(27) * << * >> * Russian * English * Content * All Issues

Vision system for the recognition of numbers of rail wagons using a modified correlation function in the Hausdorff metric

S.G. Volotovsky 1, 2, N.L. Kazanskiy 1, 2, S.B. Popov 1, 2, R.V. Khmelev 2
1Image Processing Systems Institute of RAS

2Samara State Aerospace University (SSAU)

 PDF, 496 kB

Pages: 177-184.

Full text of article: Russian language.

Abstract:
The task of automating the entire technological process of filling the railway tank cars is currently taking priority at oil loading terminals. The incoming tank cars have unique eight-digit numbers, two- or three-digit codes of the tank type (they determine the dimensions and capacity) and two-digit codes of the country of origin. This information regarding all the tanks of the arriving trains should be entered into the automated process control system as soon as possible..

Keywords:
Hausdorff metric, modified correlation, vision system, two-digit codes, automated process control system.

Citation:
Volotovsky SG, Kazanskiy NL, Popov SB, Khmelev RV. Vision system for the recognition of numbers of rail wagons using a modified correlation function in the Hausdorff metric. Computer Optics 2005; 27: 177-184.

Acknowledgements:
This work was supported by the Russian-American program Basic Research and Higher Education (BRHE), grants by the President of Russia No. НШ-1007.2003.01 and the Russian Foundation for Basic Research No. 04-07-90149 and No. 04-07-96500.

References:
  1. Volotovsky SG, Kazanskiy NL, Popov SB. System for registration of oil tank wagons. Certificate of official software registration No 2004611969 based on the application No 2004611381 of June 29, 2004. Registered in the Software registry on August 26, 2004.
  2. Volotovsky SG, Kazanskiy NL, Popov SB. Software for the recognition of numbers based on the analysis of layout topology. Certificate of official software registration No 2004611970 based on the application No 2004611382 of June 29, 2004. Registered in the Software registry on August 26, 2004.
  3. Khmelev RV, Kazanskiy NL, Popov SB. Software for the recognition of numbers based on the analysisof mutual deviations of geometric shapes of the object and sample. Certificate of official software registration No 2004611971 based on the application No 2004611383 of June29, 2004. Registered in the Software registry on August 26, 2004.
  4. Huttenlocher DP, Klauderman GA, Rucklidge WJ. Comparing images using the Hausdorff distance. IEEE Trans Pattern Anal Mach Intell 1993; 15(9): 850-863. DOI: 10.1109/34.232073.
  5. Kazanskiy NL, Khmelev RV. Comparison of an object and asample for the deviation of the contours [In Russian]. Computer Optics 2000; 20: 134-139.
  6. Mauch S. A fast algorithm for computing the closest point and distance transform. 2000. Source: http://www.acm.caltech.edu/~seanm/software/cp`t/cpt.pdf.
  7. Felzenszwalb PF, Huttenlocher DP. Distance transforms of sampled functions. cornell computing and information science Technical Report TR2004-1963 2004.
  8. Kazanskiy NL, Myasnikov VV, Khmelev RV. Algorithms for calculating distances to object pixels in binary images. Computer Optics 2000; 20: 128-133.
  9. Khmelev RV. The combined use of structural analysis and Hausdorff metrics in comparison of an object to a sample [In Russian]. Computer Optics, 2005; 27: 174-176.

© 2009, IPSI RAS
151, Molodogvardeiskaya str., Samara, 443001, Russia; E-mail: ko@smr.ru ; Tel: +7 (846) 242-41-24 (Executive secretary), +7 (846) 332-56-22 (Issuing editor), Fax: +7 (846) 332-56-20