针对出租车驾驶员蓄须行为自动检测技术展开研究，提出了一套基于视频图像的、完整的自动检测方法。首先，根据复杂交通车流的特性，提出一种用车窗区域代替完整车辆的出租车检测方法，同时基于检测到的车窗区域实现对驾驶室的精准定位。其次，利用基于亮度通道的多尺度视网膜（Multi-scale Retinex, MSR） 增强的图像预处理算法实现对图像的光线平衡及细节恢复。最后，通过libfacedetection 算法的精准下巴区域提取、结合非肤色像素点提取以及灰度阈值法的正反验证来提升算法的鲁棒性。利用交通监控视频对整套方法进行测试，研究结果表明，该方法能够对出租车驾驶员蓄须行为进行识别，进而提高执法效能。
Abstract
The automatic detection technology of taxi driver′s beard was studied, and a complete detection method of taxi driver′s beard was put forward based on video and image. First of all, according to the properties of the complex traffic flow, the method of using window area detection to replace complete vehicle was proposed, which can pinpoint the vehicle cab. Secondly, the image preprocessing algorithm of Multi-scale Retinex (MSR) enhancement based on value channel was used to achieve light balance and detail restoration of the image. Finally, the robustness of the algorithm was improved through accurate chin area extraction by libfacedetection algorithm, and combined with non-skin color pixel extraction and positive and negative verification of gray threshold method. Traffic surveillance video was used to test the method. The results show that the method proposed can be used to recognize the taxi driver′s beard, so as to improve the efficiency of law enforcement.
关键词
蓄须行为 /
车窗区域 /
MSR /
肤色 /
灰度阈值
Key words
beard /
car window area /
multi-scale retinex /
skin colour /
gray threshold
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] This Newspaper′ s Commentator. Correctly Understanding the Relationship Between the Transportation Power and the Socialist Modernization Power[N]. China Water Transport, 2017-11-13(001).
[2] HUANG Q. Current Situation and Improvement of Non-Site Enforcement[J]. Journal of Shanghai Public Security College, 2007, 17(3): 79-81.
[3] CAI X, GAN K, YANG C, et al. The Beard Detection Algorithm Based on Feature Point Location and Skin Color Segmentation[J].VideoEngineering,2016,40(3):116-121.
[4] GU Y, QIU W. Beard and Hat Detection Based on CNN[J]. Information Technology, 2017(9): 121-124.
[5] WANG J G, YAU W Y. Real- Time Beard Detection by Combining Image Decolorization and Texture Detection with Applications to Facial Gender Recognition[J]. Machine Vision & Applications , 2014 , 25(4) : 1089-1099.
[6] DHAHRI R, BELAID S. A Robust Detection and Removal of Beard from Three Dimensional Human Face[C]// 2013 5th International Conference on Modeling, Simulation and Applied Optimization(ICMSAO). Hammamet, Tunisia: IEEE, 2013: 1-5.
[7] LE T H N, LUU K, SESHADRI K, et al. Beard and Mustache Segmentation Using Sparse Classifiers on Self-Quotient Images[C]// 2012 19th IEEE International Conference on Image Processing. Coronado, U S A: IEEE, 2012: 165-168.
[8] LIENHART R, MAYDT J. An Extended Set of Haar-Like Features for Rapid Object Detection[C]// International Conference on Image Processing. New York: IEEE, 2002: 900-903.
[9] FREUND Y, SCHAPIRE R E. A Desicion-Theoretic Generalization of on-Line Learning and an Application to Boosting[J]. Journal of Computer & System Sciences, 1997, 55 (1): 119-139.
[10] LI Z, QIU H. Research on Fast Image Matching Based on Correlation Coefficient[J]. Journal of Beijing Institute of Technology, 2007, 27(11): 998-1000.
[11] JOBSON D J, RAHMAN Z, WOODELL G A. A Multiscale Retinex for Bridging the Gap Between Color Images and the Human Observation of Scenes[J]. IEEE Trans Image Process, 2002, 6 (7): 965-976.
[12] QIN X, WANG H, DU Y. Retinex Structured Light Image Enhancement Algorithm in HSV Color Space[J]. Journal of Computer-Aided Design & Computer Graphics, 2013, 25(4): 488-493.
[13] VIOLA P, JONES M. Robust Real-Time Face Detection[J]. International Journal of Computer Vision, 2004, 57(2): 137-154.
[14] LI Y, LAI J H, PONGCHI Y. Multi Template ASM and Its Application in Face Feature Points Detection[J]. Journal of Computer Research and Development, 2007, 44(1): 133-140.
[15] LIU C, ZHANG L. An Improved Feature Point Location Method for Face[J]. Journal of Fudan University (Natural Science), 2006, 45(4): 457-463.
[16] 曹建秋，王华清，蓝章礼. 基于改进Ycrcb颜色空间的肤色分割[J]. 重庆交通大学学报（自然科学报），2010，29（3）：488-492.
[17] 方晶晶，李振波，姜宇. 人体肤色区域的自适应模型分割方法[J]. 计算机辅助设计与图形学学报，2013，25（2）：229-234.
[18] 郭耸，顾国昌，蔡则苏，等. 肤色相似度和动态阈值相结合的肤色分割技术[J]. 计算机工程与应用，2010，46（18）：1-3，12.
基金
国家自然科学基金项目（61672064）；北京市自然科学基金项目（KZ201610005007）；北京市自然科学基金面上项目（4172001）；博士后科学基金项目（2016T90022, 2015M580029, 2015ZZ-23, 2016ZZ-01-05）；北京市交通委员会科技项目（2017058）
京公网安备 11010102003417号
京ICP备17006742号-3
