{"title":"Object Detection in Digital Images under Non-Standardized Conditions Using Illumination and Shadow Filtering","authors":"Waqqas-ur-Rehman Butt, Martin Servin, Marion Pause","volume":132,"journal":"International Journal of Computer and Information Engineering","pagesStart":1285,"pagesEnd":1294,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/10008298","abstract":"In recent years, object detection has gained much
\r\nattention and very encouraging research area in the field of computer
\r\nvision. The robust object boundaries detection in an image is
\r\ndemanded in numerous applications of human computer interaction
\r\nand automated surveillance systems. Many methods and approaches
\r\nhave been developed for automatic object detection in various fields,
\r\nsuch as automotive, quality control management and environmental
\r\nservices. Inappropriately, to the best of our knowledge, object
\r\ndetection under illumination with shadow consideration has not
\r\nbeen well solved yet. Furthermore, this problem is also one of
\r\nthe major hurdles to keeping an object detection method from the
\r\npractical applications. This paper presents an approach to automatic
\r\nobject detection in images under non-standardized environmental
\r\nconditions. A key challenge is how to detect the object, particularly
\r\nunder uneven illumination conditions. Image capturing conditions
\r\nthe algorithms need to consider a variety of possible environmental
\r\nfactors as the colour information, lightening and shadows varies
\r\nfrom image to image. Existing methods mostly failed to produce the
\r\nappropriate result due to variation in colour information, lightening
\r\neffects, threshold specifications, histogram dependencies and colour
\r\nranges. To overcome these limitations we propose an object detection
\r\nalgorithm, with pre-processing methods, to reduce the interference
\r\ncaused by shadow and illumination effects without fixed parameters.
\r\nWe use the Y CrCb colour model without any specific colour
\r\nranges and predefined threshold values. The segmented object regions
\r\nare further classified using morphological operations (Erosion and
\r\nDilation) and contours. Proposed approach applied on a large image
\r\ndata set acquired under various environmental conditions for wood
\r\nstack detection. Experiments show the promising result of the
\r\nproposed approach in comparison with existing methods.","references":"[1] Dawson-Howe, K. (2014), \u201cA Practical Introduction to Computer Vision\r\nwith OpenCV,\u201d Hoboken, NJ: Wiley. 235.\r\n[2] W. Y. D. K. H. Yoon (2007), \u201cFast Group Verification System for\r\nIntelligent Robot Service,\u201d IEEE Transactions on Consumer Electronics,\r\nvol. 53, pp.1731-1735.\r\n[3] W. U. R. Butt, L. Lombardi (2013), \u201cComparisons of Visual\r\nFeatures Extraction Towards Automatic Lip Reading,\u201d 5th International\r\nConference on Education and New Learning Technologies, Barcelona,\r\nSpain, vol. 5, pp.2188-2196.\r\n[4] Luca Lombardi, Waqqas ur Rehman Butt, Marco Grecuccio (2014),\r\n\u201cLip Tracking Towards an Automatic Lip Reading Approach,\u201d Journal\r\nof Multimedia Processing and Technologies, vol. 5, pp.1-11. ISSN:\r\n0976-4127. [5] W. U. R. Butt, L. Lombardi (2013), \u201cA Survey of Automatic Lip Reading\r\nApproaches,\u201d 8th ICDIM 2013 (The 8th International Conference on\r\nDigital Information Management) in Islamabad, Pakistan, pp.299-302.\r\n[6] Nusirwan A. Rahman, Kit C. Wei, John See. (2006), \u201cRGB\u2212H\u2212 CbCr\r\nSkin Colour Model for Human Face Detection,\u201d In Proceedings of The\r\nMMU International Symposium on Information and Communications\r\nTechnologies.\r\n[7] V. Vezhnevets, V. Sazonov, and A. Andreeva. (2003), \u201cA Surveyon\r\nPixel-based Skin Color Detection Techniques,\u201d 8th ICDIM 2013 (In\r\nProceedings of the GraphiCon, Moscow, Russia, pp.85-92.\r\n[8] Moscariello, Antonio, et al. (2011), \u201cCoronary CT angiography: image\r\nquality, diagnostic accuracy, and potential for radiation dose reduction\r\nusing a novel iterative image reconstruction technique comparison with\r\ntraditional filtered back projection,\u201d (European radiology, pp.2130-2138.\r\n[9] Guo, Ruiqi, Qieyun Dai, and Derek Hoiem (2011), \u201cSingle-image shadow\r\ndetection and removal using paired regions,\u201d (Computer Vision and\r\nPattern Recognition - CVPR, pp.2033-2040.\r\n[10] Ferguson, P. D., Arslan, T., Erdogan, A. T., Parmley, A. (2008),\r\n\u201cEvaluation of contrast limited adaptive histogram equalization (clahe)\r\nenhancement on a FPGA,\u201d (SoCC), pp.119-122.\r\n[11] J. Majumdar, S. Kumar K. L. (2014), \u201cModified CLAHE: An adaptive\r\nalgorithm for contrast enhancement of aerial, medical and underwater\r\nimages, \u201d International Journal of Computer Engineering and Technology\r\n(IJCET), vol. 5, pp.32-47.\r\n[12] Intel Open Source Computer Vision Library, (OPENCV)\r\n\u201chttp:\/\/sourceforge.net\/projects\/opencvlibrary\/,\r\n[13] de Dios J. J., Garcia, N.(2004), \u201cFast face segmentation in component\r\ncolor space, \u201d Int. Conf. on Image Processing, (ICIP), vol. 1, pp.191-194.\r\n[14] S. Gundimada, L. Tao, and V. Asari(2004), \u201cFace Detection Technique\r\nbased on Intensity and SkinColor Distribution, \u201d Int. Conf. on Image\r\nProcessing, (ICIP), pp.1413-1416.\r\n[15] R.-L. Hsu, M. Abdel-Mottaleb, and A. K. Jain (2002), \u201cFace Detection\r\nin Color Images, \u201d IEEE Trans. PAMI, 24(5), pp.696-706.\r\n[16] P. Peer, J. Kovac, F. Solina (2003), \u201cHuman Skin Colour Clustering for\r\nFace Detection, \u201d EUROCON1993, Ljubljana, Slovenia, pp.144-148.\r\n[17] Bassiou, Nikoletta, and Constantine Kotropoulos (2007), \u201cColor image\r\nhistogram equalization by absolute discounting back-off. \u201d Computer\r\nVision and Image Understanding 107.1 , pp.108-122.\r\n[18] Simonoff, Jeffrey S. (1998), \u201cSmoothing Methods in Statistics. \u201d 2nd\r\nedition. Springer ISBN 978-0387947167 .\r\n[19] Yiu-ming Cheung, Xin Liu, Xinge You (2012), \u201cA local region\r\nbased approach to lip tracking \u201d Pattern Recognition , vol.45 (12),\r\npp.3336-3347.\r\n[20] Guo, Ruiqi, Qieyun Dai, and Derek Hoiem. (2011), \u201cSingle-image\r\nshadow detection and removal using paired regions \u201d PComputer Vision\r\nand Pattern Recognition (CVPR), IEEE Conference.\r\n[21] S. Wang, W. Lau, S. Leung (2004), \u201cAutomatic lip contour extraction\r\nfrom colour images \u201d Pattern Recognition, vol.37(12), pp.2375-2387.\r\n[22] D. Xu, J. Liu, X. Li, Z. Liu, X. Tang (2005), \u201cInsignificant shadow\r\ndetection for video segmentation,\u201d IEEE Transactions on Circuits and\r\nSystems for Video Technology, vol.15, pp.1058-1064.","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 132, 2017"}