趙文舉,馬芳芳,馬 宏,周 春
基于無人機多光譜影像的土壤鹽分反演模型
趙文舉,馬芳芳,馬 宏,周 春
(蘭州理工大學能源與動力工程學院,蘭州 730050)
為探究不同作物覆蓋下不同深度的土壤鹽分快速反演模型,該研究采集苜蓿、玉米覆蓋下0~15、>15~30、>30~50 cm層深度的土壤鹽分含量,基于無人機多光譜影像數(shù)據(jù),提取各地塊采樣點的光譜反射率,在此基礎上引入紅邊波段計算光譜指數(shù)作為特征變量,采用支持向量機遞歸特征消除算法(Support Vector Machine-Recursive Feature Elimination,SVM-RFE)以篩選光譜指數(shù)及未經(jīng)過篩選的全指數(shù)組作為模型輸入組,共構建出36個基于隨機森林(Random Forest,RF)、極限學習機(Extreme Learning Machine,ELM)、BP神經(jīng)網(wǎng)絡(Back Propagation Neural Network)等機器學習模型,確定不同作物覆蓋下的最佳土壤鹽分反演模型。結(jié)果表明:SVM-RFE算法篩選光譜指數(shù)構建模型精度優(yōu)于未進行篩選構建的模型。對于苜蓿和玉米覆蓋土壤,整體上,RF反演效果優(yōu)于ELM模型和BPNN模型,反演結(jié)果能體現(xiàn)真實土壤鹽分含量,在0~15和>30~50 cm土層上,RF模型反演效果優(yōu)于其他模型,苜蓿樣地驗證集決定系數(shù)p2分別為0.71、0.58,驗證集均方根誤差RMSEp分別為0.026、0.033,玉米樣地p2分別為0.67、0.64,RMSEp分別為0.111、0.094,在>15~30 cm土層上ELM反演效果較好,苜蓿樣地p2為0.58,RMSEp為0.039,玉米樣地p2為0.68,RMSEp為0.059。0~15 cm是作物覆蓋下的土壤含鹽量最佳反演深度,驗證集平均決定系數(shù)2為0.65,均方根誤差RMSE為0.084。研究結(jié)果可為土壤鹽分的快速反演提供理論依據(jù)。
無人機;土壤;鹽分;多光譜;SVM-RFE;反演模型
土壤鹽堿化嚴重影響作物根系吸水能力和土壤結(jié)構,造成了大面積耕地退化和作物減產(chǎn)[1],據(jù)統(tǒng)計中國鹽堿土面積約3 600萬hm2,分布廣泛,其中耕地占26.6%[2],嚴重威脅中國1.2億hm2耕地紅線,鹽堿地治理迫在眉睫。如何高效地獲取土壤鹽分信息是鹽堿地治理的關鍵。傳統(tǒng)方法獲取土地鹽堿化分布及變化規(guī)律精度高但費時費力、易受地形限制[3],低空遙感為土壤鹽堿化反演提供科學、高效的技術手段。無人機搭載光譜儀器可快速獲取地面大范圍光譜信息,許多學者使用無人機搭載高光譜、多光譜等儀器,實現(xiàn)了土壤墑情監(jiān)測和作物生長監(jiān)測[4-5]。此外,紅邊波段可很好地反映綠色植被生長狀況,與表征植被長勢的重要參數(shù)間有較好的相關性[6-7]。
敏感光譜變量作為反演模型最重要的輸入數(shù)據(jù),其選擇方法一直是該研究領域的研究熱點。使用合適的變量選擇方法不僅可以提高模型預測精度[8],還可加快機器學習速率,孔鈺如等[9]采用連續(xù)投影算法、最佳指數(shù)法以及逐波段組合法分別進行無人機高光譜數(shù)據(jù)最佳波段篩選,結(jié)合支持向量回歸、偏最小二乘回歸和隨機森林回歸模型對冬小麥葉片LAI進行估算。Zhao等[10]使用相關系數(shù)法選取相關性高的光譜指數(shù)構建不同植被覆蓋表層土壤含鹽量反演模型,所構建模型等較好地反映真實土壤鹽分含量。楊寧等[11]通過ENET(Elastic Net)變量選擇算法構建不同深度土壤含鹽量反演模型,結(jié)果顯示ENET可以有效篩選出最優(yōu)光譜變量,且使用該方法建立的變量組構建模型精度相較于未使用變量選擇方法構建的反演模型精度有了明顯提高。近年來,支持向量機遞歸特征消除(Support Vector Machine-Recursive Feature Elimination,SVM-RFE)逐步被用于遙感領域,張睿等[12]通過對比在高光譜數(shù)據(jù)分類中SVM-RFE和OneR、Info Gain、Relief F 3種特征選擇方法對其精度影響,發(fā)現(xiàn)SVM-RFE選擇特征較優(yōu)。甄佳寧等[13]使用SVM-RFE算法選取了光譜波段及光譜指數(shù)組成的最優(yōu)變量組合,從而建立基于Sentinel-2影像的紅樹林冠層氮含量反演模型。在特征數(shù)目較少的情況下,SVM-RFE法可有效去除冗余特征,但將SVM-RFE結(jié)合機器學習算法構建作物物候期不同深度土壤鹽分反演模型的相關研究較少。
