Shadow removal on digital satellite images using wavelet transforms
DOI:
https://doi.org/10.34185/1562-9945-5-130-2020-11Keywords:
космічні знімки, вейвлет-перетворення, сегментація, текстурна характеристика, виявлення тіні, видалення тіні, кольорова метрика HSVAbstract
Shadow detection and removal in real scene images is always a challenging but yet intriguing problem. Shadows cause hindrance to correct feature extraction of image features like buildings ,towers etc. in urban areas it may also cause false color tone and shape distortion of objects, which degrades the quality of images. Hence, it is important to segment shadow regions and restore their information for image interpretation. Shadows are generated by a local and relative absence of light. Shadows are, first of all, a local decrease in the amount of light that reaches a surface. Secondly, they are a local change in the amount of light rejected by a surface toward the observer. Most shadow detection and segmentation methods are based on image analysis. This paper presents a novel algoritm for automatic shadow detection and removing shadows using HSV color model, contour segmentation and wavelet transform based on a threshold determined by wavelet coefficients in complex urban color remote sensing images for solving problems caused by shadows. In the proposed algoritm shadows are detected using S and V components which is invariant to shadow i.e., it conveys the spectral and color characteristics of image features, regardless of variations in scene illumination condition and means of wavelet coefficients. The multi-resolution property of the wavelet transform leads into four different bands without the loss of spatial information. Once the shadows are detected they are classified and a non shadow area around each shadow termed as buffer area is estimated using contour segmentation. Experiments show that the new algoritm can accurately detect shadows from urban high-resolution remote sensing images and can effectively restore shadows with a rate of over 85%. The proposed algoritm can be used for further object recognition and thematic processing of scanner images.
References
Schowengerdt R. (2007), Remote sensing: models and methods for image processing, New York: Academic Press. P.560.
Pansharpening technology of high resolution multispectral and panchromatic satellite images / V. V. Hnatushenko, Vik. V. Hnatushenko, O. O. Kavats, V. Yu. Shevchenko // Scientific Bulletin of National Mining University. – 2015. – № 4. – P. 91–98
Hnatushenko V.V., Hnatushenko Vik.V., Mozgovyi D.K., Vasyliev, V.V. “Satellite technology of the forest fires effects monitoring”. Scientific Bulletin of National Mining University, 2016. Issue 1 (151), pp. 70-76.
Marchant J.A., Onyango C.M. Shadow invariant classification for scenes illuminated by daylight // Journal Optical Society of America A. – 2000. – Vol. 17. – № 12. – P.1952-1961.
Renno, J.-P.R. Evaluation of shadow classification techniques for object detection and tracking / J.-P.R. Renno, J. Orwell, G.A. Jones // International Conference on Image Processing. – Singapore, 2004. – Vol. 1. – P.143–146.
Chen, B. Shadow Detection Based on RGB Color Model // B. Chen, D. Chen / Intelligent Computing in Signal Processing and Pattern Recognition. Lecture Notes in Control and Information Sciences. – 2006. – Vol. 345. – P. 1068–1074.
Finlayson G. D. On the Removal of Shadows from Images / G. D. Finlayson, S. D. Hordley, C. Lu, M. S. Drew // IEEE Transactions on Pattern Analysis and Machine Intelligence. – 2006. – Vol. 28, № 1. – P. 59–68.
Giles, P. Remote sensing and cast shadows in mountainous terrain / P. Giles // Photogrammetric Engineering & Remote Sensing. – 2001. – Vol. 67 (7). – P. 833–839.
Rau, J.-Y. True orthophoto generation of built-up areas using multi-view images / J.-Y. Rau, N.-Y. Chen, L.-C. Chen // Photogrammetric Engineering & Remote Sensing. – 2002. – Vol. 68 (6). – P. 581–588.
R.Cucchiara, C.Grana, M. Piccardi, A. Prati, Detecting moving objects, ghosts, and shadows in video streams//IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 25, no. 10, 2003, pp. 1337-1342.
Dare, P.M. Shadow analysis in high-resolution satellite imagery of urban areas / P.M. Dare // Photogrammetric Engineering and Remote Sensing. – 2005. – Vol. 71. – P. 169–177.
Kahtan V.Yu. Satellite Imagery Features for the Image Similarity Estimation [Electronic recourse] / Y. I. Shedlovska, V. V. Hnatushenko, V. Yu. Kashtan // International Young Scientists Forum on Applied Physics 2017, October, 16 – 20, Lviv, Ukraine : Proceedings. –Lviv, 2017. – p. 359-362.
Kahtan V.Yu. Processing technology of multispectral remote sensing images [Electronic recourse] / V.Yu. Kahtan, V. V. Hnatushenko, Y. I. Shedlovska // International Young Scientists Forum on Applied Physics. 2017. October, 16 – 20, Lviv, Ukraine: Proceedings. Lviv, 2017. – p. 355-358.
Kashtan V.Iu. Konturna sehmentatsiia tsyfrovykh suputnykovykh znimkiv z vydilenniam osoblyvykh tochok na osnovi veivlet-peretvorennia / V.Iu. Kashtan, V.V. Hnatushenko // Systemni tekhnolohii. Rehionalnyi mizhvuzivskyi zbirnyk naukovykh prats. – Vypusk 1 (120): zb. nauk. prats. – Dnipro, 2019. – s. 3-11.
Eli Arbel and Hagit Hel-Or. "Shadow Removal Using Іntensity Surfaces and Texture Anchor Points". IEEE. transactions on pattern analysis and machine intelligence, vol. 33, no. 6, June 2011.
