Аннотация.

В работе исследован новый вид признаков для задач детекции и классификации объектов на изображении. Показана устойчивость данного класса признаков в ситуациях со значительными вариациями освещенности. Исследована точность модифицированного алгоритма Виолы-Джонса в задачах идентификации колесных осей транспортных средств и распознавания цифр, нанесенных выдавливанием или чеканкой.

Ключевые слова:

компьютерное зрение, детекция и классификация объектов, устойчивые признаки, алгоритм Виолы-Джонса.

Стр. 61-72.

S.A. Gladilin, A.A. Kotov, D.P. Nikolaev, S.A. Usilin

"Construction of robust features for detection and classification of objects without characteristic brightness contrasts"

The paper describes a new class of feature for object detection and classification in images. These features are shown to be robust to significant illumination changes. The performance of the proposed method is studied on the modified Viola-Jones algorithm as applied to vehicle axle detection and embossed number recognition tasks.

Keywords: computer vision, object detection, robust features, Viola-Jones algorithm

Полная версия статьи в Формате pdf.

REFERENCES

1. Viola P. and Jones M. Robust Real-time Object Detection, International Journal of Computer Vision. – 2001.
2. P. Viola, M. Jones, and D. Snow, Detecting Pedestrians Using Patterns of Motion and Appearance, International Journal of Computer Vision, vol. 63, no. 2, 2005, pp. 153–161.
3. D.C. Lee and T. Kanade, Boosted Classifier for Car Detection, 2007.
4. S. Usilin, D. Nikolaev, V. Postnikov, Lokalizatsiya, orientatsiya i identifikatsiya dokumentov s fiksirovannoy geometriey na izobrazhenii, Trudy Instituta sistemnogo analiza Rossiyskoy akademii nauk. Obrabotka informatsionnykh i graficheskikh resursov, 2010, str. 248-261.
5. A. Grigoryev, T. Khanipov, D. Nikolaev, Determination of axle count for vehicle recognition and classification, 8th Open German-Russian
6. C.P. Papageorgiou, M. Oren, and T. Poggio, A general framework for object detection, International Conference on Computer Vision, 1998, pp. 555–562.F.
7. Kopylov I. Ye., Tassov K. L., Opredelenie marki avtomobilya po videokadru s ispolzovaniem modifitsirovannogo algoritma Violy-Dzhonsa, Nauka i Obrazovanie – iyun, 2012
8. Ahonen T, Hadid A, Pietikainen M, Face Description with Local Binary Patterns: Application to Face Recognition, Pattern Analysis and Machine Intelligence, 2006, p. 2037 – 2041
9. F. A. Kornilov, K. V. Kostousov, D. S. Perevalov, Primenenie algoritma Violy-Dzhonsa s dvukhtochechnymi priznakami dlya poiska antropogennykh obektov v trave, Tekhnicheskoe zrenie v sistemakh upravleniya, 2012, s. 214216
10. Baluja S., Rowley H. A. Boosting sex identification performance // Intern. J. Computer Vision. 2007. V. 71(1). R. 111–119.
11. Buy Tkhi Tkhu Chang, Fan Ngok Khoang, V.G. Spitsyn, Raspoznavanie lits na osnove primeneniya metoda Violy-Dzhonsa, veyvlet-preobrazovaniya i metoda glavnykh komponent // Izvestiya Tomskogo politekhnicheskogo universiteta. – 2012. – T. 320. – № 5. – s. 5459
12. Dobeshi I. Desyat lektsiy po veyvletam. — Izhevsk: RKhD, 2001. — 464 s.
13. P.A. Skorynin, Detektory osobennostey v metode Violy-Dzhonsa, postroennye na konechnykh avtomatakh // Nauchno-tekhnicheskiy vestnik Sankt-Peterburgskogo gosudarstvennogo universiteta informatsionnykh tekhnologiy, mekhaniki i optiki, 2011, № 2 (72), c. 4044
14. Mealy George H. A Method to Synthesizing Sequential Circuits. — Bell Systems Technical Journal. — P. 1045–1079
15. Papageorgiou C.P., Oren M., and Poggio T. A general framework for object detection // International Conference on Computer Vision. – 1998. – pp. 555–562.
16. Crow, Summed-area tables for texture mapping, Proceedings of SIGGRAPH, vol. 18, no. 3, 1984, pp. 207–212.Y.
17. Freund and R.E. Schapire, A decision-theoretic generalization of on-line learning and an application to boosting, Proceedings of the Second European Conference on Computational Learning Theory, Barcelona, March, 1995, pp. 23–37.
18. Schapire R. E. and Singer Y. Improved Boosting Algorithms Using Confidence-rated Predictions // Machine Learning. – Vol. 37. – No. 3. – 1999. – pp. 297–336.
19. Nock R. and Nielsen F. A Real generalization of discrete AdaBoost // Artificial Intelligence. – Vol. 171. – No. 1. – 2007. – pp. 25–41.
20. J. Canny, A Computational Approach to Edge Detection, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. pami-8, no. 6, 1986, pp. 679–698.
21. B. Yane, Tsifrovaya obrabotka izobrazheniy. – Moskva: Tekhnosfera, 2007 – 584 s.