the 基于小波变换视频能见度检测算法的研究与实现guide download.pdfVIP

第 31 卷 第 1 期 仪 器 仪 表 学 报 Vol.31 No. 1 2010 年 1 月 Chinese Journal of Scientific Instrument Jan. 2010 基于小波变换的视频能见度检测算法研究与实现* 陈钊正，周庆逵，陈启美 (南京大学通信技术研究所 南京 210093) 摘 要：由于传统能见度仪价格昂贵、架设困难、不能给出道路能见度分布，提出了基于小波变换的视频能见度检测算法。 利用摄像机自标定模型恢复图像距离信息，采用小波变换提取图像边缘特征点，使用能见度检测模型提取符合人眼的特征， 结合曲线拟合，给出能见度变化曲线。该算法及其系统已在江苏宁常镇溧高速公路上进行试点使用，其结果与人眼及能见度 仪检测误差在+10%之内，符合相关国际及国家标准。 关键词：小波变换；摄像机自标定；能见度检测模型；曲线拟合 中图分类号：TP391.41 文献标识码：A 国家标准学科分类代码：520.6040 Video visibility detection algorithm based on wavelet transformation Chen Zhaozheng, Zhou Qingkui, Chen Qimei (Institute of Communication Technology, Nanjing University, Nanjing 210093, China) Abstract ：Traditional visibility measurement sensor is expensive, hard to install and can’t give road visibility distribution. This paper presents a method of video visibility detection based on wavelet transform. Camera self-calibration is used to recover the distance information from images and wavelet transform is used to get the points that belong to the edges of the target objects; then a visibility detection model is used to get the feature that is in accordance with human eyes; finally, with curve fitting, the visibility changing curve is obtained. Now the system based on the proposed model and algorithm has been used on Ning-Chang and Zhen-Li high-way; the ap- plication results of the system are consistent with those from human eyes and visibility measurement sensors, and the error between them is with +10%, which conforms to national and foreign standards. Key words ：wavelet transform; camera self-calibration; visibility detection model; curve fitting 小算法检测值与人眼观测值之间的误差。 1 引 言