国产日韩欧美一区二区三区三州_亚洲少妇熟女av_久久久久亚洲av国产精品_波多野结衣网站一区二区_亚洲欧美色片在线91_国产亚洲精品精品国产优播av_日本一区二区三区波多野结衣 _久久国产av不卡

?

利用隨機(jī)森林法協(xié)同SIF和反射率光譜監(jiān)測小麥條銹病

2019-08-23 02:32:08白宗璠劉良云
關(guān)鍵詞:條銹病冠層微分

競 霞,白宗璠,高 媛,,劉良云

利用隨機(jī)森林法協(xié)同SIF和反射率光譜監(jiān)測小麥條銹病

競 霞1,白宗璠1,高 媛1,2,劉良云2

(1. 西安科技大學(xué)測繪科學(xué)與技術(shù)學(xué)院,西安 710054;2. 中國科學(xué)院遙感與數(shù)字地球研究所,北京 100094)

小麥?zhǔn)艿綏l銹病菌侵染后,作物的光合能力及色素含量等均會發(fā)生變化,日光誘導(dǎo)葉綠素?zé)晒?solar-induced chlorophyll fluorescence,SIF)對作物光合生理的變化比較敏感,而反射率光譜則受作物生化參數(shù)的影響較大,為了提高小麥條銹病的遙感探測精度,該文利用隨機(jī)森林(random forest,RF)等機(jī)器學(xué)習(xí)算法開展了協(xié)同冠層SIF和反射率微分光譜指數(shù)的小麥條銹病病情嚴(yán)重度的遙感探測研究。首先利用3FLD(three bands fraunhofer line discrimination)算法提取了冠層SIF數(shù)據(jù),然后結(jié)合對小麥條銹病病情嚴(yán)重度敏感的11種反射率微分光譜指數(shù)分別基于RF和后向傳播(back propagation,BP)神經(jīng)網(wǎng)絡(luò)算法構(gòu)建了反射率微分光譜指數(shù)與冠層SIF協(xié)同的小麥條銹病病情嚴(yán)重度預(yù)測模型。研究結(jié)果表明:RF算法構(gòu)建的小麥條銹病病情嚴(yán)重度預(yù)測模型優(yōu)于BP神經(jīng)網(wǎng)絡(luò)算法,3個樣本組中RF模型病情指數(shù)(disease index,DI)估測值與實(shí)測值間的決定系數(shù)2平均為0.92,比BP神經(jīng)網(wǎng)絡(luò)模型(2的平均值為0.83)提高了11%,均方根誤差(root mean square error,RMSE)平均為0.08,比同組BP神經(jīng)網(wǎng)絡(luò)模型(RMSE的平均值為0.12)減少了33%,RF算法更適合于小麥條銹病病情嚴(yán)重度的遙感探測。在反射率微分光譜指數(shù)中加入冠層SIF數(shù)據(jù)后,RF模型和BP神經(jīng)網(wǎng)絡(luò)模型精度均有所改善,其中RF模型估測值與實(shí)測值間的平均2提高了4%,平均RMSE減少了22%,BP神經(jīng)網(wǎng)絡(luò)模型估測值與實(shí)測值間的平均2提高了14%,平均RMSE減少了28%,綜合利用冠層SIF和反射率微分光譜指數(shù)能夠改善小麥條銹病病情嚴(yán)重度的遙感探測精度。研究結(jié)果可為進(jìn)一步實(shí)現(xiàn)作物健康狀況大面積高精度遙感監(jiān)測提供新的思路。

遙感;監(jiān)測;小麥;條銹??;隨機(jī)森林;日光誘導(dǎo)葉綠素?zé)晒?;微分光譜指數(shù)

0 引 言

小麥條銹?。ǎ┦且环N發(fā)生范圍廣,危害程度大的病害[1],在條銹病流行的年份,會導(dǎo)致小麥減產(chǎn)40%以上,甚至絕收[2]。因此,及時掌握小麥條銹病的發(fā)病狀態(tài),對中國農(nóng)業(yè)生產(chǎn)具有重要的意義。由于遙感技術(shù)在作物病害探測中具有快速、宏觀、無損等傳統(tǒng)田間取樣調(diào)查病害方法難以比擬的優(yōu)勢,近年來利用遙感技術(shù)探測小麥條銹病已成為眾多專家學(xué)者研究的熱點(diǎn),并取得了豐碩的成果[3-10]。其中Shi等[5]以反射率指數(shù)為自變量,建立基于小波變換的小麥條銹病估測模型,研究結(jié)果顯示該方法可以很好地預(yù)測條銹病的發(fā)病狀況。劉琦等[6]以不同的光譜特征為自變量,采用定性偏最小二乘方法建立小麥條銹病潛育期識別模型,結(jié)果表明,偽吸收系數(shù)二階導(dǎo)數(shù)為自變量的識別模型精度最高,訓(xùn)練集和檢驗(yàn)集的識別率分別為97.89%和92.98%。蔣金豹等[9]采用冠層一階微分?jǐn)?shù)據(jù)作為自變量構(gòu)建小麥條銹病估測模型,研究結(jié)果表明以紅邊峰值區(qū)與綠邊峰值區(qū)一階微分總和的比值為自變量建立的估測模型精度最高。

上述研究中,小麥條銹病的遙感探測主要是利用反射率光譜數(shù)據(jù)以及由反射率光譜計(jì)算的一階微分?jǐn)?shù)據(jù)。但反射率光譜主要反映生化組分的濃度信息,不能直接揭示植被光合生理狀態(tài)[11-12],而冠層SIF能夠更敏感地反映病害引起的生理變化[13]及受脅迫狀態(tài)[14],起到植物健康狀態(tài)“探針”作用[15]。張永江等[16]利用標(biāo)準(zhǔn)FLD方法預(yù)測了小麥條銹病不同病情嚴(yán)重度的日光誘導(dǎo)葉綠素?zé)晒猓C實(shí)了冠層SIF信息可以反映田間小麥條銹病的發(fā)病狀況。然而小麥?zhǔn)軛l銹病菌侵染后,其水分及葉綠素含量、光合速率和光能轉(zhuǎn)換率等一些生理生化指標(biāo)均會發(fā)生變化[17],僅利用反射率光譜數(shù)據(jù)或冠層SIF數(shù)據(jù)難以全面客觀的映射小麥條銹病害的真實(shí)狀況,影響小麥條銹病的遙感探測精度。

