Radiotekhnika
Publishing house Radiotekhnika

"Publishing house Radiotekhnika":
scientific and technical literature.
Books and journals of publishing houses: IPRZHR, RS-PRESS, SCIENCE-PRESS


Тел.: +7 (495) 625-9241

 

Recognition of separate test and real flat objects on basis of dimensionless marks of contours of their bitmaps by method of the Fischer’s discriminant analysis

Keywords:

S.S. Sadykov – Dr. Sc. (Eng.), Professor, Murom branch of Vladimir State University named after A.&N. Stoletovs. E-mail: sadykovss@yandex.ru Ya.Yu. Kulkov – Senior Lecturer, Murom branch of Vladimir State University named after A.&N. Stoletovs. E-mail: y_mail@mail.ru


In this article for recognition of different real flat objects it is offered to use the marks created on the basis of characteristics by contours of objects images. The purpose of work is the experimental study of recognition of flat objects by method of the Fischer’s discriminant analysis, using dimensionless marks of contours of their bitmaps and definition of a possibility to use of this method in vision systems in industrial. The technology of realization of the offered algorithm of recognition separate test and real flat objects (details and products) consists in carrying out classification a set of entrance objects by method of the linear Fischer’s discriminant analysis. The basis of reference object to a cluster is the greatest value of the so-called simple classifying function for k’s class which is a linear combination of discriminant variables. For carrying out experimental studies casual emergence of object in cameras sight of a vision system is imitated. Presentational selection on 2000 images of each class is for this purpose formed. The received results indicate that for successful use of a method Fischer’s discriminant analysis in vision system of industrial function at the solution of a problem of classification it is necessary to pick up a set of dimensionless marks of objects more carefully.
References:

 

  1. Ermakov A.A., Orlov A.A., Sadykov S.S., Starodubov D.N. Metody i algoritmy obrabotki i analiza defektoskopicheskikh i metallograficheskikh snimkov. Vladimir: Izd-vo VlGU. 2008. 112 s.
  2. Sadykov S.S., Stulov N.N. Metody i algoritmy vydelenija priznakov obektov v STZ. M.: Gorjachaja linija - Telekom. 2005. 204 s.
  3. Sadykov S.S. Formirovanie bezrazmernykh koehfficientov formy zamknutogo diskretnogo kontura // Algoritmy, metody i sistemy obrabotki dannykh. 2014. № 4 (29). S. 91−98.
  4. Sadykov S.S., Zakharova E.A., Bulanova JU.A. Tekhnologija vydelenija oblasti kisty na mammogramme // Vestnik Rjazanskogo gosudarstvennogo radiotekhnicheskogo universiteta. 2013. № 1 (43). S. 7−12.
  5. Sadykov S.S., Bulanova JU.A., Zakharova E.A., JAshkov V.S. Issledovanie markernogo vodorazdela dlja vydelenija oblasti raka molochnojj zhelezy // Algoritmy, metody i sistemy obrabotki dannykh. 2013. № 1 (23). S. 56−64.
  6. Andrianov D.E., Sadykov S.S., Simakov R.A. Razrabotka municipalnykh geoinformacionnykh sistem. M.: Mir. 2006. 109 s.
  7. Sadykov S.S., Kulkov JA.JU. EHksperimentalnoe issledovanie algoritma raspoznavanija otdelnykh testovykh ploskikh obektov na osnove ikh bezrazmernykh konturnykh priznakov // Algoritmy, metody i sistemy obrabotki dannykh. 2015. № 3 (32). S. 76−90.
  8. Duda R., KHart P. Raspoznavanie obrazov i analiz scen / Pod red. V.L. Stefanjuka. M.:Mir. 1976. 509 s.
  9. Sadykov S.S., Beljakova A.S. Regressionnye modeli stenokardii i zavisimost ikh informativnosti ot kolichestva parametrov raboty serdca // Sistemy upravlenija i informacionnye tekhnologii. 2011. T. 45. № 3.1. S. 190−194.
  10. Evstigneeva O.I., Sadykov S.S., Suslova E.E., Beljakova A.S. Kriterii vydelenija grupp riska iz lic trudosposobnogo vozrasta pri medicinskikh issledovanijakh na sisteme ASPO // Algoritmy, metody i sistemy obrabotki dannykh. 2012. № 19. S. 33−39.
  11. ZHiznjakov A.L., Sadykov S.S. Teoreticheskie osnovy obrabotki mnogomasshtabnykh posledovatelnostejj cifrovykh izobrazhenijj // Vladimir: Izd-vo VlGU. 2008. 121 s.
  12. Kanunova E.E., Orlov A.A., Sadykov S.S. Metody i algoritmy restavracii izobrazhenijj arkhivnykh testovykh dokumentov. M.: 2006. 105 s.

 

© Издательство «РАДИОТЕХНИКА», 2004-2017            Тел.: (495) 625-9241                   Designed by [SWAP]Studio