|Table of Contents|

Study on Estimation Method of Canopy Coverage for Facility Tomato Based on RGB Images

《北方园艺》[ISSN:1001-0009/CN:23-1247/S]

Issue:
2024年3
Page:
41-50
Research Field:
Publishing date:

Info

Title:
Study on Estimation Method of Canopy Coverage for Facility Tomato Based on RGB Images
Author(s):
ZOU Weijie1; HUA Shan2; XU Zhifu2; XU Minjie2; LI Shuangwei2; BAO Wenna1
(1.School of Biological and Chemical Engineering，Zhejiang University of Science and Technology，Hangzhou,Zhejiang 310023；2.Institute of Agricultural Equipment，Zhejiang Academy of Agricultural Sciences/Key Laboratory of Southeastern Hilly and Mountainous Agricultural Equipment，Ministry of Agriculture and Rural Affairs，Hangzhou,Zhejiang 310021)
Keywords:
facility tomato; RGB image; vegetation index; canopy coverage; threshold segmentation
PACS:
S 641.2
DOI:
10.11937/bfyy.20232618
Abstract:
Taking tomato as the test material，the image data was obtained by using a visible light (RGB) camera system mounted on a scaffold under the facility environment，and the accuracy of different vegetation index algorithms for segmenting tomato canopy images was studied.And the evaluation of the method for extracting the canopy coverage of facility tomato was realized，in order to provide method guidance for the estimation of canopy coverage of other facility crops.The results showed that EXG algorithm，EXGR algorithm and CIVE algorithm could well be used to estimate the canopy coverage for facility tomatoes.The root mean square error (RMSE) between estimation value and true value for facility tomato canopy coverage was 0.049 for the EXG algorithm，0.078 for the EXGR algorithm，and 0.088 for the CIVE algorithm.The coefficient of determination (R2) was 0.911，0.845，0.841 for EXG algorithm，EXGR algorithm and CIVE algorithm respectively.The estimation results of canopy coverage for facility tomato were significant different among different vegetation index segmentation algorithms.Compared with the true value image，the EXGR algorithm had low segmentation accuracy，which overestimates the canopy coverage at 10 days after transplanting.Due to over-segmentation，the CIVE algorithm underestimated the canopy coverage at 66 days after transplanting.The EXG algorithm showed a higher segmentation accuracy at all growing stages and the estimated canopy coverage was close to the true value，compared to the other two algorithms.The EXG algorithm could achieve more efficient plant-soil segmentation and more accurate estimation for tomato canopy coverage than the other two algorithms，which provided methodological guidance of canopy coverage for other crops in facility.

References:

［1］李艳，王进，褚贵新，等.基于数字图像分析监测不同氮肥处理下加工番茄生长状况的初步研究［J］.植物营养与肥料学报，2008，14(1)：139-143.［2］GITELSON A A，KAUFMAN Y J，STARK R，et al.Novel algorithms for remote estimation of vegetation fraction［J］.Remote Sensing of Environment，2002，80(1)：76-87.［3］瞿瑛，刘素红，谢云.植被覆盖度计算机模拟模型与参数敏感性分析［J］.作物学报，2008，34(11)：1964-1969.［4］秦伟，朱清科，张学霞，等.植被覆盖度及其测算方法研究进展［J］.西北农林科技大学学报(自然科学版)，2006，34(9)：163-170.［5］刘琳，郑兴明，姜涛，等.无人机遥感植被覆盖度提取方法研究综述［J］.东北师大学报(自然科学版)，2021，53(4)：151-160.［6］ZHOU Q，ROBSON M.Automated rangeland vegetation cover and density estimation using ground digital images and a spectral-contextual classifier［J］.Infernational Journal of Remote Sensing，2001，22(17)：3457-3470.［7］章文波，符素华，刘宝元.目估法测量植被覆盖度的精度分析［J］.北京师范大学学报(自然科学版)，2001，37(3)：402-408.［8］尹林江，李威，赵卫权，等.水稻多时相植被指数特征及覆盖度提取研究［J］.中国农业科技导报，2023，25(2)：83-98.［9］邢著荣，冯幼贵，杨贵军，等.基于遥感的植被覆盖度估算方法述评［J］.遥感技术与应用，2009，24(6)：849-854.［10］敖登，杨佳慧，丁维婷，等.54种植被指数研究进展综述［J］.安徽农业科学，2023，51(1)：13-21,28.［11］任杰，柏延臣，王锦地.从数码照片中快速提取植被覆盖度的方法研究［J］.遥感技术与应用，2010，25(5)：719-724.［12］吴兰兰，熊利荣，彭辉.基于RGB植被指数的大田油菜图像分割定量评价［J］.华中农业大学学报，2019，38(2)：109-113.［13］张智韬，许崇豪，谭丞轩，等.覆盖度对无人机热红外遥感反演玉米土壤含水率的影响［J］.农业机械学报，2019，50(8)：213-225.［14］史培华，王远，袁政奇，等.基于冠层RGB图像的冬小麦氮素营养指标监测［J］.南京农业大学学报，2020，43(5)：829-837.［15］周华杰，刘亚东，付金东，等.基于冠层数码图像的玉米群体长势和氮素营养状态分析［J］.青岛农业大学学报(自然科学版)，2015，32(1)：1-7.［16］李荣春，陶洪斌，张竹琴，等.基于图像处理技术的夏玉米群体长势监测研究［J］.玉米科学，2010，18(2)：128-132.［17］张雅婷.番茄幼苗形态建成对UVA的生理响应研究［D］.北京:中国农业科学院，2021.［18］HUETE A R，LIU H，DE LIRA G R，et al.A soil color index to adjust for soil and litter noise in vegetation index imagery of arid regions［C］//1994 IEEE International Geoscience and Remote Sensing Symposium.Pasadena，CA，USA：IEEE，1994：1042-1043.［19］WOEBBECKE D M，MEYER G E，VON BARGEN K，et al.Color indices for weed identification under various soil，residue，and lighting conditions［J］.Transactions of the ASAE，1995，38(1)：259-269.［20］MEYER G E，NETO J C.Verification of color vegetation indices for automated crop imaging applications［J］.Computers and Electronics in Agriculture，2008，63(2)：282-293.［21］LOUHAICHI M,BORMAN M M,JOHNSON D E.Spatially located platform and aerial photography for documentation of grazing impacts on wheat［J］.Geocarto International，2001，16(1)：65-70.［22］JOHANSEN K,MORTON M J L,MALBETEAU Y M，et al.Unmanned aerial vehicle-based phenotyping using morphometric and spectral analysis can quantify responses of wild tomato plants to salinity stress［J］.Frontiers in Plant Science，2019，10:370.［23］李翠娜，张雪芬，余正泓，等.基于图像提取技术计算夏玉米覆盖度和反演叶面积指数的精度评价［J］.中国农业气象，2016，37(4)：479-491.［24］赵静，杨焕波，兰玉彬，等.基于无人机可见光图像的夏季玉米植被覆盖度提取方法［J］.农业机械学报,2019,50(5):232-240.

Memo

Memo:
-
Last Update: 2024-03-01