龍林麗,劉 英,2,張旭陽(yáng),蘇永東,陳孝楊,2
無(wú)人機(jī)在礦區(qū)表土特征及地質(zhì)災(zāi)害監(jiān)測(cè)中的應(yīng)用
龍林麗1,劉 英1,2,張旭陽(yáng)1,蘇永東1,陳孝楊1,2
(1. 安徽理工大學(xué) 地球與環(huán)境學(xué)院,安徽 淮南 232001;2. 安徽省高潛水位礦區(qū)水土資源高效綜合利用工程實(shí)驗(yàn)室,安徽 淮南 232001)
隨著無(wú)人機(jī)的出現(xiàn)和發(fā)展,各種傳感器的小型化和智能化程度不斷提高,裝載傳感器的無(wú)人機(jī)成為獲得空間數(shù)據(jù)的高效工具。因其成本低、重訪周期短、快速高效、質(zhì)輕靈活、操作簡(jiǎn)便、影像獲取時(shí)空精度高等特點(diǎn),廣泛應(yīng)用于礦區(qū)土地?fù)p傷監(jiān)測(cè)。以“無(wú)人機(jī)(UAV)、反演(Inversion)、土壤監(jiān)測(cè)(Soil Monitoring)、地表塌陷(Surface Collapse)、地裂縫(Ground Fissure)”為關(guān)鍵詞,通過(guò)總結(jié)Web of Science、知網(wǎng)、谷歌學(xué)術(shù)等檢索系統(tǒng)中2010年1月—2021年5月發(fā)表的學(xué)術(shù)論文,對(duì)比分析無(wú)人機(jī)監(jiān)測(cè)技術(shù)與其他監(jiān)測(cè)技術(shù)的差別,綜述無(wú)人機(jī)監(jiān)測(cè)礦區(qū)重金屬、土壤含水率、含鹽量、地表塌陷、地裂縫及邊坡穩(wěn)定性的一般流程及數(shù)據(jù)處理方法,并概述無(wú)人機(jī)在礦區(qū)表土特征及地質(zhì)災(zāi)害監(jiān)測(cè)中的應(yīng)用前景,認(rèn)為未來(lái)可通過(guò)集成野外時(shí)序跟蹤調(diào)查、高精度土壤質(zhì)量監(jiān)測(cè)技術(shù)、高空間分辨率無(wú)人機(jī)監(jiān)測(cè)技術(shù)、數(shù)字模擬手段和典型工作面的試驗(yàn)監(jiān)測(cè)與分析,研究工作面自開(kāi)切眼至停采線動(dòng)態(tài)推進(jìn)中地質(zhì)災(zāi)害與土壤質(zhì)量演化耦合關(guān)系,構(gòu)建采煤沉陷區(qū)土壤質(zhì)量演化預(yù)測(cè)理論體系和時(shí)序演變模型。從而進(jìn)一步探討礦區(qū)土壤質(zhì)量與地質(zhì)災(zāi)害之間的關(guān)系,提出減緩、控制及提升礦區(qū)土壤質(zhì)量的措施,為我國(guó)煤炭生產(chǎn)基地煤炭資源開(kāi)采與生態(tài)環(huán)境的協(xié)調(diào)可持續(xù)發(fā)展提供技術(shù)支撐。
無(wú)人機(jī);采煤礦區(qū);土壤監(jiān)測(cè);地質(zhì)災(zāi)害監(jiān)測(cè)
