HOU Tao， WU Hai-ping
（School of Automation and Electrical Engineering， Lanzhou Jiaotong University， Lanzhou 730070， China）
Abstract： In case of complex textures, existing static shadow detection and removal algorithms are prone to false detection of the pixels. To solve this problem, a static shadow detection and removal algorithm based on support vector machine (SVM) and region sub-block matching is proposed. Firstly, the original image is segmented into several superpixels, and these superpixels are clustered using mean-shift clustering algorithm in the superpixel sets. Secondly, these features such as color, texture, brightness, intensity and similarity of each area are extracted. These features are used as input of SVM to obtain shadow binary images through training in non-operational state. Thirdly, soft matting is used to smooth the boundary of shadow binary graph. Finally, after finding the best matching sub-block for shadow sub-block in the illumination region based on regional covariance feature and spatial distance, the shadow weighted average factor is introduced to partially correct the sub-block, and the light recovery operator is used to partially light the sub-block. The experimental results show the number of false detection of the pixels is reduced. In addition, it can remove shadows effectively for the image with rich textures and uneven shadows and make a natural transition at the boundary between shadow and light.
Key words： shadow detection； shadow removal； feature extraction； support vector machine(SVM)； block matching； light recovery operator
CLD number： TP391.41 doi： 10.3969/j.issn.1674-8042.2020.04.005
References
［1］Khan S H, Bennamoun M, Sohel F, et al. Automatic shadow detection and removal from a single image. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2016, 38(3): 431-446.
［2］Boyadzhiev I, Paris S, Bala K. User-assisted image compositing for photographic lighting. ACM Transactions on Graphics, 2013, 32(4): 1-11.
［3］Chen L, Zhang R G, Hu J. Improved Gaussian mixture model and shadow elimination method. Journal of Computer Applications, 2013, 33(5): 1394-1397.
［4］Boroujeni H S, Charkari N M. Robust moving shadow detection with hierarchical mixture of MLP experts. Signal Image and Video Processing, 2014, 8(7): 1291-1305.
［5］Ebner M, Yuan X, Wang Z. Single-image shadow detection and removal using local colour constancy computation. Image Processing Iet, 2015, 9(2): 118-126.
［6］Khan S H, Bennamoun M, Sohel F. Automatic shadow detection and removal from a single image. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2016, 38(3): 431-446.
［7］Krishnan A, Jayadevan P, Panicker J V. Shadow removal from single image using color invariant method. In: Proceedings of International Conference on Communication & Signal Processing, Chennai, India, 2017, 6(8): 583-587.
［8］Zhang W, Wu Q M J. Moving shadow detection based on normalized eigenvalue of Wishart matrix. In: Proceedings of International Conference on Mechatronics & Automation, Xi’an, China, 2010, 4(7): 645-650.
［9］Kumar, Sanjeev, Anupreet K. Algorithm for shadow detection in real colour images. International Journal on Computer Science & Engineering, 2010, 2(7): 2444-2446.
［10］Wang J, Wang Y H, Jiang M, et al. Moving cast shadow detection using online sub-scene shadow modeling and object inner-edges analysis. Journal of Visual Communication & Image Representation, 2014, 25(5): 978-993.
［11］Achanta R, Shaji A, Smith K. SLIC superpixels compared to state-of-the-art superpixel methods. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(11): 2274-2282.
［12］Comaniciu D, Meer P. Mean shift: a robust approach toward feature space analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(5): 603-619.
［13］Panchal M H, Gamit N C. A comprehensive survey on shadow detection techniques. In: Proceedings of International Conference on Wireless Communications, Chennai, India, 2016: 23-25.
［14］Salvador E, Cavallaro A, Ebrahimi T. Cast shadow segmentation using invariant color features. Computer Vision and Image Understanding, 2004, 95(2): 238-259.
［15］Guo R, Dai Q, Hoiem D. Paired regions for shadow detection and removal. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(12): 2956-2967.
［16］Zhang L, Zhang Q, Xiao C. Shadow remover: image shadow removal based on illumination recovering optimization. IEEE Transactions on Image Processing, 2015, 24(11): 4623-4636.
［17］Shor Y, Lischinski D. The shadow meets the mask: pyramid-based shadow removal. Computer Graphics Forum, 2008, 27(2): 577-586.
［18］Liu H, Ma L Z, Cai X, et al. A closed-form solution to video matting of natural snow. Information Processing Letters, 2009, 109(18): 1097-1104.
［19］Liu S T, Jiang K H, Liu Z X. Ship target detection of aerial reconnaissance image based on region covariance and objectness. Systems Engineering and Electronics, 2019, 41(5): 972-980.
［20］Hong X, Chang H, Shan S, et al. A small second order statistical region descriptor. In: Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Miami, Florida, USA, 2009: 20-25.
静态阴影的自动检测与去除
侯 涛， 伍海萍
（兰州交通大学 自动化与电气工程学院， 甘肃 兰州 730070）
摘 要: 针对现有的静态阴影检测与去除算法容易出现像素误检及在纹理情况复杂时未能很好的去除阴影等问题， 提出了基于支持向量机与区域子块匹配的静态阴影自动检测与去除算法。 首先， 对初始图像进行超像素分割处理， 用mean-shift聚类方法在超像素特征集合内进行聚类得到n个聚类区域； 其次， 在得到的聚类区域提取像素多个特征对支持向量机进行离线训练获得阴影二值图像， 再利用soft matting对阴影二值图的边界进行平滑； 之后， 在光照区域利用区域协方差特征和空间距离这两个条件找到阴影子块的最佳匹配子块； 最后， 引入阴影加权平均因子对子块进行部分修正， 并利用光照恢复算子对子块进行局部点亮。 实验结果表明， 该算法可以很好的降低阴影像素的误检率， 对于纹理类型丰富、 阴影不均匀的图像去除阴影效果明显， 可使阴影与光照的边界效应大大减弱， 而且该算法在实际铁路场景应用中也取得了很好的效果。
关键词: 阴影检测； 阴影去除； 特征提取； 支持向量机； 子块匹配； 光照恢复算子
引用格式： HOU Tao， WU Hai-ping. Automatic detection and removal of static shadows. Journal of Measurement Science and Instrumentation, 2020, 11（4）： 343-350. ［doi： 10.3969／j.issn.1674-8042.2020.04.005］
[full text view]
