Deep convolutional neural network for damaged vegetation segmentation from RGB images based on virtual NIR-channel estimation

作     者：Artzai Picon Arantza Bereciartua-Perez Itziar Eguskiza Javier Romero-Rodriguez Carlos Javier Jimenez-Ruiz Till Eggers Christian Klukas Ramon Navarra-Mestre 

作者机构：TECNALIABasque Research and Technology Alliance(BRTA)Parque Tecnológico de BizkaiaC/Geldo.Edificio 700E-48160 Derio-BizkaiaSpain University of the Basque CountryPlaza Torres Quevedo48013 BilbaoSpain BASF Espanola S.L.Carretera A37641710 Utrera SevillaSpain BASF SESpeyererstrasse 267117 LimburgerhofGermany 

出 版 物：《Artificial Intelligence in Agriculture》 (农业人工智能（英文）)

年 卷 期：2022年第6卷第1期

页      面：199-210页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 08[工学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

主　　题：Vegetation indices estimation Vegetation coverage map Near infrared estimation Convolutional neural network Deep learning 

摘      要：Performing accurate and automated semantic segmentation of vegetation is a first algorithmic step towards more complex models that can extract accurate biological information on crop health,weed presence and phenological state,among ***,models based on normalized difference vegetation index(NDVI),near infrared channel(NIR)or RGB have been a good indicator of vegetation ***,these methods are not suitable for accurately segmenting vegetation showing damage,which precludes their use for downstream phenotyping *** this paper,we propose a comprehensive method for robust vegetation segmentation in RGB images that can cope with damaged *** method consists of a first regression convolutional neural network to estimate a virtual NIR channel from an RGB ***,we compute two newly proposed vegetation indices from this estimated virtual NIR:the infrared-dark channel subtraction(IDCS)and infrared-dark channel ratio(IDCR)***,both the RGB image and the estimated indices are fed into a semantic segmentation deep convolutional neural network to train a model to segment vegetation regardless of damage or *** model was tested on 84 plots containing thirteen vegetation species showing different degrees of damage and acquired over 28 *** results show that the best segmentation is obtained when the input image is augmented with the proposed virtual NIR channel(F1=0:94)and with the proposed IDCR and IDCS vegetation indices(F1=0:95)derived from the estimated NIR channel,while the use of only the image or RGB indices lead to inferior performance(RGB(F1=0:90)NIR(F1=0:82)or NDVI(F1=0:89)channel).The proposed method provides an end-to-end land cover map segmentation method directly from simple RGB images and has been successfully validated in real field conditions.