中国农业科学2024，Vol.57Issue(9)：1674-1686,13.DOI:10.3864/j.issn.0578-1752.2024.09.005
基于无人机多源影像数据的灌浆期人工合成小麦抗旱性评价
Drought Resistance Evaluation of Synthetic Wheat at Grain Filling Using UAV-Based Multi-Source Imagery Data
摘要
Abstract
[Objective]To evaluate the drought resistance of synthetic wheat germplasm based on multi-source images collected by unmanned aerial vehicle(UAV)and yield data,explore high-throughput indices for drought resistance evaluation,and identify synthetic wheat germplasm resources with drought resistance.This provides technical support and germplasm materials for accelerating the expansion of drought-resistant genetic resources for wheat and enhancing the level of breeding for dryland wheat.[Method]Eighty synthetic wheat germplasm and the control variety Xin Chun 37 were used as plant materials,which were sown in the field and treated with a water regime of drought stress and irrigation.Multi-source images of test materials during filling stage were collected by multi-spectral and thermal infrared cameras equipped with unmanned aerial vehicle,and the spectral index of each test material was extracted by threshold segmentation.The analyses of Pearson's correlation and principal component were performed to identify drought resistance-related spectral indices,and the drought resistance of each synthetic wheat germplasm was determined by single index and comprehensive evaluation methods.[Result]The drought resistance coefficients of 19 spectral indices of 80 synthetic wheat germplasm were calculated based on multisource imagery data obtained from unmanned aerial vehicles.The correlation analysis between the spectral indices and the yield-based drought index(DRI)showed that among the drought resistance coefficients of the 19 spectral indices,OSAVI exhibited the highest correlation with the drought index,while NDVI,CIre,and NDRE demonstrated relatively strong associations with the drought index.However,the different drought indices showed a high correlation,resulting in redundant information.The drought resistance coefficients of the 19 spectral indices were transformed into three independent comprehensive indicators through principal component analysis,with contribution rates of 59.6%,12.0%and 9.6%,respectively.The comprehensive drought resistance index(D)for each synthetic wheat germplasm were calculated by aggregating the three independent comprehensive indicators using the weighted membership function method.6 and 5 synthetic wheat germplasms with strong drought resistance were identified based on DRI and D,respectively.Among them,2 germplasms(SW004 and SW009)with high drought resistance were detected based on both DRI and D.Furthermore,the drought resistance of the 80 synthetic wheat germplasms was graded based on the drought resistance coefficient of OSAVI,and the grading results were found to be consistent with that based on the D value.Among the six strongly drought-resistant germplasms identified based on the drought resistance coefficient of OSAVI,five of them were also classified as strongly drought-resistant germplasms based on comprehensive drought resistance evaluation.[Conclusion]The spectral indices NDVI,OSAVI,CIre and NDRE extracted from UAV-based multi-source images,as well as the drought resistance comprehensive evaluation value can be used to assist in the identification of drought resistance of wheat germplasm.关键词
多源影像/光谱指数/人工合成小麦/抗旱性/灌浆期Key words
multi-source images/spectral index/synthetic wheat/drought resistance/grain filling引用本文复制引用
燕雯,金秀良,李龙,徐子涵,苏悦,张跃强,景蕊莲,毛新国,孙黛珍..基于无人机多源影像数据的灌浆期人工合成小麦抗旱性评价[J].中国农业科学,2024,57(9):1674-1686,13.基金项目
国家重点研发计划(2023YFD1201003,2022YFD1200201)、国家小麦产业技术体系(CARS-03-5) （2023YFD1201003,2022YFD1200201）
