Moving object detection by using GPS assisted image registration method

Öz

In this study, a Global Positioning System (GPS) assistance based system has been developed for unmanned ground vehicles (UGV) to detect moving objects along their route using a computer vision system. Before the real time application of the UGV, image models were created that represents a default background in specified horizontal positions of the specified coordinates on the route. This model is a type of feature matrix which is much smaller than the pure image matrices. The model matrices were recorded in the system database and a database relation was created between the model and its coordinate. The feature matrices of the images captured when the moment UGV arrived to the determined coordinates are compared with the models belong to present coordinate. As a result of the evaluation image frames are aligned with 2D image registration methods. The silhouettes of the objects are obtained by subtracting aligned frames. Thus, using this developed approach, there is no need for costly solutions to compensate for the noise generated by the moving camera. It was observed from the experiments that the system was able to detect the objects with 90% accuracy and it was able to run with 8% CPU loading and 0.057 s processing time per frame.

GPS destekli imge çakıştırma yöntemi ile hareketli nesnelerin tespiti

Öz

Bu çalışmada insansız kara araçlarının rotaları üzerinde bulunan hareketli nesneleri bilgisayar görme sistemleri kullanarak tespit edebilmesi için küresel konumlandırma sistemi destekli bir sistem geliştirilmiştir. Gerçek zamanlı çalışma öncesinde rota üzerinde belirlenen tüm koordinatlarda mobil aracın geçebileceği tüm yatay konumlardaki yalın arka planı temsil eden bir görüntü modeli oluşturulmuştur. Bu model görüntü matrislerinden çok daha küçük boyutlarda bir referans öznitelik matrisidir. Bu model elde edildiği koordinat bilgisi ile ilişkilendirilerek bilgisayar sistemi veri tabanında tutulur. Mobil aracın gerçek zamanlı hareketi sırasında belirlenen koordinatlara geldiği anda yakalan anlık görüntülerin öznitelik matrisleri ile koordinatla ilişkilendirilen model karşılaştırılır. İşlem sonucunda 2 boyutlu imge çakıştırma yöntemleri kullanılarak imge çerçeveleri hizalanır. Hizalanan çerçevelerin farkları alınarak sahneye sonradan dahil olan dinamik nesnelerin siluetleri elde edilir. Geliştirilen yaklaşım ile rota üzerindeki her bir koordinat için hafızada tutulan referans bilgiler sayesinde hareketli kameraların oluşturduğu gürültünün dengelenmesi için karmaşık ve yüksek maliyetli işlemlere gerek duyulmamaktadır. Yapılan deneysel çalışmalarda geliştirilen sistemin anlık arka plan imge çerçevesinde bulunan engelleri %90 doğrulukta algılayabildiği ve düşük maliyetli bilgisayar sistemleri ile %8 işlemci yükü ve 0.057 sn. çerçeve başına işlem süresi ile çalışabildiği anlaşılmıştır.

Anahtar Kelimeler

• Görüntü işleme, İnsansız kara aracı, Hareketli nesne tespiti, İmge çakıştırma

Kaynakça

1. Shimizu S, Yamamoto K, Wang C, Satoh Y, Tanahashi H, Niwa Y. “Moving object detection by mobile stereoomni- directional system (SOS) using spherical depth image”. Pattern Analysis and Applications, 9(2-3),113-126, 2006.
2. Gamez DAM, Devy M. “Active vision-based moving objects detection from a motion grid”. Mobile Robots (ECMR). IEEE 2013 European Conference, Barcelona, Spain, 25-27 September 2013.
3. Yu Q, Medioni G. “Map-Enhanced detection and tracking from a moving platform with local and global data association”. IEEE Workshops on Motion and Video Computing, Austin, TX, USA, 23-24 February 2007.
4. Kong H, Audibert J, Ponce J. “Detecting abandoned objects with a moving camera”. IEEE Transactions on Image Processing, 19(8), 2201-2210, 2010.
5. Foresti GL, Gentili S. “A vision based system for object detection in underwater images”. International Journal of Pattern Recognition, 14(2), 167-188, 2000.
6. Chen X. “Application of matlab in moving object detection algorithm”. Future BioMedical Information Engineering 2008 FBIE '08 International Seminar on, Wuhan, China, 18 December 2008.
7. Jarraya SK, Hammami M, Ben-Abdallah H. “Accurate background modeling for moving object detection in a dynamic scene”. Digital Image Computing: Techniques and Applications (DICTA), 2010 International Conference on, Sydney, Australia, 1-3 December 2010.
8. Spagnolo P, Orazio TD, Leo M, Distante A. “Moving object segmentation by background subtraction and temporal analysis”. Image and Vision Computing, 24(5), 411-423, 2006.

