5種煙草常用農藥的高光譜識別技術研究

打開文本圖片集

中圖分類號：S220 文獻標志碼：A 文章編號：1008-0864（2025）07-0122-11

Identification of 5 Common PesticidesUsed in Flue-tobacco Field Production Based on Hyperspectral Technology

YANGGuotao1，ZHANGShijie2，CHENChao，LIUYun2，HEChen1， NINGYinghao'，ZHANG Qing1* （1.ZhengzhouTobaccoResearch InstituteofChina NationalTobaccoCorp，Zhengzhou450ool，China;.Baoji TobaccoCompany of Shaanxi Province，Shaanxi Baoji 721oo4，China；3.China National Tobacco Coporationof Shaanxi District， Xi'an ，China）

Abstract：Torapidlyandaccuratelyobtain thepesticide typesused intobacco field production，andto improve the scientificand targeted management of tobacco field pesticide applications，hyperspectral imaging technology was applied.By comparing the spectral curves offresh tobacco leaves 48 h after spraying with 5common pesticides，the spectral differences were analyzed.Various combinations of spectral preprocessing methods，feature wavelength extraction methods，and pattern recognition techniques were tested to evaluate model accuracy.The results showed that，in the75O＼～875nm spectral range，the spectral curves of fresh tobacco leaves treated with the 5 pesticides exhibited distinct reflectance diferences.Acrossthefull spectral range，thecombinationof standard normal variate （SNV） transformation and least squares support vector machine （LSSVM） models，as well as the combination of second derivative preprocessingandrandom forest models，bothachieved high recognition accuracy，with the testset accuracy reaching 98.58 % .The continuous projection algorithm outperformed the competitive adaptive reweighting samplingalgorithmin dimensionalityreduction.Inthefeaturewavelengthrange，thecombinationofsecond derivative preprocessing，continuous projection algorithm，and random forest models achieved the best performance， with the training set accuracy reaching 100.00% and the test set accuracy at 98.22% .Thenumber of feature wavelengths was17，and the single-sample detection time was 1O.28 ms.This method could rapidlyand accurately identify the types of5commonpesticides usedin tobacco production.Aboveresultsprovided technicalsupportforthe management of pesticide applicationin tobacco field production. Key words：tobacco field production；hyperspectral imaging；pesticide identification；random forest

農藥由一種或多種化學物質組成，既可以人工合成，也可以來源于生物或其他天然物質，主要用于預防、消滅或控制農業(yè)和林業(yè)生產中的害蟲、病菌、雜草等有害生物，同時也能調節(jié)植物和昆蟲的生長。（剩余13900字）