SegVeg: Segmenting RGB Images into Green and Senescent Vegetation by Combining Deep and Shallow Methods

作     者：Mario Serouart Simon Madec Etienne David Kaaviya Velumani Raul LopezLozano Marie Weiss Frederic Baret Mario Serouart;Simon Madec;Etienne David;Kaaviya Velumani;Raul Lopez Lozano;Marie Weiss;Frédéric Baret

作者机构：ArvalisInstitutduvegetal228routedel’aerodrome-CS4050984914 Avignon Cedex 9France INRAEAvignon UniversiteUMREMMAHUMTCAPTE228routedel’aerodrome-CS4050984914 Avignon Cedex 9France CIRADUMRTETISF-34398 MontpellierFrance Hiphen SAS228routedel’aerodrome-CS4050984914 Avignon Cedex 9France 

出 版 物：《Plant Phenomics》 (植物表型组学（英文）)

年 卷 期：2022年第4卷第1期

页      面：26-42页

核心收录：

学科分类：08[工学] 080203[工学-机械设计及理论] 0802[工学-机械工程] 

基　　金：The study was partly supported by several projects including ANR PHENOME(Programme d’investissement d’avenir) Digitag(PIA Institut Convergences Agriculture Numérique ANR-16-CONV-0004) CASDAR LITERAL and P2S2 funded by CNES 

主　　题：Deep offering render 

摘      要：Pixel segmentation of high-resolution RGB images into chlorophyll-active or nonactive vegetation classes is a first step often required before estimating key traits of *** have developed the SegVeg approach for semantic segmentation of RGB images into three classes(background,green,and senescent vegetation).This is achieved in two steps:A U-net model is first trained on a very large dataset to separate whole vegetation from *** green and senescent vegetation pixels are then separated using SVM,a shallow machine learning technique,trained over a selection of pixels extracted from *** performances of the SegVeg approach is then compared to a 3-class U-net model trained using weak supervision over RGB images segmented with SegVeg as groundtruth *** show that the SegVeg approach allows to segment accurately the three ***,some confusion is observed mainly between the background and senescent vegetation,particularly over the dark and bright regions of the *** U-net model achieves similar performances,with slight degradation over the green vegetation:the SVM pixel-based approach provides more precise delineation of the green and senescent patches as compared to the convolutional nature of *** use of the components of several color spaces allows to better classify the vegetation pixels into green and ***,the models are used to predict the fraction of three classes over whole images or regularly spaced *** show that green fraction is very well estimated(R^(2)=0.94)by the SegVeg model,while the senescent and background fractions show slightly degraded performances(R^(2)=0.70 and 0.73,respectively)with a mean 95%confidence error interval of 2.7%and 2.1%for the senescent vegetation and background,versus 1%for green *** have made SegVeg publicly available as a ready-to-use script and model,along with the entire annotated grid-pixels *** thus hope to render segmentation acc