中国农业科技导报2026，Vol.28Issue(6)：104-115,12.DOI:10.13304/j.nykjdb.2024.0860
基于地膜指数与遥感影像特征的农田地膜精准识别
Accurate Identification of Agricultural Land Mulch Based on Plastic Film Index and Remote Sensing Image Characteristics
摘要
Abstract
To address the current challenges of insufficient feature utilization and limited accuracy in remote sensing identification of agricultural plastic mulch,focusing on Qixian county,Kaifeng city,Henan province as the core study area,firstly,a novel remote sensing index——the plastic film index(PFI)was developed.Simultaneously,multi-dimensional feature sets encompassing spectral data,indices,texture and surface temperature were systematically extracted from Sentinel-2 and Landsat-8 satellite imagery.Building upon this foundation,a parameter-optimized recursive feature elimination(RFE)algorithm was employed for feature selection to eliminate redundant information.Finally,a random forest classifier was utilized to extract the distribution information of agricultural plastic film.The results indicated that the RFE algorithm optimized the initial 65 feature variables to 39,significantly enhancing the model's computational efficiency and robustness.The classification model based on optimized features achieved an overall accuracy of 92%with a Kappa coefficient of 0.89,and both producer and receiver accuracy for plastic film categories exceeded 90%.Compared with other approaches,the model incorporating the PFI index and feature selection demonstrated optimal recognition performance,confirming the effectiveness of the PFI index in enhancing plastic film identification accuracy.Above results not only provided precise spatial information for agricultural producers in residual film recovery and field management,but also offered reliable technical support and data backing for pollution assessment and decision-making by agricultural environmental protection departments.关键词
农田地膜/多源遥感数据/地膜指数/温度特征/特征优化Key words
agricultural mulch/multi-source remote sensing data/plastic film index/temperature characteristics/feature optimization分类
农业科技引用本文复制引用
位盼盼,郭燕,张彦,贺佳,杨秀忠,王来刚..基于地膜指数与遥感影像特征的农田地膜精准识别[J].中国农业科技导报,2026,28(6):104-115,12.基金项目
河南省农业科学院自主创新项目(2024ZC074) （2024ZC074）
国家重点研发计划项目(2022YFD2001105) （2022YFD2001105）
中央引导地方科技发展资金项目(Z20231811179). （Z20231811179）