為此,本文基于無人機搭載多光譜成像系統(tǒng)獲取作物物候期遙感影像并對苜蓿和玉米覆蓋下不同深度土壤層鹽分含量進行同時段野外實測,通過引入紅邊波段構建出與實測土壤含鹽量相關性較高的光譜指數(shù)組,采用SVM-RFE算法對光譜指數(shù)進行篩選,將篩選前的全指數(shù)組和篩選后新指數(shù)組作為模型輸入層,構建基于隨機森林(Random Forest,RF)、極限學習機(Extreme Learning Machine,ELM)、反向神經(jīng)網(wǎng)絡(Back Propagation Neural Networks,BPNN)機器學習算法的鹽分反演模型,通過對所建36個反演模型的反演效果評估,以討賴河灌區(qū)邊灣農(nóng)場為例,確定出灌區(qū)不同作物覆蓋下適宜的土壤鹽分反演模型,以期為快速反演土壤鹽分提供理論依據(jù)。
邊灣農(nóng)場位于酒泉市肅州區(qū),隸屬于討賴河灌區(qū),農(nóng)場氣候干旱,降雨稀少,蒸發(fā)量大,海拔約1 390 m,農(nóng)場總面積約15.6 km2,其中耕地面積占34.7%,是討賴河灌區(qū)鹽堿地代表性區(qū)域,見圖1。苜蓿和玉米作為西北內(nèi)陸鹽堿地主要種植作物,將其作為主要作物覆蓋有著重要意義。本研究以邊灣農(nóng)場種植的苜蓿和玉米覆蓋土壤作為主要研究對象。

圖1 研究區(qū)示意圖
1.2.1 無人機多光譜遙感影像
使用DJI精靈4,搭載一體式多光譜成像系統(tǒng),該相機集成可見光以及紅光650 nm、綠光560 nm、藍光450 nm、近紅外840 nm以及紅邊730 nm共5個波段,2022年6月7日在酒泉市邊灣農(nóng)場進行影像實地采樣,天氣晴朗,無降雨。設置飛行高度40 m,航線重疊率70%,旁像重疊率65%,無人機平均速度4 m/s。采集好的多光譜影像導入大疆智圖進行圖像校正、拼接等。
1.2.2 野外實測采樣
2022年6月8至10日在酒泉市邊灣農(nóng)場進行野外實測采樣,采樣點分布見圖2。
此時該區(qū)苜蓿正處于盛花期,玉米處于拔節(jié)期,分別在苜蓿樣地均勻布設62個取樣點,玉米樣地布設56個取樣點,使用土鉆分別對取樣點0~15、>15~30以及>30~50 cm深度的土壤進行取樣,并記錄各點位置信息。每份樣品均稱量30 g放入鋁盒中,烘箱烘干8 h,放涼后進行研磨過篩(孔徑2 mm),篩好的土樣中加入150 mL蒸餾水并充分攪拌[14],靜止數(shù)小時后,使用雷磁DJS-1C測定土壤溶液電導率(EC1:5,mS/cm)。根據(jù)經(jīng)驗公式SSC=0.288 2EC1:5+0.018 3計算土壤鹽分含量(SSC,%)[15]。

圖2 采樣點分布圖
SVM-RFE算法是一種基于支持向量機的向后迭代遞歸特征變量選擇方法,主要是利用SVM算法對所研究特征變量進行排序,并評估每個特征變量的重要程度,按照向后迭代逐一剔除重要性低的變量[13],相較于相關性篩選法,該算法有效避免部分變量在進入模型前被過濾的情況,常用于遙感研究[16-18]。本研究將32個常見光譜指數(shù),其中包括16個鹽分指數(shù)和16個植被指數(shù),通過引入包含更加廣泛光譜信息的紅邊波段來替代紅光波段,與傳統(tǒng)光譜指數(shù)一起共計58個光譜指數(shù)參與篩選。
采用RF、ELM和BPNN共3種機器學習算法構建土壤鹽分反演模型。以苜蓿和玉米覆蓋土壤樣本的70%即苜蓿樣地43、玉米樣地39個樣本作為建模集,余下樣本作為驗證集。通過評估決定系數(shù)2、均方根誤差RMSE和標準均方根誤差nRMSE對模型建模集及驗證集進行精度評價。2介于0和1之間,越接近1說明模型精度越高,RMSE越接近0模型精度越高。nRMSE評價模型差異,當其小于10%時說明模型無差異,10%~20%間模型差異較小,20%~30%間模型差異一般,大于30%時,認為模型差異較大[19]。各參數(shù)計算公式如下:



