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

 

The Facial Recognition in the Infrared Range

DOI 10.18127/j19997493-201902-02

Keywords:

D.A. Loktev – Ph.D.(Eng.), Associate Professor, Department «Information Systems and Telecommunications», Bauman Moscow State Technical University
E-mail: loktevdan@bmstu.ru


In this paper we propose and tested mathematical and algorithmic support of complex information-measuring monitoring system, which allows solving the problem of recognition of individuals from the passenger traffic on the objects of transport infrastructure in both optical and infrared wavelengths. The study of methods of processing images obtained in the infrared range, allows understanding whether they can be used to identify objects with additional external noise, which reduce the surface of the face, suitable for creating an image for recognition. The conditions for obtaining an image with sufficient quality for detecting the desired object are formulated. To obtain the primary image of the object under study, a Fluke TIX 580 thermal imager with a sensitivity of 0.05°C was used. The algorithm of recognition of people's faces based on the aggregated use of cascade classifiers using the adaptive gain algorithm and background subtraction to determine moving objects is considered. Depending on the distance from the passenger to the camera, it is proposed to use the categories «far» and «close» objects and, depending on the speed of movement of the passenger – «fast» and «slow». When using a similar classification of recognizable objects in this study, it is assumed that the speed of movement of passenger traffic and individual passengers is quite small, and the objects of recognition are located near the infrared detector. The background subtraction algorithm used the background representation as a set of Gaussian distributions, while considering the possibility of reducing the effect of light. To determine the contour of the object on the analyzed image, the canny operator was used in the work and the main stages of working with it were described. The developed facial recognition system contains the following modules: loading a series of images of passenger traffic, entering the parameters of recognition by the user, the background subtraction algorithm, the mode of matching subsequent frames, the detection of an individual passenger, object recognition or entering into the database. The user interface of the software package is developed, which allows its operator not to have special knowledge and competencies to solve the problems of detection, tracking and recognition of objects. In general, the proposed mathematical and algorithmic software can be used in complex monitoring and control systems in both optical and infrared wavelengths, and the formulated parameters for obtaining the primary image will solve the problem of detection and recognition of individual passengers.