小麥條銹病病情嚴(yán)重度反演算法目前使用較多的是回歸分析法[6,8,18],該方法計(jì)算簡單、易于實(shí)現(xiàn),然而由于數(shù)據(jù)獲取時外界條件的差異,基于回歸分析方法建立的數(shù)學(xué)統(tǒng)計(jì)模型的普適性較差[19]。因此也有一些學(xué)者將神經(jīng)網(wǎng)絡(luò)和支持向量機(jī)等機(jī)器學(xué)習(xí)算法應(yīng)用于小麥條銹病病情嚴(yán)重度的遙感探測中,并取得了較好的研究成果[20-22]。但是目前為止將具有良好抗噪能力的RF算法應(yīng)用于小麥條銹病遙感監(jiān)測的研究鮮見報(bào)道,尤其是基于RF算法綜合利用微分光譜數(shù)據(jù)和冠層SIF數(shù)據(jù)監(jiān)測小麥條銹病的研究。為此,本文融合反射率光譜在作物生化參數(shù)探測方面的優(yōu)勢和葉綠素?zé)晒庠诠夂仙碓\斷方面的優(yōu)勢,基于RF和BP神經(jīng)網(wǎng)絡(luò)2種機(jī)器學(xué)習(xí)算法構(gòu)建了冠層SIF與微分光譜指數(shù)協(xié)同的小麥條銹病病情嚴(yán)重度預(yù)測模型,并利用保留樣本交叉檢驗(yàn)方法對模型精度進(jìn)行分析與評價,以確定小麥條銹病遙感探測的適宜方法和模型,以期為提高小麥條銹病病情嚴(yán)重度的遙感探測精度提供一種新的思路,對其他作物病害的遙感探測提供有價值的參考。

1 材料與方法

1.1 試驗(yàn)設(shè)計(jì)

試驗(yàn)區(qū)位于河北省廊坊市中國農(nóng)業(yè)科學(xué)院試驗(yàn)站(39°30¢402N,116°36¢202E),小麥品種為對條銹病比較敏感的銘賢169號。試驗(yàn)區(qū)小麥平均種植密度為113棵/m2。2018年4月9日采用濃度為9 mg/100 mL的孢子溶液對小麥進(jìn)行條銹病接種。試驗(yàn)區(qū)域小麥分為健康組(編號為A、D)和染病組(編號為B、C),每個試驗(yàn)組的面積為220 m2。每個組分為8個樣方(A1-A8、B1-B8、C1-C8、D1-D8),即健康組和染病組各16個樣方。

1.2 病情指數(shù)調(diào)查與田間冠層光譜測定

1.2.1 冠層光譜測定

本試驗(yàn)分別于2018年5月18日、5月24日和5月30日3個時期測定小麥條銹病不同病情嚴(yán)重度下的冠層光譜數(shù)據(jù),測量儀器為ASD Field Spec 4地物光譜儀,其光譜分辨率為3 nm,采樣間隔1.4 nm,采樣波長范圍350~2 500 nm,測量時間為北京時間11∶00-12∶30,測量高度距離地面1.3 m,探頭視場角25°,每個采樣點(diǎn)觀測10次并對觀測結(jié)果取平均作為該采樣點(diǎn)的光譜數(shù)據(jù),每次測量前后均用標(biāo)準(zhǔn)BaSO4參考板對冠層輻亮度數(shù)據(jù)進(jìn)行校正。

1.2.2 病情指數(shù)調(diào)查

冠層病情指數(shù)調(diào)查采用5點(diǎn)取樣法,在每個樣方內(nèi)選取對稱的5點(diǎn),每點(diǎn)約1 m2,隨機(jī)選取30株小麥,分別調(diào)查其發(fā)病情況。病情嚴(yán)重度參照國家標(biāo)準(zhǔn)“小麥條銹病測報(bào)技術(shù)規(guī)范”(GB/T 15795)[23]進(jìn)行量化。單葉嚴(yán)重度分為9個梯度,即0、1%、10%、20%、30%、45%、60%、80%和100%的葉片病斑覆蓋,分別記錄各嚴(yán)重度的小麥葉片數(shù),按式(1)計(jì)算不同梯度測試群體的病情指數(shù)[8,24]。

式中DI為病情指數(shù);為各梯度級值;為最高梯度等級值;為各梯度的葉片數(shù)。

1.2.3 日光誘導(dǎo)葉綠素?zé)晒獾?FLD算法

目前常用的日光誘導(dǎo)葉綠素?zé)晒獾墓罍y方法為基于夫瑯和費(fèi)暗線原理的標(biāo)準(zhǔn)FLD (fraunhofer line discrimi-nation) 算法。該方法假設(shè)吸收線內(nèi)外反射率和透過率相等,通過比較夫瑯和費(fèi)吸收線內(nèi)外2個波段植被冠層上行輻亮度和下行輻照度光譜差異計(jì)算夫瑯和費(fèi)吸收線處的葉綠素?zé)晒鈴?qiáng)度[25]。

式中I,I為夫瑯和費(fèi)吸收線內(nèi)、外的太陽輻照度光譜強(qiáng)度(W/cm2/nm);L,L為夫瑯和費(fèi)吸收線內(nèi)、外的植被冠層反射的輻亮度光譜強(qiáng)度 (W/cm2/nm/sr)。

標(biāo)準(zhǔn)FLD算法是基于吸收線內(nèi)外反射率和透過率相等的假設(shè)估測日光誘導(dǎo)葉綠素?zé)晒鈴?qiáng)度,但由于吸收線內(nèi)外波段的反射率和熒光值實(shí)際上存在差異,影響了熒光的估測精度[26]。為了克服標(biāo)準(zhǔn)FLD方法局限性,Maier等[27]提出了一種改進(jìn)的3FLD熒光預(yù)測算法,該算法認(rèn)為在吸收線波段周圍葉綠素?zé)晒夂头瓷渎使庾V是線性變化的,利用吸收線左右各一個波段的加權(quán)平均值代替標(biāo)準(zhǔn)FLD算法中的單一波段值,從而在一定程度上減小標(biāo)準(zhǔn)FLD方法中熒光和反射率恒定假設(shè)所帶來的誤差,提高了冠層SIF的預(yù)測精度[28],而且3FLD方法也是SIF估測的最魯棒算法[29],基于此本文采用3FLD方法計(jì)算冠層SIF值,計(jì)算公式如式(3)~(5)所示。