試驗共采集0~15、>15~30、>30~50 cm土層的苜蓿樣地樣本點62個,玉米樣地樣本點56個,共計354個采樣點,所有樣本點的土壤含鹽量特性分析見表1。由表1可知,苜蓿樣地0~15 cm表層土壤樣本重度鹽化(土壤含鹽量在0.5%~1%)、中度鹽化(土壤含鹽量在0.2%~0.5%)、輕度鹽化(土壤含鹽量<0.2%)[15]的樣本分別占總樣本的1.6%、83.9%、14.5%;>15~30 cm土層中3個等級的樣本占比為0%、83.9%、16.1%;30~50 cm土層的樣本占比為0%、72.6%、27.4%。玉米樣地0~15 cm表層土壤中重度鹽化、中度鹽化、輕度鹽化樣本分別占總樣本的62.5%、37.5%、0%。>15~30 cm和>30~50 cm中3個等級的樣本占比均為51.8%、48.2%、0%。3個土層的樣本數(shù)量變異系數(shù)均小于0.4,變異程度小,最值均在合理范圍內(nèi),無異常值,可全部用于模型數(shù)據(jù)集。

表1 土壤含鹽量特性描述分析
本研究選取常見的16個傳統(tǒng)鹽分指數(shù),分別為NDSI、Int1、Int2、SI、SI1、SI2、SI3、SI-T、S1、S2、S3、S4、S5、S6、SR、BI和16個傳統(tǒng)植被指數(shù)NDVI、DVI、EVI、RVI、GCI、GSAVI、GRVI、GOSAVI、GNDVI、GLI、GDVI、LAI、IPVI、MSAVI、NNIP、NLI作為模型輸入,計算公式見表2。
通過研究引入紅邊波段對作物參數(shù)或土壤水鹽反演精度的影響,發(fā)現(xiàn)引入紅邊波段可有效提升反演模型精度[20-22]。本文通過引入紅邊波段,將以上傳統(tǒng)光譜指數(shù)中的紅光波段用紅邊波段替代(記為X-reg,如EVI-reg),從而衍生出26個改進光譜指數(shù),共構成58個光譜指數(shù)。

表2 光譜指數(shù)計算公式表
注:、NIR代表多光譜影像中紅光、綠光、藍光、近紅外波段的反射值。
Note:,,, NIR represent the reflection values of red band, green band, blue band and near-infrared band in multispectral images.
采用SVM-RFE算法對原光譜指數(shù)和引入紅變波段后共58個光譜指數(shù)進行篩選(該算法用MATLAB R2018a實現(xiàn)),通過逐次增加篩選個數(shù),發(fā)現(xiàn)設置保留特征為14個時,反演模型精度最好,為避免篩選特征數(shù)量對模型精度的影響,對每層土壤均篩選14個光譜指數(shù)作為模型的輸入量。SVM-RFE使用SVM機器學習模型精度作為度量對特征進行排序,去掉最小特征得分的特征,用剩余特征再次訓練模型,進行下一次迭代,直到?jīng)]有特征,排在前面的單個特征并不一定是最優(yōu)特征子集,而是特征變量組合在一起才使得模型達到最優(yōu)學習效果,通過不停迭代,每次減少一個保留特征數(shù)量,得到重要程度排序表3。
從表3可知,苜蓿樣地篩選結(jié)果多為植被指數(shù),且引入紅邊波段計算出的改進光譜指數(shù)占比較大,而玉米樣地多為鹽分指數(shù)。對篩選后的光譜指數(shù)組進行相關性檢驗(圖3),發(fā)現(xiàn)大部分光譜指數(shù)與實測值相關性顯著,可用于后續(xù)建模。
2.3.1 全指數(shù)組模型反演結(jié)果分析
為探究SVM-RFE在光譜指數(shù)篩選中的適用性,用SVM-RFE算法篩選光譜指數(shù)建立模型與未篩選光譜指數(shù)組建立模型進行對比,全指數(shù)組模型反演結(jié)果見表 4,可知將紅邊波段引入不同作物覆蓋下不同深度土壤鹽分反演模型的構建中是可行的。全指數(shù)組構建模型中,苜蓿和玉米覆蓋地0~15 cm土壤層反演效果為最優(yōu),其他2個深度層反演效果接近,苜蓿樣地RF模型驗證集p2在0.50~0.57間,RMSEp在0.030~0.036間,玉米樣地RF模型驗證集p2在0.60~0.62間,RMSEp在0.044~0.102間,nRMSEc和nRMSEp均小于20%,模型差異小,RF模型表現(xiàn)能力突出,其次是ELM模型。

表3 基于SVM-RFE篩選的光譜指數(shù)重要性排序

圖3 光譜指數(shù)與實測土壤含鹽量相關系數(shù)圖
