基于SVM和决策树的自然图像低能见度天气现象识别
Low Visibility Weather Recognition via SVM and Decision Tree in Single Image

作者: 徐冠雷 * , 王孝通 , 邵利民 , 周立佳 , 徐晓刚 :海军大连舰艇学院军事海洋系,辽宁 大连;

关键词: 低能见度天气图像支持向量机决策树训练Low Visibility Weather Image Support Vector Machine (SVM) Decision Tree Training

摘要:
一个地区的能见度不但反映了该地区大气环境的质量,并且与人们的生活有着密切相关的联系。通常,低能见度天气也严重地影响了人们的经济生产,因此其观测具有重要意义。大气环境能见度较低的原因与气象条件有着密切的关联,低能见度的天气现象主要有雨、雪、雾霾、沙尘等。本文提出了一种基于室外单幅自然图像的低能见度天气现象识别算法,该算法通过低能见度天气现象对图像光学信息的影响,提取图像的对比度、饱和度、亮度等特征参数信息进行训练和分类,在训练过程中根据各类别特征之间的距离建立分类决策树,并为决策树构建支持向量机(SVM)分类器,对低能见度天气进行自动分类识别。通过对互联网上的大量低能见度天气光学图像的训练和测试,算法对低能见度的天气现象的平均识别率可达70%。该算法可以为分布式识别提供技术支持,然后采用分布式识别投票,最终可以把识别正确率提高到95%以上。

Abstract: The visibility in a region not only reflects the quality of the atmospheric environment, but also has close relationship with people’s life. In general, low visibility weather affects people’s economic development, so the real-time observation of low visibility is of much signification. The reason of low visibility is closely associated with meteorological conditions. The low visibility weather phenomena mainly contain rain, snow, fog, etc. This paper proposes a recognition method which is based on low visibility weather phenomenon by means of the influence of low visibility weather phenomenon on the image information such as the image contrast, saturation and brightness that can be employed for training and classification. We establish a classification decision tree according to the distance between the different categories in the process of training and building support vector machine (SVM) classifier for the decision tree. It can classify the low visibility weather image automatically and intelligently. Through testing a huge amount of images downloaded from the internet, the experimental results show that weather image mean recognition rate is over 70%. After adopting the voting scheme via distributed recognition, the final low visibility weather recognition rate is more than 95%.

文章引用: 徐冠雷 , 王孝通 , 邵利民 , 周立佳 , 徐晓刚 (2016) 基于SVM和决策树的自然图像低能见度天气现象识别。 图像与信号处理, 5, 155-165. doi: 10.12677/JISP.2016.54018

参考文献

[1] 谢兴生, 陶善昌, 周秀骥. 数字摄像法测量气象能见度[J]. 科学通报, 1999(44): 97-100.

[2] 孙学金, 王晓蕾, 李洁, 等. 大气探测学[M]. 气象出版社, 北京: 2009.

[3] 章毓晋. 图像处理[M]. 北京: 清华大学出版社, 2006: 165-172.

[4] 吴立德. 计算机视觉[M]. 上海: 复旦大学出版社, 1993.

[5] 宋小宁, 赵英时. MODIS图像的云检测及分析[J]. 中国图象图形学报, 2003, 8(9): 1079-1083.

[6] 霍娟, 吕达仁. 全天空数字相机观测云量的初步研究[J]. 南京气象学院学报, 2002(2): 242-246.

[7] Xiquan, D., Patrick, M., Gerald, G.M., et al. (2002) Comparison of Status Cloud Properties Deduced from Surface, GOES, and Aircraft Data during the March 2000 ARM Cloud IOP. Journal of the Atmospheric Science, 59, 3265- 3284. http://dx.doi.org/10.1175/1520-0469(2002)059<3265:COSCPD>2.0.CO;2

[8] Goodman, A.H. (1988) Henderson-Sellers A. Cloud Detection and Analysis: A Review of Recent Progress. Atmospheric Research, 21, 203-228. http://dx.doi.org/10.1016/0169-8095(88)90027-0

[9] 师春香, 吴蓉璋, 项续康. 多阈值和神经网络卫星云图自动分割实验[J]. 应用气象学报, 2001, 12(1): 70-78.