礦產(chǎn)資源在國(guó)民經(jīng)濟(jì)發(fā)展中占有重要地位[1],隨著工業(yè)的快速發(fā)展,人們對(duì)礦產(chǎn)資源的需求量顯著增加[2]。通常礦物的開(kāi)采方式有露天開(kāi)采和地下開(kāi)采2種,這2種方式在開(kāi)采過(guò)程中都會(huì)影響礦區(qū)生態(tài)環(huán)境,如地下開(kāi)采會(huì)引起地表沉陷,形成大量地裂縫,破壞耕地,加劇水土流失、破壞植被等。露天開(kāi)采產(chǎn)生的矸石、尾礦固體廢物及直接剝離表層土壤、形成排土場(chǎng)壓占土地等都會(huì)影響土地可持續(xù)利用[3]。因此,在采礦過(guò)程中,利用現(xiàn)代高新技術(shù)對(duì)礦區(qū)生態(tài)環(huán)境進(jìn)行實(shí)時(shí)監(jiān)測(cè)是合理規(guī)劃可持續(xù)開(kāi)采的關(guān)鍵。
礦區(qū)常用的監(jiān)測(cè)技術(shù)包括野外現(xiàn)場(chǎng)監(jiān)測(cè)技術(shù)、全球?qū)Ш叫l(wèi)星技術(shù)(GNSS)、合成孔徑雷達(dá)干涉測(cè)量技術(shù)(InSAR)、三維激光掃描技術(shù)、無(wú)人機(jī)航測(cè)技術(shù)[4]等。傳統(tǒng)的野外現(xiàn)場(chǎng)監(jiān)測(cè)的一般步驟都是野外考察,布設(shè)監(jiān)測(cè)點(diǎn),儀器監(jiān)測(cè),最后通過(guò)周期性的監(jiān)測(cè)得到結(jié)果,其數(shù)據(jù)收集工作強(qiáng)度大,數(shù)據(jù)時(shí)效性差。GNSS雖然具有精度高、布網(wǎng)迅速、可全天候工作等特點(diǎn),但易受衛(wèi)星信號(hào)差的影響[5]。InSAR具有空間分辨率高、全天候、監(jiān)測(cè)范圍大等優(yōu)點(diǎn),然而受大氣延遲的影響較大[6]。三維激光掃描技術(shù)存在的主要問(wèn)題是數(shù)據(jù)質(zhì)量不佳,點(diǎn)云數(shù)據(jù)匹配精度不高[7]等。與其他監(jiān)測(cè)方法相比,無(wú)人機(jī)具有操作簡(jiǎn)單、高效快速、靈活方便、分辨率高、數(shù)據(jù)時(shí)效性強(qiáng)[8]等特點(diǎn),在低空飛行時(shí)可快速獲取到厘米級(jí)圖像[9]。無(wú)人機(jī)航測(cè)作為動(dòng)態(tài)、經(jīng)濟(jì)、連續(xù)的一種數(shù)據(jù)采集方法,可對(duì)采后風(fēng)險(xiǎn)源進(jìn)行持續(xù)監(jiān)測(cè)[10],在礦區(qū)生態(tài)環(huán)境監(jiān)測(cè)中有巨大的應(yīng)用前景。目前,基于無(wú)人機(jī)數(shù)據(jù)的科學(xué)研究已經(jīng)取得一定成果,但無(wú)人機(jī)在礦區(qū)生態(tài)監(jiān)測(cè)中的應(yīng)用拓展有待進(jìn)一步研究。
筆者利用“無(wú)人機(jī)(UAV)、反演(Inversion)、土壤監(jiān)測(cè)(Soil Monitoring)、地表塌陷(Surface Collapse)、地裂縫(Ground Fissure)”等檢索詞,在Web of Science、知網(wǎng)、谷歌學(xué)術(shù)中檢索近十年的文獻(xiàn),通過(guò)無(wú)人機(jī)在礦區(qū)監(jiān)測(cè)目的不同進(jìn)行分類,對(duì)其進(jìn)行綜合分析,綜述了無(wú)人機(jī)在礦區(qū)表土特征監(jiān)測(cè)、地質(zhì)災(zāi)害監(jiān)測(cè)方面的應(yīng)用,以期為無(wú)人機(jī)在礦區(qū)生態(tài)環(huán)境監(jiān)測(cè)中的應(yīng)用與推廣提供借鑒與參考。