9. Yu X, Chen X, Gao M. “Motion detection in dynamic scenes based on fuzzy c-means clustering”. International Conference on Communication Systems and Network Technologies, Rajkot, India, 11-13 May 2012.
10. Kim J, Ye G, Kim D. “Moving object detection under free-moving camera”. IEEE 17th International Conference on Image Processing, Hong Kong, China, 26-29 September 2010.
11. Zhang Y, Kiselewich SJ, Bauson WA, Hammoud R. “Robust moving object detection at distance in the visible spectrum and beyond using a moving camera”. Conference on Computer Vision and Pattern Recognition, New York, USA, 17-22 June 2006.
12. Weng M, Huang G, Da X. “A new ınterframe difference algorithm for moving target detection”. Image and Signal Processing (CISP) 3rd International Congress, Yantai, China, 16-18 October 2010.
13. Roth S, Black MJ. “On the spatial statistics of optical flow”. International Journal of Computer Vision, 74(1), 33-50, 2007.
14. H Liu, Hong TH, Herman M, Chellappa R. Accuracy vs. Efficiency Trade-Offs in Optical Flow Algorithms. Editors: Buxton B, Cipolla R. Lecture Notes in Computer Science, 271-286, Cambridge, UK, Springer Berlin Heidelberg, 1996.
15. Ren Y, Chua CS, Ho YK. “Statistical background modeling for non-stationary camera”. Pattern Recognition Letters, 24(1-3), 183-196, 2003.
16. Sappa AD, Dornaika F, Ger´onimo D, L´opez A. “Registration-Based moving object detection from a moving camera”. Workshop on Perception Planning and Navigation for Intelligent Vehicles, Nice, France, 26 September 2008.
17. Cheraghi SA, Sheikh UU. “Moving object detection using image registration for a moving camera platform”. Control System Computing and Engineering (ICCSCE) IEEE International Conference on, Penang, Malaysia, 23-25 November 2012.
18. Bay H, Tuytelaars T, Van Gool L. Surf: Speeded up Robust Features. Editors: Leonardis A, Bischof H, Pinz A. Computer Vision-ECCV 2006, 404-417, Graz, Austria: Springer Berlin Heidelberg, 2006.
19. Harris C, Stephens M. “A combined corner and edge detector”. Alvey Vision Conference, Manchester, UK, 31 August - 2 September 1988.
20. Rosten E, Drummond T. Machine Learning for High-Speed Corner Detection. Editors: Leonardis A, Bischof H, Pinz A. Computer Vision-ECCV 2006, 430-443, Graz, Austria, Springer Berlin Heidelberg, 2006.
21. Wilson HR, Giese SC. “Threshold visibility of frequency gradient patterns”. Vision Research, 17(10), 1177-1190 1977.
22. Erhan C, Tazehkandi AA, Yalcin H, Bayram I. “Traffic sign detection and recognition fusing feature descriptors”. Signal Processing and Communications Applications Conference (SIU), Haspolat, TRNC, 24-26 April 2013.
23. Martin J, Crowley JL. Comparison of Correlation Techniques. Editors: Rembold U, Dillmann R, Hertzberger LO, Kanade T. Intelligent Autonomous Systems IAS 4, 86-93, Karlsruhe, Germany, IOS Press, 1995.
24. Otsu N. “A Threshold selection method from gray-level histograms”. Automatica, 11(285-296), 23-27, 1975.
25. Gonzalez RC, Woods RE, Eddins SL. Digital Image Processing Using MATLAB. 2nd ed. Gatesmark, USA, 2009.
26. Stauffer C, Grimson WEL. “Adaptive background mixture models for real-time tracking”. IEEE Computer Society Conference on ComputerVision and Pattern Recognition, Fort Collins, CO, USA, 23-25 June 1999.