[10] Yamashita, M., Yoshimura, M. and Nakashizuka, T. (2004) Cloud Cover Estimation Using Multitemporal Hemisphere Imageries. Proceedings of 7th Congress of the International Society for Photogrammertry and Remote Sensing (ISPRS04), Istanbul, 818-821.

[11] Souza-Echer, M.P., Pereir, A.E.B., Bins, L.S., et al. (2006) A Simple Method for the Assessment of the Cloud Cover State in High Latitude Regions by a Ground-Based Digital Camera. Journal of Atmospheric and Oceanic Technology, 23, 437-447. http://dx.doi.org/10.1175/JTECH1833.1

[12] 胡树贞, 马舒庆, 陶法, 等. 基于红外实时阈值的全天空云量观测[J]. 应用气象学报, 2013, 24(2): 179-188.

[13] Yang, Y.-X. and Hu, X.-H. (2009) Miriam PWD20 Visibility and Comparative and Analysis of Visibility and Visual. Journal of Lanzhou University: Natural Science, 45, 61-63.

[14] Steffens, C. (1949) Measurement of Visibility by Photographic Photometry. Industrial Engineering Chemistry, 41, 2396-2399. http://dx.doi.org/10.1021/ie50479a015

[15] Legal, T., Legal, L. and Lehn, W. (1994) Measureing Visibility Using Digital Remote Videocameras. American Meteorological Society, 87-89.

[16] 安明伟, 陈启美, 郭宗良. 基于路况视频的气象能见度检测方法与系统设[J]. 仪器仪表学报, 2010, 31(5): 1148- 1153.

[17] 李勃, 董蓉, 陈启美. 无需人工标记的视频对比度道路能见度检测[J]. 计算机辅助设计与图形学报, 2009, 21(11): 1575-1582.

[18] 谢兴生, 陶善昌, 周秀骥. 数字摄像法测量气象能见度[J]. 科学通报, 1999, 44(1): 97-100.

[19] Lu, C., Lin, D., Jia, J. and Tang, C.-K. (2014) Two-Class Weather Classification. 2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Columbus, 23-28 June 2014, 3718-3725. http://dx.doi.org/10.1109/CVPR.2014.475

[20] 李骞, 范茵, 张璟, 李宝强. 基于室外图像的天气现象识别方法[J]. 计算机应用, 2011, 31(6): 1-2.

[21] 吴小季. 基于SVM图像分类方法的研究[D]: [硕士学位论文]. 南京: 南京信息工程大学, 2011.

[22] 魏志静. 基于人工神经网络的分类方法研究及其在个人信用评估中的应用[D]: [硕士学位论文]. 济南: 山东师范大学, 2007.

[23] 孙秀亮. 基于属性加权的选择性朴素贝叶斯分类研究[D]: [硕士学位论文]. 哈尔滨: 哈尔滨工程大学, 2013.

[24] 许国根, 贾瑛. 模式识别与智能计算的MATLAB实现[M]. 北京: 北京航空航天大学出版社, 2012: 125-130.

[25] 朱乾根. 天气学原理和方法[M]. 南京: 南京气象出版社, 1992.

[26] Xu, G., Wang, X. and Xu, X. (2009) Improved Bi-Dimensional EMD and Hilbert Spectrum for the Analysis of Textures, Pattern Recognition, 42, 718-734. http://dx.doi.org/10.1016/j.patcog.2008.09.017

[27] Xu, G., Wang, X. and Xu, X. (2012) On Analysis of Bi-Dimensional Component Decomposition via BEMD. Pattern Recognition, 45, 1617-1626. http://dx.doi.org/10.1016/j.patcog.2011.11.004

分享
Top