1) 無(wú)人機(jī)監(jiān)測(cè)表土重金屬及數(shù)據(jù)預(yù)處理方法
煤炭資源的開(kāi)發(fā)、生產(chǎn)及運(yùn)輸過(guò)程中礦區(qū)污水的排放、酸性廢水的淋濾、運(yùn)輸中的灑落礦物和煤矸石的堆放等,在風(fēng)蝕、雨蝕、沉降等作用導(dǎo)致礦區(qū)及周邊土壤中重金屬大量富集,對(duì)土壤造成污染,而重金屬不易被土壤微生物降解,卻可以被動(dòng)植物富集,某些重金屬還能轉(zhuǎn)化為毒性更嚴(yán)重的甲基化合物,通過(guò)食物鏈在人體內(nèi)蓄積,嚴(yán)重影響人體健康[11]。因此,監(jiān)測(cè)礦區(qū)土壤重金屬元素是礦區(qū)生態(tài)環(huán)境監(jiān)測(cè)的重要一環(huán)。目前研究重金屬對(duì)土壤的污染的方法主要包括現(xiàn)場(chǎng)采樣法和遙感監(jiān)測(cè)法。傳統(tǒng)的現(xiàn)場(chǎng)調(diào)查可從特定的時(shí)刻和位置提供有限的信息,無(wú)法提供重金屬濃度分布時(shí)空動(dòng)態(tài)變化[12],而無(wú)人機(jī)遙感的反射光譜法通過(guò)電磁輻射反演可快速獲取土壤重金屬含量時(shí)空分布特征,是當(dāng)前熱門且主流的方法之一,一般可分為直接反演和間接反演,直接反演通常是利用無(wú)人機(jī)獲取的光譜信息與實(shí)測(cè)重金屬的相關(guān)性進(jìn)行,監(jiān)測(cè)流程是光譜數(shù)據(jù)獲取、光譜預(yù)處理、提取特征波段、研究處理后的光譜與實(shí)測(cè)重金屬濃度的相關(guān)性、構(gòu)建反演模型;間接反演則是根據(jù)不同土壤組分之間的相關(guān)關(guān)系,構(gòu)建間接反演模型。
光譜數(shù)據(jù)在監(jiān)測(cè)時(shí)會(huì)受到試驗(yàn)條件與土壤自身?xiàng)l件的影響,需要對(duì)光譜數(shù)據(jù)進(jìn)行預(yù)處理,從而有效提升光譜信息[13]。目前常用的預(yù)處理方法包括平滑處理[14]、包絡(luò)線去除(CR)[15]、一階導(dǎo)數(shù)(FD)[16]、微分校正(DR)[17]、連續(xù)小波變換(CWT)[17]、標(biāo)準(zhǔn)正態(tài)變量校正(SNV)[18]、多元散射校正(MSC)[19]等諸多方法。高光譜反演重金屬含量的常用光譜波段是400~2 500 nm[20],波段范圍廣,而且土壤光譜反射率不高。為了簡(jiǎn)化模型和提高反演精度,更好地解釋土壤重金屬與光譜數(shù)據(jù)的相關(guān)性,需要對(duì)光譜數(shù)據(jù)進(jìn)行特征分析,分為特征選擇、特征提取和特征增強(qiáng)[21]。特征選擇主要是從特征中選擇起主要作用的子集,方法有Pearson相關(guān)性分析、競(jìng)爭(zhēng)性自適應(yīng)重加權(quán)算法(CARS)、封裝法、嵌入法等[22-24]。特征提取是在原來(lái)的特征中,產(chǎn)生一個(gè)新子集,也可以起空間降維作用,有主成分分析法(PCA)[25]、最小噪聲分離(MNF)[25]、小波變換(WT)[26]等。特征增強(qiáng),主要是利用已有變量通過(guò)綜合分析而塑造出新變量,通常和預(yù)處理一起,特征提取后的光譜能夠更好地捕獲地面重金屬濃度的信息。