為了減弱冠層光譜數(shù)據(jù)測試時不同時間段太陽光照強(qiáng)度等外界因素對日光誘導(dǎo)葉綠素?zé)晒夤浪阒档挠绊?,提高日光誘導(dǎo)葉綠素?zé)晒忸A(yù)測精度,本文將計(jì)算得到的日光誘導(dǎo)葉綠素?zé)晒獾慕^對強(qiáng)度(W/cm2/nm/sr)分別除以參考板獲取的夫瑯和費(fèi)吸收線內(nèi)的太陽入射輻照度,得到該吸收線處的日光誘導(dǎo)葉綠素?zé)晒獾南鄬?qiáng)度[30-31]。如式(6)所示

1.2.4 高光譜微分指數(shù)

為了快速尋找植被光譜曲線的彎曲點(diǎn)及最大最小反射率的波長位置等特征參量以及分解重疊的吸收波段,常常對原始光譜數(shù)據(jù)進(jìn)行微分處理而得微分光譜。通過對反射率光譜進(jìn)行微分處理能夠增強(qiáng)光譜曲線在坡度上的細(xì)微變化[32],去除部分線性或接近線性的背景、噪聲光譜對目標(biāo)光譜的影響[33],本文結(jié)合已有的高光譜微分指數(shù)監(jiān)測小麥條銹病方面的研究成果[18,34],選取的微分光譜指數(shù)如表1所示。

表1 微分光譜指數(shù)

由于光譜采樣間隔的離散性,一階微分光譜通常用差分方法近似計(jì)算[33]

1.3 模型構(gòu)建

1.3.1 BP神經(jīng)網(wǎng)絡(luò)模型

神經(jīng)網(wǎng)絡(luò)模型中,BP神經(jīng)網(wǎng)絡(luò)是應(yīng)用范圍比較廣泛的一種建模方法。BP神經(jīng)網(wǎng)絡(luò)模型的拓?fù)浣Y(jié)構(gòu)包括輸入層(input)、隱含層(hide layer)和輸出層(output layer),層與層之間的神經(jīng)元通過相應(yīng)的網(wǎng)絡(luò)權(quán)重系數(shù)相互聯(lián)系,每層內(nèi)的神經(jīng)元之間不連接。其核心思想是通過調(diào)整各神經(jīng)元之間的權(quán)值,將誤差由隱含層向輸入層逐層反傳,對誤差函數(shù)進(jìn)行“鏈?zhǔn)角髮?dǎo)”,使誤差逼近最小值[35]。本文中BP神經(jīng)網(wǎng)絡(luò)模型由輸入層(對小麥條銹病敏感的各類特征參量)、隱含層和輸出層(小麥條銹病病情嚴(yán)重度)構(gòu)成。為了得到更好的網(wǎng)絡(luò)訓(xùn)練效果,本文在建模前對各輸入變量進(jìn)行了歸一化處理。

1.3.2 隨機(jī)森林模型

隨機(jī)森林是一種基于分類回歸樹的機(jī)器學(xué)習(xí)算法[36],能夠?qū)⒍喾N決策樹算法結(jié)合起來,對同一現(xiàn)象進(jìn)行重復(fù)預(yù)測[37],其基本思想是通過bootstrap重采樣的方法在原始訓(xùn)練集中抽取多個樣本,對每個抽取出的樣本均進(jìn)行決策樹建模,最后通過多數(shù)投票法得到最終的預(yù)測結(jié)果[38]。本文中RF模型的輸入變量為小麥條銹病的各個敏感因子,輸出變量為小麥條銹病的病情嚴(yán)重度。

2 結(jié)果與分析

2.1 相關(guān)性分析

2.1.1 冠層SIF與DI的相關(guān)性分析

由于O2-A (760 nm)波段氧氣吸收形成的夫瑯和費(fèi)暗線特征明顯[39],熒光估測精度高[40]。因此本文利用測定的小麥條銹病不同病情嚴(yán)重度下太陽及冠層輻亮度數(shù)據(jù)通過3FLD方法計(jì)算O2-A暗線處的日光誘導(dǎo)葉綠素?zé)晒庀鄬?qiáng)度,并在此基礎(chǔ)上分析小麥條銹病病情嚴(yán)重度和葉綠素?zé)晒庀鄬?qiáng)度的關(guān)系(圖1)。由圖1可以看出,O2-A波段處的日光誘導(dǎo)葉綠素?zé)晒庀鄬?qiáng)度與小麥條銹病病情嚴(yán)重度達(dá)到了極顯著負(fù)相關(guān)。這是由于O2-A波段的冠層SIF主要受植被吸收光合有效輻射APARchl(absorbed photosynthetic active radiation)影響[41],隨著小麥條銹病病情嚴(yán)重度的增加,光合作用活性減弱,進(jìn)而導(dǎo)致APARchl降低[42]和冠層SIF值減小,因此冠層SIF與小麥條銹病病情嚴(yán)重度之間呈現(xiàn)極顯著負(fù)相關(guān)關(guān)系。

注:**表示1%水平極顯著。下同

2.1.2 一階微分光譜指數(shù)與DI的相關(guān)性分析

為了篩選對小麥條銹病病情嚴(yán)重度敏感的微分光譜指數(shù),本文首先利用公式(7)對小麥條銹病不同病情嚴(yán)重度下的冠層光譜數(shù)據(jù)進(jìn)行一階微分處理,得到各微分光譜指數(shù),并將其與小麥條銹病病情嚴(yán)重度進(jìn)行相關(guān)性分析(表2)。由表2可以看出除SDg、Dg與小麥條銹病病情嚴(yán)重度的相關(guān)性不顯著外,其他11個微分光譜指數(shù)均與小麥條銹病病情嚴(yán)重度達(dá)到了極顯著相關(guān),可以作為自變量構(gòu)建小麥條銹病病情嚴(yán)重度的預(yù)測模型。

表2 DI與微分光譜指數(shù)之間的關(guān)系

2.2 模型構(gòu)建與精度分析

在進(jìn)行小麥條銹病病情嚴(yán)重度模型構(gòu)建和精度評價時,為了使評價結(jié)果更客觀,本文將53個樣本數(shù)據(jù)(47個染病樣本,6個健康樣本)重復(fù)進(jìn)行3次隨機(jī)分組(記為a、b、c),每組中40個數(shù)據(jù)(35個染病樣本,5個健康樣本)作為訓(xùn)練樣本用于模型構(gòu)建,剩余的13個數(shù)據(jù)(12個染病樣本,1個健康樣本)作為驗(yàn)證樣本用以模型評價。

2.2.1 病情嚴(yán)重度預(yù)測模型的構(gòu)建

本文分別基于RF算法和BP神經(jīng)網(wǎng)絡(luò)算法構(gòu)建小麥條銹病病情嚴(yán)重度估測模型,并對2種算法的模型精度進(jìn)行評價,以確定小麥條銹病病情嚴(yán)重度遙感探測的適宜算法和模型。

