1.School of Computer and Communication Engineering,University of Science and Technology Beijing,Beijing 100083 2.Interactive Digital Media Technology Research Center,Institute of Automation,Chinese Academy of Sciences,Beijing 100190 3.Meteorological Observation Center,China Meteorological Administration,Beijing 100081
Abstract High hardware cost,complex operation and narrow application are main problems of existing measuring methods of atmospheric visibility. In this paper,machine learning is introduced into the study of visibility measurement and a method of daytime visibility measurement is proposed based on image understanding. Firstly,image features and vectors based on pixel contrast are designed and extracted grounded on the segmentation of the regions of interest in measured scene images. Then,the relational model between image features and visibility is constructed by training support vector regression. Finally,visibility of images to be measured is computed according to the model. Experimental results show that the proposed method has both high visibility measuring precision and good flexibility. Moreover, it reduces the limitations of existing visibility measurement methods.
[1]Zeng Shu′er,Wang Gaili. Observation and Instrument of Visibility. Journal of Applied Meteorology,1999,10(2): 207-212 (in Chinese) (曾书儿,王改利.能见度的观测及其仪器.应用气象学报,1999,10(2): 207-212) [2]Steffens C. Measurement of Visibility by Photographic Photometry. Industrial Engineering Chemistry,1949,41(11): 2396-2399 [3]Kwon T M. Atmospheric Visibility Measurements Using Video Cameras: Relative Visibility[EB/OL]. [2012-07-20]. http://www.conservancy.umn.edu/bitstream/1027/1/CTS-04-03.pdf [4]Hautière N,Aubert D,Dumont ,et al. Experimental Validation of Dedicated Methods to In-Vehicle Estimation of Atmospheric Visibility Distance. IEEE Trans on Instrumentation and Measurement,2008,57(10): 2218-2225 [5]Chen Zhaozheng,Chen Qimei. Video Contrast Visibility Detection Algorithm and Its Implementation Based on Camera Self-Calibration. Journal of Electronic Information Technology,2010,32(12): 2907-2912 (in Chinese) (陈钊正,陈启美.基于摄像机自标定的视频对比度能见度检测算法与实现.电子与信息学报,2010,32(12): 2907-2912) [6]Kwon T M. An Automatic Visibility Measurement System Based on Video Cameras[EB/OL]. [2012-07-20]. http//www.tc.umn.edu/~kwonw059/Research/images/VisibilityReportPh-1-V1.pdf [7]Hautière N,Tarel J P,Lavenant J,et al. Automatic Fog Detection and Estimation of Visibility Distance through Use of an Onboard Camera. Machine Vision and Applications Journal,2006,17(1): 8-20 [8]Lü Weitao,Tao Shanchang,Liu Yifeng,et al. Measuring Meteorological Visibility Based on Digital Photography-Dual Differential Luminance Method and Experimental Study. Chinese Journal of Atmospheric Sciences,2004,28(4): 559-570 (in Chinese) (吕伟涛,陶善昌,刘亦风,等.基于数字摄像技术测量气象能见度——双亮度差方法和试验研究.大气科学,2004,28(4): 559-570) [9]Dumont E,Cavallo V. Extended Photometric Model of Fog Effects on Road Vision. Transportation Research Record,2004,1862: 77-81 [10]Malm W,Pitchford M,Pitchford A. Site Specific Factors Influencing the Visual Range Calculated from Teleradiometer Measurements. Atmospheric Environment,1982,16(10): 2323-2333 [11]Molenar J V. Uncertainty Analysis of Calculated Extinction from Apparent Contrast Measurements from Images of Natural Targets[EB/OL]. [2012-08-02]. ttp://vista.cira.colostate.edu/improve/publications/graylit/BextFromPhoto/Bext_Uncertainty_Image.pdf [12]Kwon T M. Video Camera-Based Visibility Measurement System: USA,US6853453. 2005-08-02 [13]Jourlin M,Pinoli J C. Logarithmic Image Processing: The Mathematical and Physical Framework for the Representation and Processing of Transmitted Images // Hawkes P W,ed. Advances in Imaging and Electron Physics. New York,USA: Academic Press,2001,115: 129-196 [14]Lowe D G. Distinctive Image Features from Scale-Invariant Keypoints. International Journal of Computer Vision,2004,60(2): 91-110 [15]Wang Dingcheng,Fang Tingjian,Tang Yi,et al. Review of Support Vector Machines Regression Theory and Control. Pattern Recognition and Artificial Intelligence,2003,16(2): 192-197 (in Chinese) (王定成,方廷健,唐 毅,等.支持向量机回归理论与控制的综述.模式识别与人工智能,2003,16(2): 192-197) [16]Tang Yaohua,Guo Weimin,Gao Jinghuai. SVM Parameter Selection Algorithm Based on Maximum Kernel Similarity Diversity. Pattern Recognition and Artificial Intelligence,2010,23(2): 210-215 (in Chinese) (唐耀华,郭为民,高静怀.基于核相似性差异最大化的支持向量机参数选择算法.模式识别与人工智能,2010,23(2): 210-215) [17]Li Hao,Sun Xuejin. Theoretical Analysis on Measurement Error of Forward Scattering Visibility Meter. Infrared and Laser Engineering,2009,38(6): 1094-1098 (in Chinese) (李 浩,孙学金.前向散射能见度仪测量误差的理论分析.红外与激光工程,2009,38(6): 1094-1098)
2013, Vol. 26 