利用以上光譜數(shù)據(jù)預(yù)處理方法,突出光譜特征,進(jìn)而能求取與重金屬最大相關(guān)波段,以得到土壤光譜中能表現(xiàn)重金屬元素的敏感波段,如Zn的最大相關(guān)波段為515 nm[27],Cr、As、Cu的敏感波段分別是379、1 778、2 018 nm[28]。在土壤重金屬反演研究中,單一敏感波段數(shù)據(jù)所得到的特征表達(dá)能力有限;通過(guò)對(duì)多波段線性地組合會(huì)使特征表達(dá)能力得到提升,如宋練等[29]使用2 320 nm和1 755 nm、2 260 nm和2 210 nm、1 920 nm和480 nm的波段進(jìn)行反演,發(fā)現(xiàn)As、Cd和Zn與上述6個(gè)波段沒(méi)有很好的相關(guān)性,但將這6個(gè)波段進(jìn)行組合可得到較好的相關(guān)性,預(yù)測(cè)模型相關(guān)系數(shù)2分別為0.80、0.71、0.61。需要指出的是,并不是所有重金屬元素都有光譜響應(yīng),土壤中的黏土礦物、鐵錳氧化物和有機(jī)質(zhì)等對(duì)重金屬離子有吸附作用,這些組分對(duì)光譜響應(yīng)較強(qiáng),故仍可通過(guò)監(jiān)測(cè)其與土壤中其他光譜響應(yīng)強(qiáng)的組分共同變化來(lái)監(jiān)測(cè)光譜特征不明顯的重金屬[30]。
2) 土壤重金屬反演模型
土壤重金屬反演模型主要分為物理模型和數(shù)學(xué)經(jīng)驗(yàn)?zāi)P汀N锢砟P褪峭ㄟ^(guò)確定的物理機(jī)理闡述光譜反射率與重金屬含量之間的關(guān)系,但在實(shí)際應(yīng)用中,建立物理模型所需的參數(shù)很難獲得,且受環(huán)境因素影響很大,故通常利用數(shù)學(xué)經(jīng)驗(yàn)?zāi)P蛯?duì)土壤重金屬含量與光譜反射率進(jìn)行模型擬合,常用的統(tǒng)計(jì)模型包括偏最小二乘回歸(PLSR)[17-18]、多元逐步回歸(MSR)[19]、多元線性回歸(MLR)[31]等方法,MLR和MSR可以用于建立重金屬含量與光譜特性之間的線性關(guān)系,當(dāng)變量自相關(guān)的時(shí)候,則需要PLSR來(lái)解釋,PLSR可以認(rèn)為是回歸分析、主成分分析、相關(guān)性分析的綜合,通常應(yīng)用于測(cè)量重金屬豐度[32]。此外,一些非線性的數(shù)學(xué)統(tǒng)計(jì)方法如人工神經(jīng)網(wǎng)絡(luò)(ANN)[25]、支持向量機(jī)法(SVM)[33]、決策樹(shù)法(DT)、超限學(xué)習(xí)機(jī)(ELM)[34]等,也被廣泛應(yīng)用于反演中?,F(xiàn)將文獻(xiàn)[19,27-28,34]中,利用無(wú)人機(jī)高光譜影像反演重金屬含量成果總結(jié)見(jiàn)表1。
表1 單波段反演重金屬文獻(xiàn)總結(jié)
綜上所述,通過(guò)構(gòu)建重金屬含量與光譜的相關(guān)關(guān)系模型反演重金屬含量的方法可行。因?yàn)楣庾V數(shù)據(jù)的采集會(huì)受各種各樣因素的影響,如大氣吸收和散射、土壤含水量和粒徑、植被覆蓋、凋落物等[35],利用光譜數(shù)據(jù)反演土壤重金屬的含量具有很大的挑戰(zhàn)性,因此,在進(jìn)行反演時(shí),選擇合適的波段和建模方法至關(guān)重要。根據(jù)重金屬的吸附和聚集過(guò)程,重金屬的聚集受到多種因素綜合影響,如氧化錳、二氧化硅、氧化鋁、有機(jī)質(zhì)和離子環(huán)境等,單一條件下的預(yù)測(cè)仍有其局限性?,F(xiàn)有研究表明,隨著地表類型的復(fù)雜化,重金屬反演精度越來(lái)越低[25],故在地形復(fù)雜地區(qū)的可行性還有待進(jìn)一步研究。