論文通過對訓(xùn)練樣本的多次仿真確定隨機(jī)森林算法中決策樹的數(shù)量(ntree)為500,內(nèi)部節(jié)點(diǎn)隨機(jī)選擇屬性個數(shù)(mtry)取默認(rèn)值。BP算法采用3層網(wǎng)絡(luò)標(biāo)準(zhǔn)結(jié)構(gòu),其中隱含層設(shè)置為5個神經(jīng)元,最大訓(xùn)練次數(shù)為5 000,訓(xùn)練間隔為10,最小均方根誤差為0.001,學(xué)習(xí)步長為0.1。

在構(gòu)建小麥條銹病病情嚴(yán)重度預(yù)測模型時,本文分別以反射率微分光譜指數(shù)以及冠層SIF協(xié)同反射率微分光譜指數(shù)作為預(yù)測模型的輸入?yún)?shù),利用RF和BP神經(jīng)網(wǎng)絡(luò)算法構(gòu)建了小麥條銹病病情嚴(yán)重度預(yù)測模型(表3)。

表3 訓(xùn)練集模型預(yù)測結(jié)果

Table 3 Training set model prediction result

由表3可以看出,加入冠層SIF數(shù)據(jù)后,3個樣本組中RF模型和BP神經(jīng)網(wǎng)絡(luò)模型DI估測精度較單一反射率微分光譜數(shù)據(jù)均有一定程度的提高。無論是以反射率微分光譜還是冠層SIF協(xié)同微分光譜作為模型的輸入變量,RF模型的預(yù)測精度均高于同組BP神經(jīng)網(wǎng)絡(luò)模型,RF算法更適合于小麥條銹病病情嚴(yán)重度的遙感探測。

2.2.2 模型評價

為了保證評價結(jié)果的可靠性和穩(wěn)定性,本文采用保留樣本交叉檢驗(yàn)方式,利用建模剩余的13個數(shù)據(jù)作為驗(yàn)證樣本,分別對RF算法以及BP神經(jīng)網(wǎng)絡(luò)算法所構(gòu)建的小麥條銹病病情嚴(yán)重度預(yù)測模型進(jìn)行檢驗(yàn),結(jié)果如圖2和圖3所示。

由圖2和圖3可以看出,同時把SIF和微分光譜指數(shù)作為自變量,RF模型和BP神經(jīng)網(wǎng)絡(luò)模型的小麥條銹病病情嚴(yán)重度的預(yù)測精度較同組僅以微分光譜指數(shù)為自變量的模型精度都有一定的提高,其中RF模型估測DI值和實(shí)測DI值間的平均2提高了4%,平均RMSE減少了22%。BP神經(jīng)網(wǎng)絡(luò)的估測DI值與實(shí)測DI值間的平均2提高了14%,平均RMSE提高了28%,無論是微分光譜指數(shù)單獨(dú)作為自變量還是SIF與微分光譜指數(shù)共同作為自變量,RF算法構(gòu)建的小麥條銹病病情嚴(yán)重度的預(yù)測模型均優(yōu)于同組BP神經(jīng)網(wǎng)絡(luò)模型,3個樣本組中RF模型DI估測值與實(shí)測值間的決定系數(shù)2平均為0.92,比BP神經(jīng)網(wǎng)絡(luò)模型(2的平均值為0.83)提高了11%,RMSE平均為0.08,比同組BP神經(jīng)網(wǎng)絡(luò)模型(RMSE的平均值為0.12)減少了33%。因此RF算法構(gòu)建的小麥條銹病病情嚴(yán)重度的預(yù)測模型優(yōu)于BP神經(jīng)網(wǎng)絡(luò)模型,更適合于小麥條銹病病情嚴(yán)重度的遙感探測。

圖2 小麥條銹病病情嚴(yán)重度RF預(yù)測模型驗(yàn)證

圖3 小麥條銹病病情嚴(yán)重度BP神經(jīng)網(wǎng)絡(luò)預(yù)測模型驗(yàn)證

3 討 論

冠層SIF和反射率光譜數(shù)據(jù)在小麥條銹病病情嚴(yán)重度的遙感探測中各有優(yōu)勢和局限性。反射率光譜受土壤顏色、陰影或者其它非綠色景觀成分等背景噪聲的影響較大[43],且對作物光合活性不敏感,但可以很好的監(jiān)測作物色素含量變化[44-45],而SIF則受背景噪聲的影響較小[46],對作物光合生理上的變化比較敏感,SIF和光合作用之間的直接聯(lián)系能夠改善植物脅迫遙感監(jiān)測精度[43]。本文的研究結(jié)果也表明協(xié)同冠層SIF和反射率微分光譜指數(shù)為自變量構(gòu)建的RF模型和BP神經(jīng)網(wǎng)絡(luò)模型精度均高于反射率微分光譜指數(shù)的模型精度。其中,RF模型預(yù)測DI值與實(shí)測DI值間的RMSE平均減小了22%,BP神經(jīng)網(wǎng)絡(luò)模型的RMSE平均減小了28%。與已有研究結(jié)果相比,本文綜合利用冠層SIF數(shù)據(jù)和反射率微分光譜指數(shù)所建模型預(yù)測DI值與實(shí)測DI值間的RMSE分別減少了14%[9]和25%[10],加入冠層SIF數(shù)據(jù)能夠提高小麥條銹病的遙感探測精度,這與陳思媛等的研究結(jié)論相一致[47]。這是因?yàn)樾←準(zhǔn)艿綏l銹病菌侵染后,APARchl和葉綠素含量降低,冠層SIF強(qiáng)度與APARchl密切相關(guān),反射率微分光譜指數(shù)則對植物生化組分的濃度信息敏感,協(xié)同反射率微分光譜指數(shù)和日光誘導(dǎo)葉綠素?zé)晒鈹?shù)據(jù)能夠充分利用反射率光譜在作物生化參數(shù)探測方面的優(yōu)勢及葉綠素?zé)晒庠诠夂仙碓\斷方面的優(yōu)勢,從而提高小麥條銹病病情嚴(yán)重度的遙感估測精度。