2.3.2 篩選后指數(shù)組模型反演分析
用SVM-RFE算法篩選光譜指數(shù)后的14個光譜指數(shù)建立模型,對比表4和表5,經(jīng)過指數(shù)篩選構建的模型精度優(yōu)于未進行指數(shù)篩選的全指數(shù)構建的模型。由表5和圖4a可知,對于苜蓿樣地,從深度看,0~15、>15~30 和>30~50 cm土層構建的反演模型p2分別在0.61~0.71、0.43~0.58、0.54~0.58之間,RMSEp分別在0.026~0.69、0.030~0.54、0.032~0.39間,0~15 cm土壤鹽分反演明顯優(yōu)于其他深度,其次是>30~50 cm。從模型看,對于0~15和>30~50 cm土層鹽分反演模型,RF反演效果優(yōu)于ELM和BPNN,p2分別為0.71、0.58,RMSEp分別為0.026、0.033,其次是ELM;對于>15~30 cm土層,ELM反演精度最高,p2為0.58,RMSEp為0.039,其次是RF模型。綜上,苜蓿覆蓋下土壤鹽分反演最佳深度是0~15 cm,其次是>30~50 cm,且RF整體表現(xiàn)最優(yōu),如圖 5所示。

表4 基于全指數(shù)組的不同深度土壤鹽分含量反演結(jié)果
注:c2、RMSEc、nRMSEc分別為建模集的決定系數(shù)、均方根誤差和標準均方根誤差,p2、RMSEp、nRMSEp分別為驗證集的決定系數(shù)、均方根誤差和標準均方根誤差。下同。
Note:c2、RMSEcand nRMSEcare the determination coefficient, root mean square error and standard root mean square error of the modeling set, andp2、RMSEpand nRMSEpare the determination coefficient, root mean square error and standard root mean square error of the validation set. The same below.

表5 基于SVM-RFE篩選的不同深度土壤鹽分反演模型

圖4 基于SVM-RFE變量篩選的機器學習模型土壤含鹽量實測值與預測值對比

圖5 苜蓿樣地土壤含鹽量實測值與預測值對比
由表5和圖4b可知,對于玉米樣地0~15 、>15~30 和>30~50 cm土層反演模型p2分別在0.60~0.67、0.52~0.68、0.53~0.64之間,RMSEp分別在0.110~0.134、0.059~0.87、0.044~0.122間,0~15 cm、>15~30 cm反演效果優(yōu)于>30~50 cm。從模型看,玉米與苜蓿樣地反演結(jié)果一致,在0~15 和30~50 cm土層,RF反演效果最優(yōu),p2分別為0.67、0.64,RMSEp分別為0.111、0.094,其次是ELM;對于>15~30 cm土壤,ELM反演精度最高,p2為0.68,RMSEp為0.059,其次是RF模型。綜上,玉米覆蓋下土壤鹽分反演最佳深度是0~15 cm,其次是15~30 cm,RF總體反演效果依然是最佳(如圖6所示)。且0~15 cm是作物覆蓋下的土壤含鹽量最佳反演深度,苜蓿、玉米樣地平均p2為0.65,平均RMSE為0.084。

圖6 玉米樣地土壤含鹽量實測值與預測值對比
作物冠層光譜反射率敏感響應土壤鹽分含量,可利用植被指數(shù)和鹽分指數(shù)等建立與實測土壤含鹽量的關系,從而實現(xiàn)土壤鹽分反演[26]。本文選取苜蓿和玉米覆蓋土壤作為研究對象,研究不同植被覆蓋下不同深度土壤層含鹽量的反演,以無人機搭載多光譜相機為載體,構建了多作物、多深度、多模型的土壤鹽分反演模型,實現(xiàn)農(nóng)耕區(qū)不同深度土壤含鹽量的快速獲取。
通過SVM-RFE變量選擇方法篩選出的苜蓿和玉米光譜指數(shù)有較明顯差異,苜蓿樣地篩選后光譜指數(shù)大部分為植被指數(shù),玉米樣地篩選后光譜指數(shù)大部分為鹽分指數(shù),主要原因是作物光譜特征因為作物類型不同、植被覆蓋度差異較大表現(xiàn)出明顯區(qū)別(采樣時苜蓿覆蓋程度高,玉米覆蓋程度低,大面積的土壤處于裸露狀態(tài)),故由光譜波段構建的光譜指數(shù)也有顯著差異[10,27]。張智韜等[28]在研究植被覆蓋程度對土壤含鹽量反演的影響時,證實了這一點,隨著植被覆蓋度增加,鹽分指數(shù)敏感性逐漸降低,植被指數(shù)敏感性逐漸增加。