References:
  1. Devyaterikov E.A., Mikhailov B.B. Sistema tekhnicheskogo zreniya dlya izmereniya puti mobilnogo robota. Mekhanika, upravlenie i informatika. 2012. № 8. S. 219−224.
  2. Loktev D.A. Opredeleniya parametrov ob'ekta po serii ego izobrazhenii v kompleksnoi sisteme monitoringa. Put i putevoe khozyaistvo. 2014. № 2. S. 31−33.
  3. Sun Z., Bebis G., Miller R. On-road vehicle detection using optical sensors: A review. Proceeding of the IEEE International Conference on Intelligent Transportation Systems. 2004. V. 6. P. 125−137.
  4. Favorskaya M.N., Shilov A.S. Algoritmy realizatsii otsenki dvizheniya v sistemakh videonablyudeniya. Sistemy upravleniya i informatsionnye tekhnologii (IPU RAN, VGTU). M.-Voronezh. 2008. № 3.3(33). S. 408−412.
  5. Hahne Uw. Real-Time Depth Imaging. Tu Berlin, Fakultät Iv, Computer Graphics. 2012. 108 p.
  6. Lindner M., Kolb A. Calibration of the Intensity-Related Distance Error of the PMD ToF-Camera. Proceedings of Spie, Intelligent Robots and Computer Vision. Boston, MA. USA. 2007. V. 6764. P. 56−64.
  7. Gil P., Pomares J., Torres F. Analysis and Adaptation of Integration Time in PMD Camera for Visual Servoing. International Conference on Pattern Recognition. 2010. P. 311−315.
  8. Muljowidodo K., Rasyid M.A., Sapto A.N., Budiyono A. Vision based distance measurement system using single laser pointer design for underwater vehicle. Indian J Mar Sci., 2009. V. 38(3). P. 324−331.
  9. Deschenes F., Ziou D., Fuchs P. Enhanced Depth from Defocus Estimation: Tolerance to Spatial Displacements. Rapport technique. № 256. Sherbrooke, Qc. Canada. Université de Sherbrooke. 2000. P. 34−42.
  10. Hu H., Haan G. Low Cost Robust Blur Estimator. Proceedings IEEE International Conference on Image Processing. Atlanta. 2006. P. 617−620.
  11. Zaman T. Depth Estimation from Blur Estimation. http://www.timzaman.com. 2012.
  12. Lin H.-Y., Chang C.-H. Depth from motion and defocus blur. Optical Engineering. December 2006. V. 45(12), N127201. P. 1−12.
  13. Borachi G., Caglioti V. Motion blur estimation at corners// Computer Vision Theory and Applications (VISAPP). 2008. P. 389−395.
  14. Murygin K.V. Osobennosti realizatsii algoritma AdaBoost dlya obnaruzheniya ob'ektov na izobrazheniyakh. Shtuchnii іntelekt (Donetsk). 2009. № 3. S. 573−581.
  15. Teplov A.V. Postroenie sistemy raspoznavaniya malorazmernykh ob'ektov v infrakrasnom diapazone opticheskogo spektra. Fundamentalnye i prikladnye issledovaniya: problemy i rezultaty. 2015. № 20. S. 93−95.
  16. Loktev A.A., Alfimtsev A.N., Loktev D.A. Algoritm raspoznavaniya ob'ektov. Vestnik MGSU. 2012. № 5. S. 194−201.
  17. Loktev A.A., Loktev D.A. Metod opredeleniya rasstoyaniya do ob'ekta putem analiza razmytiya ego izobrazheniya. Vestnik MGSU. 2015. № 6. S. 140−151.
  18. Tsurkov V.I., Grinchuk O.V. Tsiklicheskie generativnye neironnye seti dlya uluchsheniya raspoznavaniya lits v nestandartnykh domenakh. Izvestiya Rossiiskoi akademii nauk. Teoriya i sistemy upravleniya. 2018. № 4. S. 153−159.
  19. Rossikhin Yu.A., Shitikova M.V., Loktev A.A. The analysis of thin-walled building structures subjected to impact excitation. 4th International Ph.D. Symposium in Civil Engineering. 2002. S. 487−492.
  20. Loktev A.A., Zaletdinov A.V. Opredelenie tochek vzaimodeistviya pryamykh i otrazhennykh voln v plastinke. Vestnik MGSU. 2010. № 4−3. S. 303−308.
  21. Alfimtsev A.N., Loktev D.A., Loktev A.A. Sravnenie metodologii razrabotki sistem intellektualnogo vzaimodeistviya. Vestnik MGSU. 2013. № 5. S. 200−208.
  22. Alfimtsev A.N., Loktev D.A., Loktev A.A. Razrabotka polzovatelskogo interfeisa kompleksnoi sistemy videomonitoringa. Vestnik MGSU. 2012. № 11. S. 242−252.
  23. Loktev A.A., Bakhtin V.F., Chernikov I.Yu., Loktev D.A. Metodika opredeleniya vneshnikh defektov sooruzheniya putem analiza serii ego izobrazhenii v sisteme monitoringa. Vestnik MGSU. 2015. № 3. S. 7−16.
  24. Loktev D.A., Loktev A.A. Determination of object location by analyzing the image blur. Contemporary Engineering Sciences. 2015. T. 8. № 9. S. 467−475.
  25. Loktev A.A., Sychev V.P., Loktev D.A. K zadache proektirovaniya modulya vizualnogo raspoznavaniya elementov verkhnego stroeniya puti na vysokoskorostnykh magistralyakh. Transport Rossiiskoi Federatsii. 2017. № 1 (68). S. 22−26.
  26. Loktev D.A., Loktev A.A. Development of a user interface for an integrated system of video monitoring based on ontologies. Contemporary Engineering Sciences. 2015. T. 8. № 20. S. 789−797.

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