在構(gòu)建小麥條銹病病情嚴(yán)重度估測模型時,RF模型估測DI值與實(shí)測DI值間的2比BP神經(jīng)網(wǎng)絡(luò)模型提高了11%,RMSE減少了33%,其中,以反射率微分光譜指數(shù)為自變量構(gòu)建的RF模型預(yù)測DI值和實(shí)測DI值之間2平均為0.90,RMSE平均為0.09,2比BP神經(jīng)網(wǎng)絡(luò)平均提高了17%,RMSE平均減少了36%。加入冠層SIF數(shù)據(jù)后,RF模型預(yù)測DI值和實(shí)測DI值之間2平均為0.94,RMSE平均為0.07,2比BP神經(jīng)網(wǎng)絡(luò)平均提高了6%,RMSE平均減少了30%。因此無論是否在模型特征參量中加入冠層SIF數(shù)據(jù),RF模型預(yù)測精度都優(yōu)于BP神經(jīng)網(wǎng)絡(luò)模型,這與姚雄等關(guān)于林地葉面積指數(shù)的遙感估算的研究結(jié)果相一致[19],王麗愛等在估算小麥葉綠素含量的研究中也表明RF模型的估測效果優(yōu)于BP神經(jīng)網(wǎng)路模型[48]。這主要是由于RF算法具有良好的抗噪能力,不易陷入過度擬合[37],而BP神經(jīng)網(wǎng)絡(luò)在訓(xùn)練過程中由于學(xué)習(xí)了過多的樣本細(xì)節(jié),學(xué)習(xí)出的模型難以體現(xiàn)樣本內(nèi)含的規(guī)律[19]。

本文將冠層SIF和微分光譜指數(shù)作為自變量構(gòu)建小麥條銹病病情嚴(yán)重度估測模型時,沒有考慮不同參數(shù)對模型貢獻(xiàn)率的問題,如何確定各參量因子的權(quán)重并應(yīng)用于模型的構(gòu)建還有待繼續(xù)研究。

論文在構(gòu)建小麥條銹病病情嚴(yán)重度的遙感探測模型時,僅對比分析了RF和BP神經(jīng)網(wǎng)絡(luò)2種機(jī)器學(xué)習(xí)算法,如果使用更多的模型構(gòu)建算法參與病情嚴(yán)重度估測精度的對比分析,RF模型是否仍是小麥條銹病遙感探測的最優(yōu)模型,還需要進(jìn)行進(jìn)一步的探測。

本文在構(gòu)建小麥條銹病病情嚴(yán)重度估測模型時,僅將冠層SIF和反射率微分光譜指數(shù)直接拼接作為模型的輸入?yún)⒘?,并沒有考慮不同特征參量與病情指數(shù)之間的映射關(guān)系,如何找到不同特征參量與病情指數(shù)之間的最優(yōu)映射函數(shù),針對融合后的反射率光譜-冠層SIF特征空間,利用多核學(xué)習(xí)算法建立基于特征最優(yōu)核映射的小麥條銹病病情嚴(yán)重度的遙感探測模型則是下步要研究的問題。

4 結(jié) 論

為了提高小麥條銹病的遙感探測精度,本文綜合利用反射率和冠層SIF數(shù)據(jù)在植物生理和生化探測中的優(yōu)勢,基于BP神經(jīng)網(wǎng)絡(luò)和RF算法開展了冠層SIF與反射率微分光譜指數(shù)協(xié)同的小麥條銹病遙感探測研究。主要結(jié)論如下:

1)反射率微分光譜指數(shù)以及冠層SIF協(xié)同反射率微分光譜指數(shù)均能實(shí)現(xiàn)小麥條銹病的遙感探測。但自變量中加入冠層SIF數(shù)據(jù)后,RF模型和BP神經(jīng)網(wǎng)絡(luò)模型的探測精度均有一定程度的提高,其中RF模型預(yù)測DI值和實(shí)測DI值間的RMSE平均減少了22%,BP神經(jīng)網(wǎng)絡(luò)模型的RMSE平均減小了28%。

2)無論利用反射率微分光譜指數(shù)還是冠層SIF協(xié)同反射率微分光譜指數(shù)作為自變量,RF模型的估測精度均高于BP神經(jīng)網(wǎng)絡(luò)模型,更適合小麥條銹病的探測。其中,3個樣本組中RF模型的DI估測值與實(shí)測值的RMSE比同組BP神經(jīng)網(wǎng)絡(luò)模型減少了33%。以反射率微分光譜指數(shù)為自變量構(gòu)建的RF模型預(yù)測DI值和實(shí)測DI值間的RMSE比BP神經(jīng)網(wǎng)絡(luò)模型平均減小了36%,加入冠層SIF數(shù)據(jù)后RF模型的RMSE比BP神經(jīng)網(wǎng)絡(luò)模型至少減少了30%。

[1] 王靜,景元書,黃文江,等.冬小麥條銹病嚴(yán)重度不同估算方法對比研究[J]. 光譜學(xué)與光譜分析,2015,35(6):1649-1653.Wang Jing, Jing Yuanshu, Huang Wenjiang, et al. Comparative study on different estimation methods of severity of stripe rust of winter wheat[J]. Spectroscopy and Spectrum Analysis. 2015, 35(6): 1649-1653. (in Chinese with English abstract)

[2] 王利民,劉佳,楊福剛,等. 基于GF-1/WFV數(shù)據(jù)的冬小麥條銹病遙感監(jiān)測[J]. 農(nóng)業(yè)工程學(xué)報(bào),2017,33(20):153-160.Wang Limin, Liu Jia, Yang Fugang, et al. Winter wheat stripe rust remote sensing monitoring based on GF-1/WFV [J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2017, 33(20): 153-160. (in Chinese with English abstract)

[3] Liang D, Liu N, Zhang D, et al. Discrimination of powdery mildew and yellow rust of winter wheat using high-resolution hyperspectra and imageries[J]. Infrared and Laster Enginee-ring. 2017, 46(1): 1-8

[4] Zheng Q, Huang W, Cui X, et al. New spectral index for detecting wheat yellow rust using sentinel-2 multispectral imagery[J]. Sensors 2018, 868(18): 1-19

[5] Shi Y, Huang W, Gonzalez-Moreno P, et al. Wavelet-based rust spectral feature set(WRSFs):A novel spectral feature set based on continuous wavelet transformation for tracking progressive host-pathogen interaction of yellow rust on wheat [J]. Remote Sensing 2018, 252(10): 1-19

[6] 劉琦,谷醫(yī)林,王翠翠,等. 基于偏最小二乘法的小麥條銹病潛育期冠層高光譜分析[J]. 植物保護(hù)學(xué)報(bào),2018,45(1):138-145.Liu Qi, Gu Yilin, Wang Cuicui, et al. Canopy hyperspectral features analysis of latent period wheat stripe rustbased on discriminant partial least squares [J]. Journal of Plant Protection 2018, 45(1): 138-145. (in Chinese with English abstract)

[7] Hui W, Feng Q, Liu R, et al.Identification and severity determination of wheat stripe rust and wheat leaf rust based on hyperspectral data acquired using a black-paper- based measuring method[J]. Plos one. 2016, 11(4): e0154648.

