基于多通道海洋激光雷达的海陆波形分类  

Classification of Sea and Land Waveform Based on Multi-Channel Ocean Lidar

作　　者：黄田程 陶邦一[1] 毛志华[1] 贺岩[2] 胡善江[2] 王聪聪[3] 俞家勇 陈鹏[1] 

机构地区：[1]国家海洋局第二海洋研究所卫星海洋环境动力学国家重点实验室,浙江杭州310012 [2]中国科学院上海光学精密机械研究所上海市全固态激光器与应用技术重点实验室,上海201800 [3]浙江大学计算机科学与技术学院,浙江杭州310013 [4]山东科技大学测绘科学与工程学院,山东青岛266590

出　　处：《中国激光》2017年第6期288-297,共10页Chinese Journal of Lasers

基　　金：国家重大科学仪器设备开发专项任务(2013YQ120343);国家重点研发计划(2016YFC1400901);中央级公益性科研院所基本科研业务费专项资金资助项目(QNYC201602)

摘　　要：针对国产机载激光雷达测深系统海陆波形分类识别的需求,基于多通道海洋激光雷达波形数据的特点,通过提取多通道波形的特征参数,采用支持向量机的方法构建分类模型对海陆波形进行分类。通过检验结果证明,该分类方法总体精度和Kappa系数分别达99.03%和0.9805。该方法海陆波形分类精度满足工程应用需求,适用于国产机载激光雷达测深系统的波形分类处理,为后续深度计算过程中水体介质光速校正和潮汐波浪改正奠定基础。To solve the problem of sea and land waveform classification for domestic airborne lidar bathymetric system, support vector machine method is used to build the model for classification of sea and land waveform. This method is based on the features of multi-channel ocean lidar waveform data, and characteristic parameters of multi- channel waveform are extracted. The validation results indicate that overall accuracy and Kappa coefficient of the classification method are 99. 03M and 0. 9805, respectively. The accuracy of the sea and land waveform classification of the proposed method is good enough for the engineering application. And it is suit for the waveform classification treatment of domestic airborne lidar bathymetric system. Which can lay the foundation for correcting the velocity of light in the water medium during the depth calculation procedure, tide and wave correction.

关 键 词：遥感 激光雷达 海陆 波形分类 支持向量机 

分 类 号：O436[机械工程—光学工程]