首页|Reports on Machine Learning Findings from Nanning Normal University Provide New Insights (Inversion of winter wheat leaf area index from UAV multispectral image s: classical vs. deep learning approaches)
Reports on Machine Learning Findings from Nanning Normal University Provide New Insights (Inversion of winter wheat leaf area index from UAV multispectral image s: classical vs. deep learning approaches)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-A new study on artificial intelligence is now available. According to news reporting from Nanning,People's Republic o f China,by NewsRx journalists,research stated,"Precise and timely leaf area i ndex (LAI) estimation for winter wheat is crucial for precision agriculture. The emergence of high-resolution unmanned aerial vehicle (UAV) data and machine lea rning techniques offers a revolutionary approach for fine-scale estimation of wh eat LAI at the low cost." The news correspondents obtained a quote from the research from Nanning Normal U niversity: "While machine learning has proven valuable for LAI estimation,there are still model limitations and variations that impede accurate and efficient L AI inversion. This study explores the potential of classical machine learning mo dels and deep learning model for estimating winter wheat LAI using multispectral images acquired by drones. Initially,the texture features and vegetation indic es served as inputs for the partial least squares regression (PLSR) model and ra ndom forest (RF) model. Then,the ground-measured LAI data were combined to inve rt winter wheat LAI. In contrast,this study also employed a convolutional neura l network (CNN) model that solely utilizes the cropped original image for LAI es timation. The results show that vegetation indices outperform the texture featur es in terms of correlation analysis with LAI and estimation accuracy. However,t he highest accuracy is achieved by combining both vegetation indices and texture features to invert LAI in both conventional machine learning methods. Among the three models,the CNN approach yielded the highest LAI estimation accuracy (R2 = 0.83),followed by the RF model (R2 = 0.82),with the PLSR model exhibited the lowest accuracy (R2 = 0.78). The spatial distribution and values of the estimat ed results for the RF and CNN models are similar,whereas the PLSR model differs significantly from the first two models."
Nanning Normal UniversityNanningPeop le's Republic of ChinaAsiaCyborgsEmerging TechnologiesMachine Learning
NETL
NSTL
