基于引导滤波的Mie散射光谱分析  

Analysis of Mie scattering spectrum based on guided filtering

作　　者：肖韶荣 周炜炜 

机构地区：[1]江苏大气海洋光电探测重点实验室,江苏南京210044 [2]江苏省大气环境与装备技术协同创新中心,江苏南京210044 [3]南京信息工程大学物理与光电工程学院,江苏南京210044

出　　处：《应用光学》2017年第6期923-930,共8页Journal of Applied Optics

基　　金：公益性(气象)行业专项(GYHY201206037);江苏省科技支撑计划(BE2010733)

摘　　要：为提高白光散射法反演粒子粒径精度,消除散射谱噪声,使用一维引导滤波对不同角度的Mie散射光谱数据进行滤波,并分析了引导滤波的特性。为解决引导数据引起的局部噪声问题,提出了小波-引导滤波联合算法。在小散射角时,引导滤波的结果与小波滤波类似,但在大散射角时,引导滤波的效果要比小波滤波好,并且滤除了脉冲噪声。小波-引导联合滤波的结果在15°附近与引导滤波相同,在45°到90°时的滤波效果明显比单独的小波滤波或引导滤波效果好,并且去除了由引导数据引起的局部噪声。In order to improve the accuracy of particle size inversion, eliminate the scattering spectrum noise, the Mie scattering spectrum data of different angles were filtered by one dimensional guided filter, and the characteristics of the guided filter were analyzed. In order to solve the problem of local noise caused by the guidance data, a wavelet-guided filtering algorithm was proposed. The results show that the filtering result is similar to that of the wavelet filter in the small scattering angle, but the filtering effect is better than the wavelet filter in the large scattering angle, and the pulse noise may be filtered out. The result of wavelet-guided joint filtering is the same as that of the guided filter at 15°, the filtering effect is better than that of the wavelet filter or the guided filter at 45° to 90°,and the local noise caused by the guidance data can be removed.

关 键 词：引导滤波 Mie散射光谱 小波变换 脉冲噪声 

分 类 号：TN29[电子电信—物理电子学] TH741[机械工程—光学工程]