[8] 魯軍景,黃文江,蔣金豹,等. 小波特征與傳統(tǒng)光譜特征預(yù)測冬小麥條銹病病情嚴(yán)重度的對比研究[J]. 麥類作物學(xué)報(bào),2015,35(10):1456-1461.Lu Junjing, Huang Wenjiang, Jiang Jinbao, et al. A comparative study on the severity of stripe rust disease in winter wheat estimated by little baud sign and traditional spectral features [J]. Journal of Wheat Crops, 2015, 35(10): 1456-1461. (in Chinese with English abstract)

[9] 蔣金豹,陳云浩,黃文江,等. 冬小麥條銹病嚴(yán)重度高光譜遙感反演模型研究[J]. 南京農(nóng)業(yè)大學(xué)學(xué)報(bào),2007,30(3):63-67. Jiang Jinbao, Chen Yunhao, Huang Wenjiang, et al. Study on hyperspectral remote sensing retriveralmodels about winter wheat stripe rust severity[J]. Journal of Nanjing Agricultural University. 2007, 30(3): 63-67. (in Chinese with English abstract)

[10] 陳云浩,蔣金豹,黃文江,等. 主成分分析法與植被指數(shù)經(jīng)驗(yàn)方法估測冬小麥條銹病嚴(yán)重度的對比研究[J]. 光譜學(xué)與光譜分析,2009,29(8):2161-6165. Chen Yunhao, Jiang Jinbao, Huang Wenjiang, et al. Comparison of Principal Component Analysis with VI-EmpiricalApproach for Estimating Severity of Yellow Rust of Winter Wheat[J]. Spectroscopy and Spectral Analysis, 2009, 29(8): 2161-6165. (in Chinese with English abstract)

[11] McMurtrey J E, Middleton E M, Crop L A, et al. Optical reflectance and fluorescence for detecting nitrogen needs in zea mays L[C] //Geoscience and Remote Sensing Symposium, 2003. IGARSS’03.Proceedings. 2003 IEEE international. IEEE, 2003, 7: 4602-4604

[12] Ashour loo D, Mobasheri M R, Huete A. Developing two spectral disease indices for detection for wheat leaf rust ()[J]. Remote Sensing, 2014, 6(6): 4723-4740.

[13] Subhash N, Subhash N, Ravi V, et al. Detection of mosaic virus disease in cassava plants by sunlight-induced fluorescence imaging: A pilot study for proximal sensing[J]. International Journal of Remote Sensing, 2015, 36(11): 2880-2889.

[14] Liu L, Zhao J, Guan L. Tracking photosynthetic injury of Paraquat-treated crop using chlorophyll fluorescence from hyperspectral data[J]. European Journal of Remote Sensing, 2013, 46: 459-473.

[15] Krause G H, Weis E. Chlorophyll fluorescence and photosynthesis: the basics[J]. Annual Review of Plant Biology, 1991, 42(1): 313-349.

[16] 張永江,黃文江,王紀(jì)華,等. 基于Fraunhofer 線的小麥條銹病熒光遙感探測[J]. 中國農(nóng)業(yè)科學(xué),2007,40(1):78-83. Zhang Yongjiang, Huang Wenjiang, Wang Jihua, et al. Chlorophyll fluorescence sensing to detect stripe rust in wheat(L.)fields based on fraunhofer lines[J]. Scientia Agricultura Sinica, 2007, 40(1): 78-83. (in Chinese with English abstract)

[17] 劉琦,王翠翠,王睿,等. 潛育期小麥條銹菌的高光譜定性識別[J]. 植物保護(hù)學(xué)報(bào),2018,45(1):153-160. Liu Qi, Wang Cuicui, Wang Rui, et al. Hyperspectral qualitative identification on latent period of wheat stripe rust[J]. Journal of Plant Protection, 2018, 45(1): 153-160. (in Chinese with English abstract)

[18] 蔣金豹,陳云浩,黃文江,等. 利用高光譜微分指數(shù)進(jìn)行冬小麥條銹病病情的診斷研究[J]. 光學(xué)技術(shù),2007,4(39):620-623.Jiang Jinbao, Chen Yunhao, Huang Wenjiang, et al. Diagnosis of stripe rust disease in winter wheat by using hyperspectral differential index[J]. Optics Technology, 2007, 4(39): 620-623. (in Chinese with English abstract)

[19] 姚雄,余坤勇,楊玉潔,等. 基于隨機(jī)森林模型的林地葉面積指數(shù)遙感估算[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2017,48(5):159-166.Yao Xiong, Yu Kunyong, Yang Yujie, et al. Remote sensing estimation of woodland leaf area index based on random forest model [J]. Transactions of the Chinese Society for Agricultural Machinery, 2017, 48(5): 159-166. (in Chinese with English abstract)

[20] 胡小平,楊之為,李振岐,等. 漢中地區(qū)小麥條銹病的BP神經(jīng)網(wǎng)絡(luò)預(yù)測[J]. 西北農(nóng)業(yè)學(xué)報(bào),2000,9(3):28-31.Hu Xiaoping, Yang Zhiwei, Li Zhenqi, et al. BP neural network prediction of wheat stripe rust in hanzhong region [J]. Journal of Northwest Agriculture, 2000, 9(3): 28-31. (in Chinese with English abstract)

[21] L Mo. Prediction of wheat stripe rust using neural network[C]. IEEE International conference on intelligent computing & intelligent systems. 2010, 3: 475-479.

[22] Wang H, Ma Z. Prediction of wheat stripe based on support vector machine[C]. International conference on natural computation. 2011, 1: 378-382.

[23] 中華人民共和國國家質(zhì)量監(jiān)督檢驗(yàn)檢疫總局中國國家標(biāo)準(zhǔn)化管理委員會. GB/T 15795-2011小麥條銹病測報(bào)技術(shù)規(guī)范[S]北京:中國標(biāo)準(zhǔn)出版社,2011-11-19.

[24] 王紀(jì)華,趙春江,黃文江.農(nóng)業(yè)定量遙感基礎(chǔ)與應(yīng)用[M]. 北京:科學(xué)出版社,2008.

[25] Plascyk K A, Gabriel F C. The Fraunhofer line discriminator MKII-an airborne instrument for precise and standardized ecological luminescence measurement[J]. IEEE Transactions on Instrumentation and Measurement. 1975, 24(4): 306-313.