Wang等[29]以庫車綠洲為研究區(qū)研究土壤鹽分反演算法比較時發(fā)現(xiàn),在0~10 、>10~30 和>30~50 cm三個土層深度中構建的13個鹽分反演模型2最高的是0~10 cm層(0.60~0.74),其次是>30~50 cm(0.30~0.47),最低的是10~30 cm(0.15~0.31),這與本研究苜蓿覆蓋下土壤鹽分反演結(jié)果一致。而對于玉米,反演模型效果最好、精度最高的是0~10 cm層,其次是>15~30 cm,最低的是>30~50 cm,這種差異應該是由于作物不同、覆蓋度不同以及覆膜產(chǎn)生的[28,30]。
本研究得出,植被覆蓋下土壤鹽分反演模型在0~15 和30~50 cm土層RF模型表現(xiàn)最好,在15~30 cm土壤層ELM反演效果略優(yōu)于RF,許多相關研究也發(fā)現(xiàn)ELM模型反演精度高[11,30]。綜合評估構建模型表現(xiàn)最佳的是RF模型,2種作物0~15 cm土層的RF模型驗證集2和RMSE平均值分別為0.69、0.069,>15~30 cm土層的RF模型驗證集2和RMSE平均值為0.59、0.048,>30~50 cm土層的RF模型驗證集2和RMSE平均值為0.61、0.039。很多學者在做相似反演模型比較時,均認為基于RF算法構建的模型反演精度高,穩(wěn)定性好,抗過擬合能力強[31-32]。Wei等[33]在進行土壤鹽漬化監(jiān)測模型研究時同樣得出在基于RF、BPNN和SVR算法的植被覆蓋條件下的含鹽量反演機器學習模型中,基于光譜指數(shù)組的RF反演模型測試集2為0.67,RMSE為0.112,基于RF算法模型表現(xiàn)最佳。杜瑞麒等[34]在研究基于Sentinel-2多光譜衛(wèi)星反演植被覆蓋下的土壤鹽分時同樣得出了類似結(jié)論。故相比于ELM和BPNN,基于RF算法構建的土壤鹽分反演模型更能真實的表達討賴河流域作物覆蓋下土壤的含鹽量。
本文主要研究了基于無人機多光譜影像的不同作物覆蓋下不同土壤深度鹽分反演模型并取得了較好的反演結(jié)果,但多源遙感及環(huán)境因素對模型的影響有待進一步研究,在今后的學習中,可考慮與衛(wèi)星遙感相結(jié)合,同時將氣溫、土壤濕度等參數(shù)加入到模型構建中,以期確定出精準性更高、適用性更強的土壤鹽分反演模型。
本文采集了苜蓿、玉米覆蓋下不同深度土壤含鹽量,以全指數(shù)組和用SVM-RFE篩選光譜指數(shù)組結(jié)合RF、ELM和BPNN 3種機器學習算法,建立36個作物覆蓋下土壤鹽分反演模型,并以決定系數(shù)2、均方根誤差RMSE、標準均方根誤差nRMSE評估各模型反演效果,得出以下結(jié)論:
1)對比全指數(shù)組構建模型,通過SVM-RFE變量選擇方法篩選出的光譜指數(shù)組作為模型輸入數(shù)據(jù),所構建的鹽分反演模型反演效果更好,均方根誤差更小,訓練速度也明顯提高,表明該變量選擇方法用于不同深度層土壤鹽分反演及相關研究中來是可行的。
2)對于基于SVM-RFE變量篩選的苜蓿覆蓋下的土壤,0~15 cm反演效果最好,其次是>30~50 cm,最后是>15~30 cm,驗證集2依次為在0.61~0.71、0.43~0.58、0.54~0.58之間。玉米覆蓋下的土壤在0~15 cm土層同樣反演效果最好,其次是>15~30 cm,最后是>30~50 cm,驗證集2依次在0.60~0.67、0.52~0.68、0.53~0.64間。故作物覆蓋下的土壤含鹽量的最佳反演深度是0~15 cm,反演結(jié)果最能體現(xiàn)真實土壤鹽分值。
3)在0~15和>30~50 cm土層RF模型的反演效果最好,苜蓿樣地驗證集2分別為0.71、0.58,玉米樣地驗證集2分別為0.67、0.64。>15~30 cm土層ELM模型的反演效果最好,苜蓿樣地驗證集2為0.58,玉米樣地驗證集2為0.68。整體看,苜蓿和玉米樣地所建反演模型中RF模型總體反演效果優(yōu)于ELM模型,BPNN模型表現(xiàn)一般。
[1] 王佳麗,黃賢金,鐘太洋,等. 鹽堿地可持續(xù)利用研究綜述[J]. 地理學報,2011,66(5):673-684.
