Machine learning-based spectral and spatial analysis of hyper-and multi-spectral leaf images for Dutch elm disease detection and resistance screening  

作　　者：Xing Wei Jinnuo Zhang Anna O.Conrad Charles E.Flower Cornelia C.Pinchot Nancy Hayes-Plazolles Ziling Chen Zhihang Song Songlin Fei Jian Jin 

机构地区：[1]Department of Agricultural and Biological Engineering,Purdue University,West Lafayette,IN 47907,United States [2]USDA Forest Service,Northern Research Station,Hardwood Tree Improvement and Regeneration Center,West Lafayette,IN 47906,United States [3]USDA Forest Service,Northern Research Station,Delaware,OH 43015,United States [4]Department of Forestry and Natural Resources,Purdue University,West Lafayette,IN 47907,United States

出　　处：《Artificial Intelligence in Agriculture》2023年第4期26-34,共9页农业人工智能（英文）

摘　　要：Diseases caused by invasive pathogens are an increasing threat to forest health,and early and accurate disease detection is essential for timely and precision forest management.The recent technological advancements in spectral imaging and artificial intelligence have opened up new possibilities for plant disease detection in both crops and trees.In this study,Dutch elm disease(DED;caused by Ophiostoma novo-ulmi,)and American elm(Ulmus americana)was used as example pathosystem to evaluate the accuracy of two in-house developed high-precision portable hyper-and multi-spectral leaf imagers combined with machine learning as new tools for forest disease detection.Hyper-and multi-spectral images were collected from leaves of American elm geno-types with varied disease susceptibilities after mock-inoculation and inoculation with O.novo-ulmi under green-house conditions.Both traditional machine learning and state-of-art deep learning models were built upon derived spectra and directly upon spectral image cubes.Deep learning models that incorporate both spectral and spatial features of high-resolution spectral leaf images have better performance than traditional machine learning models built upon spectral features alone in detecting DED.Edges and symptomatic spots on the leaves were highlighted in the deep learning model as important spatial features to distinguish leaves from inoculated and mock-inoculated trees.In addition,spectral and spatial feature patterns identified in the machine learning-based models were found relative to the DED susceptibility of elm genotypes.Though further studies are needed to assess applications in other pathosystems,hyper-and multi-spectral leaf imagers combined with machine learning show potential as new tools for disease phenotyping in trees.

关 键 词：American elm Dutch elm disease Hyperspectral imaging Multispectral imaging Support vector machine Convolution neural network Disease phenotyping Digital forestry 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] O43[自动化与计算机技术—控制科学与工程]