土壤含水率作為土壤重要的理化性質(zhì)之一,直接影響作物生長(zhǎng),對(duì)土壤水分進(jìn)行實(shí)時(shí)、準(zhǔn)確監(jiān)測(cè),有利于農(nóng)作物管理和提高水資源利用率。土壤水分監(jiān)測(cè)方法主要有傳統(tǒng)測(cè)定方法和遙感監(jiān)測(cè)方法。傳統(tǒng)測(cè)定方法有烘干稱重法、張力計(jì)法、電阻法、中子法[36]等,需要在實(shí)驗(yàn)室測(cè)定,雖然精度高但測(cè)定范圍有限,工作量大,難以滿足大范圍的監(jiān)測(cè)要求[37]。遙感技術(shù)利用特定波段下土壤反射率與土壤水分的關(guān)系對(duì)土壤含水率進(jìn)行估算。常用到的遙感波段有可見(jiàn)光–近紅外[38]、熱紅外[39]和微波[40]。雖然上述方法能大范圍監(jiān)測(cè)土壤含水率,但存在精度低、時(shí)效性差、成本高等問(wèn)題。無(wú)人機(jī)遙感技術(shù)具有靈活性強(qiáng)、分辨率高、采集數(shù)據(jù)速度快等優(yōu)點(diǎn),正好彌補(bǔ)了傳統(tǒng)檢測(cè)方法和衛(wèi)星遙感監(jiān)測(cè)的不足。
研究表明,土壤水分與土壤光譜反射率呈負(fù)相關(guān)關(guān)系,具有較高反射率的土壤較為干燥,而熱紅外反射率隨土壤含水率的增加而增加,呈正相關(guān)關(guān)系[39]。無(wú)人機(jī)遙感監(jiān)測(cè)方法一般有2種,其一是基于熱紅外影像通過(guò)熱慣量法和土壤水分指數(shù)法反演土壤含水率。土壤熱慣量是引起土壤表層溫度變化的內(nèi)在因素,與土壤含水量有著密切聯(lián)系,同時(shí)又控制著土壤溫度日較差的大小。土壤水分較大時(shí),土壤具有較大的熱慣量。在實(shí)際應(yīng)用當(dāng)中,熱慣量常用ATI表示,公式如下[41]:
土壤水分指數(shù)法則是利用無(wú)人機(jī)搭載輕型熱相機(jī)和RGB相機(jī)等,獲得復(fù)合地表溫度和彩色圖像,得到地表–空氣溫差圖、歸一化紅綠差值指數(shù)(NGRDI)、水分虧缺指數(shù)(WDI)等,建立土壤含水量預(yù)測(cè)模型,并用計(jì)算出的指數(shù)繪制水分缺失圖,分析土壤水分分布情況[42]。上述參數(shù)的計(jì)算方法如下:
其二是直接利用土壤反射率與實(shí)測(cè)土壤含水率建立相關(guān)關(guān)系進(jìn)行反演,或利用土壤光譜反射率提取的參數(shù)與實(shí)測(cè)土壤水分含量的相關(guān)關(guān)系來(lái)建立反演模型,如歸一化光譜斜率吸收指數(shù)(NSSAI)[37]、增強(qiáng)性植被指數(shù)(EVI)[42]、垂直干旱指數(shù)(PDI)[43]等。張智韜等[44]、王海峰等[45]通過(guò)野外采樣、室內(nèi)理化分析、光譜數(shù)據(jù)的采集和處理等一系列工作,利用無(wú)人機(jī)搭載多光譜相機(jī)采集6個(gè)波段(490、550、680、720、800、900 nm)的土壤光譜反射率,每次拍攝完后使用稱量法測(cè)量土壤含水率,再利用偏最小二乘回歸法、逐步回歸法和嶺回歸法的一元回歸模型建立土壤含水率與光譜反射率的相關(guān)回歸模型。