Wang Jiali, Huang Xianjin, Zhong Taiyang, et al. Review on sustainable utilization of saline-alkali land[J]. Journal of Geography, 2011, 66(5): 673-684. (in Chinese with English abstract)
[2] Aldabaa A A A, Weindorf D C, Chakraborty S, et al. Combination of proximal and remote sensing methods for rapid soil salinity quantification[J]. Geoderma, 2015, 239: 34-46.
[3] Ding J, Yu D. Monitoring and evaluating spatial variability of soil salinity in dry and wet seasons in the Werigan-Kuqa Oasis, China, using remote sensing and electromagnetic induction instruments[J]. Geoderma, 2014, 235: 316-322.
[4] Ge X, Wang J, Ding J, et al. Combining UAV-based hyperspectral imagery and machine learning algorithms for soil moisture content monitoring[J]. Peer J, 2019, 7: e6926.
[5] 王翔宇,楊菡,李鑫星,等. 基于無人機可見光譜遙感的玉米長勢監(jiān)測[J]. 光譜學與光譜分析,2021,41(1):265-270.
Wang Xiangyu, Yang Han, Li Xinxing, et al. Maize growth monitoring based on UAV visible spectrum remote sensing[J]. Spectroscopy and Spectral Analysis, 2021, 41(1): 265-270. (in Chinese with English abstract)
[6] Kaplan G, Avdan U. Evaluating the utilization of the red edge and radar bands from sentinel sensors for wetland classification[J]. Catena, 2019, 178: 109-119.
[7] 陳鵬飛,Nicolas Tremblay,王紀華,等. 估測作物冠層生物量的新植被指數(shù)的研究[J]. 光譜學與光譜分析,2010,30(2):512-517.
Chen Pengfei, Nicolas Tremblay, Wang Jihua, et al. A study on the new vegetation index for estimating crop canopy biomass[J]. Spectroscopy and Spectral Analysis, 2010, 30(2): 512-517. (in Chinese with English abstract)
[8] Hong Y, Chen Y, Yu L, et al. Combining fractional order derivative and spectral variable selection for organic matter estimation of homogeneous soil samples by VIS-NIR spectroscopy[J]. Remote Sensing, 2018, 10(3): 479.
[9] 孔鈺如,王李娟,馮海寬,等. 無人機高光譜波段選擇的葉面積指數(shù)反演[J]. 光譜學與光譜分析,2022,42(3):933-939.
Kong Yuru, Wang Lijuan, Feng Haikuan, et al. Leaf area index inversion of UAV hyperspectral band selection[J]. Spectroscopy and Spectral Analysis, 2022, 42(3): 933-939. (in Chinese with English abstract)
[10] Zhao W, Zhou C, Zhou C, et al. Soil salinity inversion model of oasis in arid area based on UAV multispectral remote sensing[J]. Remote Sensing, 2022, 14(8): 1-13.
[11] 楊寧,崔文軒,張智韜,等. 無人機多光譜遙感反演不同深度土壤鹽分[J]. 農(nóng)業(yè)工程學報,2020,36(22):13-21.
Yang Ning, Cui Wenxuan, Zhang Zhitao, et al. Inversion of soil salinity at different depths by UAV multispectral remote sensing[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2020, 36(22): 13-21. (in Chinese with English abstract)
[12] 張睿,馬建文. 一種SVM-RFE高光譜數(shù)據(jù)特征選擇算法[J]. 武漢大學學報(信息科學版),2009,34(7):834-837.
Zhang Rui, Ma Jianwen. SVM-RFE feature selection algorithm for hyperspectral data[J]. Journal of Wuhan University (Information Science Edition), 2009, 34(7): 834-837. (in Chinese with English abstract)
[13] 甄佳寧,蔣俠朋,趙德梅,等. 利用Sentinel-2影像超分辨率重建的紅樹林冠層氮含量反演[J]. 遙感學報,2022,26(6):1206-1219.
Zhen Jianing, Jiang Xiapeng, Zhao Demei, et al. Inversion of mangrove canopy nitrogen content using Sentinel-2 image super-resolution reconstruction[J]. Journal of Remote Sensing, 2022, 26(6): 1206-1219. (in Chinese with English abstract)
[14] Zhang F, Tashpolat T, Ding J L, et al. Relationships between soil salinization and spectra in the delta oasis of Weigan and Kuqa Rivers[J]. Research of Environmental Sciences, 2009, 22(2): 227-235.
[15] 黃權中,徐旭,呂玲嬌,等. 基于遙感反演河套灌區(qū)土壤鹽分分布及對作物生長的影響[J]. 農(nóng)業(yè)工程學報,2018,34(1):102-109.
Huang Quanzhong, Xu Xu, Lyu Lingjiao, et al. Soil salinity distribution based on remote sensing and its effect on crop growth in Hetao Irrigation District[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2018, 34(1): 102-109. (in Chinese with English abstract)
[16] Hankui Z, Bo H, Le Y. Kernel function in SVM-RFE based hyperspectral data band selection[J]. Remote Sensing Technology and Application, 2014, 28(5): 747-752.
