夜间激光雷达成像数据的降噪技术  

Noise reduction technique of night lidar imaging data

作　　者：马瑞敏 杜丽美 MA Ruimin;DU Limei(Changzhi University,Changzhi Shanxi 046000,China)

机构地区：[1]长治学院,山西长治046000

出　　处：《激光杂志》2023年第5期205-209,共5页Laser Journal

基　　金：山西省高等学校大学生创新创业训练计划重点项目(No.2020600);长治学院校级科研项目(No.XJZZ202101)。

摘　　要：为改善图像质量,设计夜间激光雷达成像数据的降噪技术。测算目标与雷达之间的距离信息,生成目标激光雷达图像;通过快速混合蛙跳算法增强目标激光雷达图像,采用Contourlet变换算法进行图像去噪,通过拉普拉斯塔形分解与方向滤波器组完成图像的多分辨率多方向分解,并利用Contourlet逆变换重构激光雷达图像,获取去噪后的结果。经实验验证:该技术在测量激光雷达图像时误差较小,最高为0.6像素,图像增强后的夜间激光雷达成像峰值信噪比与白天的差值为11 dB,且降噪后图像结构相似度可达到0.9以上。In order to improve the image quality,the noise reduction technology of night lidar imaging data is de-signed.Calculate the distance information between the target and the radar and generate the target lidar image;the tar-get lidar image is enhanced by fast hybrid leapfrog algorithm,and the Contourlet transform algorithm is used for image denoising,the multi-resolution and multi-directional decomposition of the image is completed by Laplacian decompo-sition and directional filter bank,and the inverse contourlet transform is used to reconstruct the lidar image to obtain the denoised results.The experimental results show that the error of this technology is small when measuring lidar ima-ges,up to 0.6 pixels.The difference between the peak signal-to-noise ratio of lidar imaging at night and that during the day after image enhancement is 11 dB,and the similarity of image structure after noise reduction can reach more than 0.9.

关 键 词：夜间激光雷达 成像数据 降噪技术 方向滤波器 快速混合蛙跳 CONTOURLET变换 

分 类 号：TN953[电子电信—信号与信息处理]