也有學(xué)者使用Spequoia多光譜相機(jī)和ECH2O土壤水分傳感器采集4個(gè)波段(550、660、735、790 nm)的土壤光譜反射率和土壤含水率數(shù)據(jù),結(jié)合PLSR、嶺回歸法、BP神經(jīng)網(wǎng)絡(luò)3種方法建立土壤含水率反演模型[46],為礦區(qū)地表土壤水分監(jiān)測(cè)提供理論依據(jù)和實(shí)踐應(yīng)用參考。以上光譜參數(shù)的計(jì)算方法如下:
式中:為土壤基線的斜率,土壤基線是近紅外和紅波段范圍內(nèi)土壤光譜變化近似的曲線。
綜上所述,無(wú)人機(jī)遙感監(jiān)測(cè)技術(shù)獲取土壤水分是基于土壤表面的遙感數(shù)據(jù),通過(guò)研究遙感數(shù)據(jù)與土壤水分的關(guān)系,建立土壤水分與遙感數(shù)據(jù)的關(guān)系模型來(lái)反演土壤水分信息,近些年來(lái)應(yīng)用無(wú)人機(jī)監(jiān)測(cè)逐漸增多,但針對(duì)于礦區(qū)土壤含水率反演的研究還較少,并且對(duì)土壤水分進(jìn)行估算時(shí)考慮的影響因素較少,如未考慮植被、土地利用類型等。
土壤鹽漬化是指土壤底層或地下水的鹽分隨毛管上升到地表,水分蒸發(fā)后,使鹽分積累在表層土壤中的過(guò)程,是自然因素和人為因素綜合作用的結(jié)果。及時(shí)準(zhǔn)確地獲取土壤鹽分的空間分布及時(shí)空演變規(guī)律對(duì)防止土壤鹽漬化、提高農(nóng)作物生產(chǎn)力具有指導(dǎo)意義。傳統(tǒng)土壤含鹽量的測(cè)定方法是測(cè)定土壤的電導(dǎo)率,需要現(xiàn)場(chǎng)定點(diǎn)采樣和實(shí)驗(yàn)室分析。此類方法雖測(cè)量精度較高,但費(fèi)時(shí)費(fèi)力,難以實(shí)現(xiàn)大尺度、實(shí)時(shí)、動(dòng)態(tài)地檢測(cè)土壤鹽分信息。近些年,隨著遙感技術(shù)的快速發(fā)展,具有綜合、宏觀等特點(diǎn)的衛(wèi)星遙感及便攜式地物光譜儀等遙感技術(shù)已被廣泛應(yīng)用。但衛(wèi)星遙感和便攜式地物光譜儀自身存在缺點(diǎn),如衛(wèi)星遙感有重返周期長(zhǎng)、分辨率分布不均勻等缺點(diǎn)[47],便攜式地物光譜儀只適用于小范圍的土壤鹽分含量監(jiān)測(cè)。馮文哲等[48]將實(shí)測(cè)含鹽量與無(wú)人機(jī)、GF-1衛(wèi)星2種數(shù)據(jù)的光譜特征因子進(jìn)行相關(guān)性分析,利用MLR、SR、RR分別對(duì)2種數(shù)據(jù)進(jìn)行建模反演,驗(yàn)證了利用無(wú)人機(jī)數(shù)據(jù)進(jìn)行反演精度高,效果好。無(wú)人機(jī)遙感技術(shù)同時(shí)避免了衛(wèi)星遙感和便攜式地物光譜儀的缺點(diǎn),具有較大的覆蓋面積和可觀的數(shù)據(jù),為土壤鹽漬化監(jiān)測(cè)提供了新的可能。
