based mobile robot indoor localization / Oleg Shipitko [и др.] // Institute for Information Transmission Problems – 2018.

Kunina I. A., Models and algorithms for calibration of the radial distortion of the camera according to the features of the Half-spectrum of images in uncontrolled shooting // Dissertation for the degree of Candidate of Technical Sciences (2020).

Rodionov S.A. Fundamentals of optics. Lecture notes.– St. Petersburg: SPb GITMO (TU), 2000. - 167 p.

Nghia T. Vo, Robert C. Atwood, and Michael Drakopoulos. Radial lens distortion correction with sub-pixel accuracy for x-ray micro-tomography. Opt. Express, 23(25):32859–32868, Dec 2015. URL: https://www.osapublishing.org/abstract.cfm?uri=oe-23-25-32859, doi:10.1364/OE.23.032859.

T. A. Clarke and J. G. Fryer. The development of camera calibration methods and models. The Photogrammetric Record, 16(91):51–66, 1998. URL: https://onlinelibrary.wiley.com/doi/epdf/10.1111/0031-868X.00113, doi:10.1111/0031-868X.00113

Carlos Ricolfe-Viala and Antonio-Jose Sanchez-Salmeron. Lens distortion models evaluation. Appl. Opt., 49(30):5914–5928, Oct 2010. URL: http://www.osapublishing.org/ao/abstract.cfm?URI=ao-49-30-5914, doi:10.1364/AO.49.005914

A. Criminisi, I. Reid, and A. Zisserman. A plane measuring device. Image and Vision Computing, 17(8):625–634, 1999. URL: https://www.sciencedirect.com/science/article/pii/S0262885698001838, doi:https://doi.org/10.1016/S0262-8856(98)00183-8

Nghia T. Vo. Python implementation of distortion correction methods for x-ray tomography. zenodo.org, 2018. URL: https://zenodo.org/record/1322720#.YV7_k9rTVaQ, doi:10.5281/zenodo.1322720.

Donald G. Bailey. A new approach to lens disortion correction. Proceedings of Image and Vision Computing New Zealand Conference, pages 59–64, 2002. URL: https://www-ist.massey.ac.nz/dbailey/sprg/pdfs/2002_IVCNZ_59.pdf.


Refbacks

  • There are currently no refbacks.


Abava  Кибербезопасность MoNeTec 2024

ISSN: 2307-8162