[26] 劉良云.植被定量遙感原理與應(yīng)用[M].北京:科學(xué)出版社,2014.

[27] Maier S W, Günther K P, Stellmes M. Sun-induced fluorescence: A new tools for precision farming[M]// Mcdonald M, Schepers J, Tartly L, et al. Digital imaging and spectral techniques: Aplications to precision agriculture and crop physiology. Madison, WI, USA: American Society of Agronomy Special Publication, 2003: 209-222.

[28] 劉新杰.日光誘導(dǎo)葉綠素?zé)晒獾倪b感反演研究[D]. 北京:中國科學(xué)院大學(xué),2016. Liu Xinjie. Study of Sun-Induced Cholorophy Retrieval Using Remote Sensing Methods[D]. Beijing: University of Chinese Academy of Sciences, 2016. (in Chinese with English abstract)

[29] Liu X, Liu L. Improving chlorophyll fluorescence retrieval using reflectance reconstruction based on principal components analysis[J]. IEEE Geoscience & Remote Sensing Letters, 2015, 12(8): 1645-1649.

[30] Liu L, Cheng Z. Detection of vegetation light-use efficiency based on solar-induced chlorophyll fluorescence separated from canopy radiance spectrum[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2010, 3(3): 306-312.

[31] Cordon G, Lagorio M G, Paruelo J M. Chlorophyll fluorescence, photochemical reflective index and normalized difference vegetative index during plant senescence[J]. Journal of Plant Physiology, 2016, 199, 100-110.

[32] 童慶禧,張兵,鄭芬蘭,等.高光譜遙感——原理、技術(shù)與應(yīng)用[M],北京:高等教育出版社,2015.

[33] 浦瑞良,宮鵬.高光譜遙感及應(yīng)用[M],北京:高等教育出版社,2003.

[34] 蔣金豹,陳云浩,黃文江. 用高光譜微分指數(shù)預(yù)測條銹病脅迫下小麥冠層葉綠素密度[J]. 光譜學(xué)與光譜分析,2010,30(8):2243-2247.Jiang Jinbao, Chen Yunhao, Huang Wenjiang. Estimation of canopy chlorophyll density of wheat under stripe rust stress by using hyperspectral differential index[J]. Spectroscopy and Spectrum Analysis, 2010, 30(8): 2243-2247. (in Chinese with English abstract)

[35] 袁冰清,程功,鄭柳剛. BP神經(jīng)網(wǎng)絡(luò)基本原理[J].數(shù)字通信世界,2008,8(17),28-29.Yuan Bingqing, Cheng Gong, Zheng Liugang. Basic principles of BP neural network [J]. Digital communication world. 2008, 8(17): 28-29. (in Chinese with English abstract)

[36] Breiman L. Random forest[J]. Machine Learning, 2001, 45(1): 5-32.

[37] Rodriguez-Galiano V, Mendes M P, Garcia-Soldado M J, et al. Precictive modeling of groungwater nitrate pollution using random forest and multisource variables related to intrinsic and specific vulnerability:A case study in an agricultural setting(Southern Spain)[J]. Science of the Total Environment.2014, 476-477: 189-206.

[38] 李健麗,董瑩瑩,師越,等. 基于隨機(jī)森林模型的小麥白粉病遙感監(jiān)測方法[J]. 植物保護(hù)學(xué)報(bào),2018,45(2):395-396.Li Jianli, Dong Yingying, Shi Yue, et al. Remote sensing monitoring method of wheat powdery mildew based on random forest model [J]. Journal of plant protection, 2008, 45(2): 395-396. (in Chinese with English abstract)

[39] 劉良云,張永江,王紀(jì)華,等. 利用夫瑯禾費(fèi)暗線探測自然光條件下的植被光合作用熒光研究[J]. 遙感學(xué)報(bào),2006,10(1):130-137.Liu Liangyun, Zhang Yongjiang, Wang Jihua, et al. Fluorescence study under natural light detection by using Fraunhofer dark line [J]. Journal of remote sensing.2006, 10(1): 130-137. (in Chinese with English abstract)

[40] 胡姣嬋,劉良云,劉新杰. FluorMOD模擬葉綠素?zé)晒夥颥樅藤M(fèi)暗線反演算法不確定性分析[J]. 遙感學(xué)報(bào),2015,19(4):594-608.Hu Jiaochan, Liu Liangyun, Liu Xinjie. Uncertainty analysis of FluorMOD simulated chlorophyll fluorescence fraunhofer inverse algorithm [J]. Journal of Remote Sensing, 2015, 19(4): 594-608. (in Chinese with English abstract)

[41] Du S, Liu L, Liu X, et al. Response of canopy solar-induced chlorophyll fluorescence to the absorbed photosynthetically active radiation absorbed by chlorophyll[J]. Remote Sensing, 2017, 9(9): 911.

[42] Porcarcastell A, Tyystj?rvi E, Atherton J, et al. Linking chlorophyll a fluorescence to photosynthesis for remote sensing applications: mechanisms and challenges[J]. Journal of Experimental Botany, 2014, 65(15): 4065-4095.

[43] Song L, Guanter L, Guan K, et al. Satellite sun-induced chlorophyll fluorescence detects early response of winter wheat to heat stress in the Indian Indo-Gangetic Plains[J]. Global Change Biology, 2018, 24: 4023-4037.

[44] Beck P S A, Goetz S J. Satellite observations of high northern latitude vegetation productivity changes between 1982 and 2008: Ecological variability and regional differences[J]. Environmental Research Letters, 2011, 6(4): 049501.

[45] Gamon J A, Kovalchuck O, Wong C Y S, et al. Monitoring seasonal and diurnal changes in photosynthetic pigments with automated PRI and NDVI sensors[J]. Biogeosciences, 2015, 12(3): 4149-4159.

[46] Baker, Neil R. Chlorophyll Fluorescence: A Probe of Photosynthesis In Vivo[J]. Annu Rev Plant Biol, 2008, 59(1): 89-113.