目前,土壤鹽分反演模型常借助數(shù)學(xué)模型來(lái)構(gòu)建,包括以下步驟:① 光譜波段的提??;②光譜指數(shù)的建立;③變量的確定;④定量反演模型的建立。由于土壤中化學(xué)元素的吸收光譜不同,一些主要鹽離子(Na+、Cl–)與光譜反射率存在一定程度的相關(guān)關(guān)系,因此,利用可見(jiàn)光和近紅外光譜可以在一定程度上反演土壤鹽離子,得到土壤鹽分信息。借助遙感光譜探測(cè)土壤鹽度的方法有2種,在植被覆蓋度低或鹽漬化較為嚴(yán)重的裸土地區(qū),可利用無(wú)人機(jī)與高光譜儀、多光譜相機(jī)獲得的光譜影像,提取光譜信息,建立鹽分指數(shù),定量描述土壤鹽度信息。常用于表征土壤鹽漬化的鹽分指數(shù)包括:鹽度指數(shù)(Salinity Index, SI)[49]、歸一化鹽度指數(shù)(Normalized Difference Salinity Index, NDS)[50]等。鹽漬化土壤因其鹽分含量超過(guò)正常閾值進(jìn)而影響植被的生理參數(shù),因此出現(xiàn)紅光波段反射率增加,近紅外波段反射率降低的現(xiàn)象,鑒于此,有諸多學(xué)者在植被覆蓋度較高的地區(qū)利用植被光譜信息,間接預(yù)測(cè)土壤中鹽分含量。涉及的植被指數(shù)例如歸一化植被指數(shù)(Normalized Difference Vegetation Index,NDVI)[51]、增強(qiáng)性植被指數(shù)(Enhanced Vegetation Index,EVI)[52]、土壤調(diào)節(jié)植被指數(shù)(Soil-Adjusted Vegetation Index,SAVI)[53]等,以上植被指數(shù)都已被證明可以有效地表征植被光譜與土壤含鹽量的相關(guān)關(guān)系,具體鹽分指數(shù)與植被指數(shù)計(jì)算方式見(jiàn)表2。
表2 用于土壤含鹽量監(jiān)測(cè)的鹽分指數(shù)與植被指數(shù)
注:、、、分別為綠光、紅光、近紅外和藍(lán)光波段光譜反射率;為蓋度背景調(diào)節(jié)因子,取0.5。
在利用光譜指數(shù)建立土壤反演預(yù)測(cè)模型過(guò)程中,會(huì)產(chǎn)生多余的信息,可利用變量選擇的方法對(duì)冗余信息進(jìn)行篩選,從而達(dá)到提高模型預(yù)測(cè)精度的目的。確定敏感變量的方法主要有變量投影重要性分析、灰度關(guān)聯(lián)分析、逐步回歸分析、連續(xù)投影算法等,其中變量投影重要性分析是基于PLSR法的一種變量篩選方法,通過(guò)計(jì)算VIP得分確定變量的重要性,實(shí)現(xiàn)自變量的排序[54];灰度關(guān)聯(lián)分析法根據(jù)各因素之間發(fā)展趨勢(shì)的相似程度或差異程度來(lái)衡量各因素之間相關(guān)性的一種方法[55];逐步回歸分析是根據(jù)自變量對(duì)因變量的作用和顯著程度,剔除無(wú)價(jià)值的波段[56],這3種方法目前被證明可以有效篩選模型輸入變量。包括PLSR在內(nèi)的線性回歸方程是目前常用于估算土壤含鹽量的數(shù)理統(tǒng)計(jì)模型,但實(shí)際上,光譜變量與土壤含鹽量很少線性相關(guān),因此引入機(jī)器學(xué)習(xí)算法,如反向傳播神經(jīng)網(wǎng)絡(luò)(BPNN)、支持向量回歸(SVR)、隨機(jī)森林(RF)、極限學(xué)習(xí)機(jī)(ELM)和多元混合回歸模型等。魏廣飛[54]、王海峰[56]、張智韜等[57]的研究發(fā)現(xiàn),同一模型在不同變量篩選方法下,預(yù)測(cè)精度會(huì)相差很大,且選取的參數(shù)并不一定是適宜模型的最優(yōu)參數(shù)。所以,在減少數(shù)據(jù)冗余度、降低模型復(fù)雜度、拓展模型普適性的基礎(chǔ)上,提高模型的反演精度和運(yùn)算效率等方面仍有很大的探索空間。
