Study on Vegetation Classification Based on Spectral Knowledge Base

Peng Liu,Kaihua Wu,Xuexue Zhang,Jingcheng Zhang,Bin Wang

doi:10.1007/978-3-030-06179-1_32

Abstract

A framework about spectral based vegetation classification was proposed, which serves as a core methodology of the vegetation spectral knowledge base. The hyperspectral reflectances of 13 types of plants were measured by an ASD FieldSpec 4 spectroradiometer. Two forms of spectral features were used for representing the key spectral characteristics of plants, including Vegetation index (VI) and spectral shape features. Based on these spectral features, a sensitivity analysis was performed to identify the most important features for establishing the classifier. The analysis of variance (ANOVA) and the cross-correlation analysis were applied to derive the sensitivity of features and remove features that have high correlations. Then, a classification method for differentiating plants was established by coupling some spectral similarity measures (e.g., ED) with some classification methods (e.g., BPANN and SVM). The results of discrimination analysis showed that a highest accuracy was produced by SVM with the OAA over 99% when using 7 sensitive VIs. The results suggested the framework about spectral based vegetation classification can form a basis for spectral knowledge base and application technology and further achieve a wide range of plant classification based on remote sensing.

Highlights

Recent advances in Hyperspectral provide opportunities to map plant species and vegetation at various scales and resolutions
Based on hyperspectral measurements of a number of plant species, this paper focuses on the extraction and selection of spectral features for vegetation classification
10-25 measurements were made in the field of view, and 13 species of plant hyperspectral data (n = 324) was Measured, including Loropetalum chinense var.rubrum, Platycladus orientalis (L.) Franco, Ilex crenata Thunb., Sedum sarmentosum Bunge, Rhododendron simsii Planch., Nelumbo nucifera Gaertn. , Pyracantha, fortuneana (Maxim.) Li, Buxus megistophylla, Ligustrum vicaryi, Paspalum thunbergii Kunth ex Steud., Mahonia fortune (Lindl.) Fedde, Photinia serrulata Lindl. and Buxus sinica (Rehd. et Wils.) Cheng(Fig.1)

Summary

Introduction

Recent advances in Hyperspectral provide opportunities to map plant species and vegetation at various scales and resolutions. The study of vegetation classification methods based on hyperspectral data is an important part of plant spectrum library. Some detailed changes in spectral curves of hyperspectral data can be detected by spectral feature selection and extraction methods such as continuum removal or derivative analysis . Based on the pretreatment of original spectral data, this research uses a series of methods to analyze and compare the spectral curve, including derivative, Log(1/R), continuum removing an so on. It uses the characteristic parameters of spectral to classify and extract the plant. The artificial neural network method and factor analysis method were used to classify and extract typical vegetation

Methods

Results

Conclusion