[17] Zhang R, Ma J. Feature selection for hyperspectral data based on recursive support vector machines[J]. International Journal of Remote Sensing, 2009, 30(14): 3669-3677.
[18] Chen L, Xing M, He B, et al. Estimating soil moisture over winter wheat fields during growing season using machine-learning methods[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14: 3706-3718.
[19] 譚丞軒,張智韜,許崇豪,等. 無人機多光譜遙感反演各生育期玉米根域土壤含水率[J]. 農(nóng)業(yè)工程學報,2020,36(10):63-74.
Tan Chengxuan, Zhang Zhitao, Xu Chonghao, et al. Soil water content inversion model in field maize root zone based on UAV multispectral remote sensing[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2020, 36(10): 63-74. (in Chinese with English abstract)
[20] Kaplan G, Avdan U. Evaluating the utilization of the red edge and radar bands from sentinel sensors for wetland classification[J]. Catena, 2019, 178: 109-119.
[21] Wang J, Ding J, Yu D, et al. Capability of Sentinel-2 MSI data for monitoring and mapping of soil salinity in dry and wet seasons in the Ebinur Lake region, Xinjiang, China[J]. Geoderma, 2019, 353: 172-187.
[22] Zheng Q, Huang W, Cui X, et al. New spectral index for detecting wheat yellow rust using Sentinel-2 multispectral imagery[J]. Sensors, 2018, 18(3): 868.
[23] Khan N M, Rastoskuev V V, Sato Y, et al. Assessment of hydrosaline land degradation by using a simple approach of remote sensing indicators[J]. Agricultural Water Management, 2005, 77(1/2/3): 96-109.
[24] Allbed A, Kumar L, Aldakheel Y Y. Assessing soil salinity using soil salinity and vegetation indices derived from IKONOS high-spatial resolution imageries: Applications in a date palm dominated region[J]. Geoderma, 2014, 230: 1-8.
[25] 蘇偉,王偉,劉哲,等.無人機影像反演玉米冠層LAI和葉綠素含量的參數(shù)確定[J].農(nóng)業(yè)工程學報,2020,36(19):58-65.
Su Wei, Wang Wei, Liu Zhe, et al. Determination of parameters for inversion of maize canopy LAI and chlorophyll content from UAV images[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2020, 36(19): 58-65. (in Chinese with English abstract)
[26] 張智韜,魏廣飛,姚志華,等. 基于無人機多光譜遙感的土壤含鹽量反演模型研究[J]. 農(nóng)業(yè)機械學報,2019,50(12):151-160.
Zhang Zhitao, Wei Guangfei, Yao Zhihua et al. Research on inversion model of soil salt content based on UAV multispectral remote sensing[J]. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(12): 151-160. (in Chinese with English abstract)
[27] 陳俊英,姚志華,張智韜,等. 大田葵花土壤含鹽量無人機遙感反演研究[J]. 農(nóng)業(yè)機械學報,2020,51(7):178-191.
Chen Junying, Yao Zhihua, Zhang Zhitao, et al. Research on soil salinity of sunflower in field by UAV remote sensing inversion[J]. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(7): 178-191. (in Chinese with English abstract)
[28] 張智韜,臺翔,楊寧,等. 不同植被覆蓋度下無人機多光譜遙感土壤含鹽量反演[J]. 農(nóng)業(yè)機械學報,2022,53(8):220-230.
Zhang Zhitao, Tai Xiang, Yang Ning, et al. Inversion of soil salt content by UAV multispectral remote sensing under different vegetation coverage[J]. Transactions of the Chinese Society for Agricultural Machinery, 2022, 53(8): 220-230. (in Chinese with English abstract)
[29] Wang F, Shi Z, Biswas A, et al. Multi-algorithm comparison for predicting soil salinity[J]. Geoderma, 2020, 365: 114211.
[30] 姚志華,陳俊英,張智韜,等. 覆膜對無人機多光譜遙感反演土壤含鹽量精度的影響[J]. 農(nóng)業(yè)工程學報,2019,35(19):89-97.
Yao Zhihua, Chen Junying, Zhang Zhitao, et al. Effect of film mulching on inversion accuracy of soil salinity by UAV multispectral remote sensing[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2019, 35(19): 89-97. (in Chinese with English abstract)
[31] Chen X W, Liu M. Prediction of protein-protein interactions using random decision forest framework[J]. Bioinformatics, 2005, 21(24): 4394-4400.
[32] Lindner C, Bromiley P A, Ionita M C, et al. Robust and accurate shape model matching using random forest regression-voting[J]. IEEE transactions on pattern analysis and machine intelligence, 2014, 37(9): 1862-1874.