[47] 陳思媛,競霞,董瑩瑩,等. 基于日光誘導(dǎo)葉綠素?zé)晒馀c反射率光譜的小麥條銹病探測研究[J]. 遙感技術(shù)與應(yīng)用,2019,34(3):511-520. Chen Siyuan, Jing Xia, Dong Yingying, et al. Detection of wheat stripe rust using solar-induced chlorophyll fluorescence and reflectance spectral indices[J]. Remote Sensing Technology and Application, 2019, 34(3): 511-520. (in Chinese with English abstract)

[48] 王麗愛,馬昌,周旭東,等. 基于隨機(jī)森林回歸算法的小麥葉片SPAD值遙感估算[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2015,46(1):259-265. Wang Liai, Ma Chang, Zhou Xudong, et al. Estimation of wheat leaf spad value using rf algorithmic model and remote sensing data[J]. Transactions of the Chinese Society for Agricultural Machinery, 2015, 46(1):259-265. (in Chinese with English abstract)

Wheat stripe rust monitoring by random forest algorithm combined with SIF and reflectance spectrum

Jing Xia1, Bai Zongfan1, Gao Yuan1,2, Liu Liangyun2

(1.,’,710054,; 2.,,100094,)

The prevalence of wheat stripe rust has a significant impact on the production of winter wheat all over the world. An effective monitoring and warning of this disease is imperative to ensure the quality of wheat production. Remote sensing detection of wheat stripe rust is important for agriculture management and decision. The reflectance spectrum is closely related to the changes of biomass. It cannot, however, directly reveal the photosynthetic physiological state of vegetation. Solar-induced chlorophyll fluorescence(SIF) can sensitively reflect the photosynthetic vitality of crops, and the canopy’s solar-induced chlorophyll fluorescence signal includes the fluorescence characteristics of physiological changes caused by plant disease stress. In order to improve detection precision of wheat stripe rust, this study made full use of the advantages of reflectance spectroscopy for the detection of crop biochemical parameters and the advantages of chlorophyll fluorescence in photosynthetic physiological diagnosis, a remote sensing study on the severity of wheat stripe rust was carried out by using random forest (RF) and other machine learning algorithms synergistic SIF and reflectance differential spectral index in the canopy of wheat. Firstly, based on Fraunhofer line principle, three bands fraunhofer line discrimination(3FLD) algorithm was used to predict the intensity of chlorophyll fluorescence in O2-A band (760 nm). Then 11 reflectance differential spectral indices, which are sensitive to the severity of wheat stripe rust disease were selected. Based on RF and back propagation(BP) neural network algorithm, a model for predicting the severity of wheat stripe rust with differential reflectance spectral index and canopy SIF was established. The study incorporated a cross-checking method based on measurements of control samples. Fifty-two raw crop samples were randomly divided into two parts three times, the first part including 39 datasets was used as the training set for the model building, and the remaining 13 data samples were used to evaluate the accuracy of the models. The results showed that: 1) There is a significant negative correlation between SIF and the disease severity of wheat stripe rust. Remote sensing detection of wheat stripe rust severity can both be realized using the differential spectral index alone or by using the differential spectral index and the solar-induced chlorophyll fluorescence in combination. However, the accuracy of the estimates made by the RF and BP neural network models using the combination of data from the differential spectral index and the solar-induced chlorophyll fluorescence were all higher than that for the models constructed using the differential spectral index alone. In the three sample groups, average determination coefficient between the estimated DI using the RF model and the BP neural network model and the measured DI increased by 4% and 14% respectively, and the average RMSE decreased by 33% and 28% respectively. The detection accuracy of wheat stripe rust severity can be improved using solar-induced chlorophyll fluorescence combined reflectance differential spectral index. 2) The canopy solar-induced chlorophyll fluorescence synergistic differential spectral index were used as sensitive factors, the coefficients of determination between the estimated DI using the RF model and the measured DI were 0.90, 0.93, and 0.98, respectively, which were greater than the coefficients produced when using the BP neural network model for the same group (0.88, 0.84, and 0.92). Similarly, the RMSEs were 0.09, 0.07, and 0.04, respectively, which were smaller than the RMSEs (0.10, 0.11, and 0.09) using the BP neural network model for the same group. Therefore, the model using the RF algorithm was better at estimating wheat stripe rust severity than the BP neural network-based model, and it is more suitable for the remote sensing detection of wheat stripe rust severity. These results have important significance for improving the accuracy of the real-world remote sensing detection of wheat stripe rust, and the analysis provides new ideas for further realizing large-area remote sensing monitoring of crop health.

remote sensing; monitoring; wheat; stripe rust; random forest; solar-induced chlorophyll fluorescence; differential spectral index

10.11975/j.issn.1002-6819.2019.13.017

S512.1+1

A

1002-6819(2019)-13-0154-09

2018-11-18

2019-05-24

國家自然科學(xué)基金(41601467)

競 霞,博士,副教授,研究方向?yàn)檗r(nóng)業(yè)定量遙感。Email:jingxiaxust@163.com

競 霞,白宗璠,高 媛,劉良云.利用隨機(jī)森林法協(xié)同SIF和反射率光譜監(jiān)測小麥條銹病[J]. 農(nóng)業(yè)工程學(xué)報(bào),2019,35(13):154-161. doi:10.11975/j.issn.1002-6819.2019.13.017 http://www.tcsae.org

Jing Xia, Bai Zongfan, Gao Yuan, Liu Liangyun. Wheat stripe rust monitoring by random forest algorithm combined with SIF and reflectance spectrum [J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2019, 35(13): 154-161. (in Chinese with English abstract) doi:10.11975/j.issn.1002-6819.2019.13.017 http://www.tcsae.org

猜你喜歡
條銹病冠層微分
小麥條銹病田間為害損失的初步分析
湖北植保(2022年4期)2022-08-23 10:51:52
基于低空遙感的果樹冠層信息提取方法研究
基于激光雷達(dá)的樹形靶標(biāo)冠層葉面積探測模型研究
陜西安康:農(nóng)技專家開展小麥條銹病普防工作
擬微分算子在Hp(ω)上的有界性
安徽省淮南森林冠層輻射傳輸過程的特征
上下解反向的脈沖微分包含解的存在性
小麥條銹病持續(xù)控制的策略
215份小麥材料抗條銹病基因分子檢測初報(bào)
施氮水平對冬小麥冠層氨揮發(fā)的影響
乌海市| 兴海县| 城固县| 陕西省| 灵川县| 和龙市| 特克斯县| 南郑县| 朔州市| 泸溪县| 安泽县| 进贤县| 西和县| 仙居县| 醴陵市| 武鸣县| 纳雍县| 长沙市| 高唐县| 吴江市| 阿拉善盟| 昌平区| 兴义市| 中牟县| 门头沟区| 云阳县| 乌兰浩特市| 绥滨县| 平山县| 寿宁县| 玛曲县| 龙游县| 鹤庆县| 平度市| 思茅市| 本溪| 临沂市| 马龙县| 理塘县| 凤庆县| 应城市|