基于数学形态滤波的植被光谱去噪方法研究

发布时间：2018-01-04 12:24

本文关键词：基于数学形态滤波的植被光谱去噪方法研究　出处：《遥感技术与应用》2016年05期 　论文类型：期刊论文

【摘要】：光谱维噪声使地物光谱扭曲或变形,中心波长偏移,影响地物信息提取和地表参量反演的精度。对光谱维噪声进行滤波处理,有利于改善遥感数据定量应用的效果。由于数学形态滤波的原理简单且较易实现,被应用到植被光谱以及有机化合物光谱的研究中。运用数学形态滤波对地面实测小麦光谱去噪,一方面对滤波后的光谱进行噪声和波形相似度的直观分析,另一方面通过植被指数反演小麦理化参量进行定量应用评价。结果表明,与传统Savitzky-Golay滤波相比,在可见-近红外波段范围内,数学形态滤波去噪后的光谱能够保持可见—近红外波段原始光谱的固有特征,叶面积指数和叶绿素的反演精度比去噪前有小幅提升,主要原因是实测光谱在该谱段范围的噪声影响很小;在短波红外波段范围内,数学形态滤波能有效去除短波红外大尺度噪声,提高叶片含水量的反演精度。而传统Savitzky-Golay滤波只能削弱短波红外大尺度噪声。广义形态滤波去噪后植被指数和叶片含水量之间的R2最高可达0.5130(去噪前0.3753),叶片含水量的反演值与实测值之间的R2最高可达0.4221(去噪前0.3097),RMSE为0.0243(去噪前0.0318),优于传统Savitzky-Golay滤波。
[Abstract]:Spectral dimension noise distorts or deforms the spectrum of ground objects, and the center wavelength shifts, which affects the accuracy of ground object information extraction and surface parameter inversion. The spectral dimension noise is filtered and processed. It is helpful to improve the effect of quantitative application of remote sensing data, because the principle of mathematical morphological filtering is simple and easy to realize. It is applied to the study of vegetation spectrum and organic compound spectrum. On the one hand, the noise and waveform similarity of the filtered spectrum are analyzed directly by using mathematical morphological filter to remove noise from the measured wheat spectrum on the ground. On the other hand, the physical and chemical parameters of wheat were quantitatively evaluated by vegetation index inversion. The results showed that compared with the traditional Savitzky-Golay filter, it was in the range of visible and near infrared band. The spectrum after denoising by mathematical morphological filtering can maintain the inherent characteristics of the original spectrum in the visible-near infrared band. The retrieval accuracy of leaf area index and chlorophyll is slightly higher than that before denoising. The main reason is that the noise of the measured spectrum in the range of the spectrum is very small. In the range of short-wave infrared band, the mathematical morphological filter can effectively remove the short-wave infrared large-scale noise. The inversion accuracy of leaf water content is improved, while the traditional Savitzky-Golay filter can only weaken the large scale noise of short wave infrared. R2 between vegetation Index and Leaf Water content after denoising by Generalized morphological filter. Up to 0.5130 (. Before denoising 0.3753). The maximum R2 between the inversion value of leaf water content and the measured value was 0.4221 (the RMSE of 0.3097m before denoising was 0.0243 (0.0318 before denoising). It is superior to the traditional Savitzky-Golay filter.
【作者单位】：中国科学院遥感与数字地球研究所高光谱研究室;中国科学院大学资源环境学院;
【基金】：国家自然科学基金项目(41671360、40971205)
【分类号】：S512.1;TP79
【正文快照】： 1引言在地物光谱测量的过程中,由于受到光照条件、大气吸收、传感器信噪比下降等干扰因素的影响,目标地物光谱中往往存在不同尺度的噪声。为提高光谱信息提取和反演的精度,需要对含噪光谱进行去噪,以消除或减弱光谱维噪声的干扰,增强光谱的有效信息。根据去噪原理的不同,目前

【相似文献】