[33] Wei G, Li Y, Zhang Z, et al. Estimation of soil salt content by combining UAV-borne multispectral sensor and machine learning algorithms[J]. PeerJ, 2020, 8(2): e9087.
[34] 杜瑞麒,陳俊英,張智韜,等. Sentinel-2多光譜衛(wèi)星遙感反演植被覆蓋下的土壤鹽分變化[J]. 農(nóng)業(yè)工程學報,2021,37(17):107-115.
Du Ruiqi, Chen Junying, Zhang Zhitao, et al. Sentinel-2 multispectral satellite remote sensing inversion of soil salt changes under vegetation cover[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2021, 37(17): 107-115. (in Chinese with English abstract)
Soil salinity inversion model based on the multispectral images of UAV
Zhao Wenju, Ma Fangfang, Ma Hong, Zhou Chun
(,,730050,)
Soil salinization has posed a serious threat to the growth and yield of crops in the national food security. Among them, the Taolai River basin with the widely distributed saline-alkali land has been one of the most important planting areas in northwest China. It is a high demand for the timely acquisition of soil salinity information during the salinization control. In this study, a representative sampling area of soil salinization was taken as the Bianwan Farm in Suzhou District, Jiuquan City, Gansu Province, China. A rapid inversion model of soil salinity was proposed at the soil depths of 0-15, 15-30, and 30-50 cm under the crop cover of alfalfa and corn in the phenological period. The multi-spectral image data of the Unmanned Aerial Vehicle (UAV) was also collected at the same time. The reflectance of the spectral band was extracted in the different acquisition points of plots. The red edge band was also introduced to calculate the spectral index, in order to effectively improve the inversion accuracy. A total of 58 spectral indices were involved in the modeling. The Support Vector Machine-Recursive Feature Elimination (SVM-RFE) was selected to screen the spectral index. Specifically, the SVM was used to sort the feature variables, and then evaluate the importance of each feature variable. The variables with low importance were removed, according to the backward iteration. As such, a better performance was achieved to effectively remove the redundant features for the high running speed of the model. A total of 36 models were constructed to evaluate the accuracy and inversion effect of the models, including Random Forest (RF), Extreme learning machine (ELM), and Back-propagation neural network (BPNN). The model input was taken as the unfiltered full and filtered new variable group. Finally, the best soil-salinity inversion model was determined for the optimal inversion depth under crop coverage. The results show that the SVM-RFE variable selection significantly improved the accuracy of each soil-salinity inversion model. A better performance was achieved in the coefficient of determination (2), root-mean-square error (RMSE), and training speed of the improved model, compared with the model without variable screening. Overall, the inversion effect of the RF model was better than that of ELM and BPNN models. Among them, the inversion effect was one of the best indicators for real soil salt. Specifically, the RF model presented the best inversion effect in the 0-15 and 30-50 cm soil layers under crop cover, where thep2values of the validation set in the alfalfa field were 0.71 and 0.58, respectively, RMSEpvalues were 0.026 and 0.033, respectively; thep2values in the corn field were 0.67 and 0.64, respectively, the RMSEpvalues were 0.111 and 0.094, respectively. In the 15-30 cm layer, the ELM model presented the best inversion effect, where thep2values in the alfalfa and corn fields were 0.58, and 0.039, respectively; the RMSEpvalues were 0.68, and 0.059, respectively. In the terms of inversion depth, the inversion effects of 0-15 cm and 30-50 cm were better than that of 15-30 cm for the alfalfa-covered soil. The inversion effects of 0-15 cm and 15-30 cm were better than that of 30-50 cm for the corn-covered soil. The comprehensive analysis showed that the 0-15 cm soil layer was the best inversion depth for the soil salt content under crop cover, where the average2of the validation set was 0.65, and the RMSE was 0.084. A strong reference was offered to manage the saline-alkali land in the arid area of northwest China. The finding can also provide a scientific basis for the rapid inversion of salt in the different soil depth layers under crop mulching.
UAV; soils; salinity; multispectral; SVM-RFE; inversion model
10.11975/j.issn.1002-6819.2022.24.010
S127
A
1002-6819(2022)-24-0093-09
趙文舉,馬芳芳,馬宏,等. 基于無人機多光譜影像的土壤鹽分反演模型[J]. 農(nóng)業(yè)工程學報,2022,38(24):93-101.doi:10.11975/j.issn.1002-6819.2022.24.010 http://www.tcsae.org
Zhao Wenju, Ma Fangfang, Ma Hong, et al. Soil salinity inversion model based on the multispectral images of UAV[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2022, 38(24): 93-101. (in Chinese with English abstract) doi:10.11975/j.issn.1002-6819.2022.24.010 http://www.tcsae.org
2022-09-12
2022-11-08
國家自然科學基金項目(51869010)
趙文舉,博士,教授,博士生導師,研究方向為寒旱區(qū)生態(tài)水利。Email:wenjuzhao@126.com