煤炭地下開(kāi)采破壞煤層覆巖原有的應(yīng)力平衡狀態(tài),導(dǎo)致覆巖從下至上產(chǎn)生垮落、裂隙和彎曲下沉,表現(xiàn)為采空區(qū)上方地表發(fā)生大面積沉陷。目前,針對(duì)無(wú)人機(jī)監(jiān)測(cè)采煤礦區(qū)地表沉陷的研究已取得很多成果。
地表沉陷的監(jiān)測(cè)流程:首先根據(jù)煤礦的地形和地表沉陷特征確定測(cè)量點(diǎn)和控制點(diǎn)的布設(shè)方案,然后在研究區(qū)內(nèi)均勻地布設(shè)地表沉陷監(jiān)測(cè)標(biāo)志點(diǎn),外業(yè)采集及內(nèi)業(yè)處理點(diǎn)云數(shù)據(jù);其次進(jìn)行地面數(shù)字模型的構(gòu)建;最后對(duì)多期建模數(shù)據(jù)進(jìn)行疊加分析,獲取地表沉陷的精細(xì)特征,確定地表移動(dòng)下沉情況[58]。通常在無(wú)人機(jī)上搭載激光雷達(dá),對(duì)礦區(qū)地面的周期性掃描可得到采煤沉陷區(qū)高精度、高分辨率的點(diǎn)云數(shù)據(jù),經(jīng)過(guò)濾波和插值處理,得到數(shù)字高程模型(DEM),通過(guò)對(duì)多期DEM進(jìn)行疊加可得數(shù)字高程變化模型,即沉陷DEM。現(xiàn)常用的點(diǎn)云插值方法有專業(yè)化數(shù)字高程模型插值(ANU-DEM)、反距離權(quán)重插值(IDW)、克里金插值(Kriging)、自然鄰域插值(NN)、樣條函數(shù)插值(Spline)、改進(jìn)謝別德插值算法(SPD)、徑向基函數(shù)插值算法(RBF)、局部多項(xiàng)式插值算法等[59-61]。在構(gòu)建DEM時(shí),通常使用濾波技術(shù)分離點(diǎn)云地面點(diǎn),理論依據(jù)是基于鄰近腳點(diǎn)的高程突變。根據(jù)濾波的算法原理不同可以將其分成基于坡度、曲面擬合、分割、不規(guī)則三角網(wǎng)、形態(tài)學(xué)及機(jī)器學(xué)習(xí)的濾波算法6大類[62-64]。
無(wú)人機(jī)攝影測(cè)量技術(shù)和機(jī)載LiDAR技術(shù)作業(yè)時(shí)受儀器精度、地形特征、操作者經(jīng)驗(yàn)等多種影響,獲取的點(diǎn)云數(shù)據(jù)通常含有大量的噪聲點(diǎn),會(huì)嚴(yán)重影響濾波的準(zhǔn)確性和分類結(jié)果。點(diǎn)云數(shù)據(jù)的去噪方法主要包含網(wǎng)格化去噪和模型去噪2種方法。網(wǎng)格化去噪是利用網(wǎng)格把點(diǎn)云數(shù)據(jù)進(jìn)行分割,能提高去噪的效率,但是需要重建網(wǎng)格,并且在去噪過(guò)程中,需要重新建立網(wǎng)格的原有拓?fù)湫畔?,極大地增加了算法的復(fù)雜度,在一定的程度上可能還會(huì)導(dǎo)致重建的網(wǎng)格出現(xiàn)拓?fù)湫畔㈠e(cuò)誤。模型去噪是直接利用去噪方法對(duì)點(diǎn)云數(shù)據(jù)進(jìn)行去噪,可以有效地防止網(wǎng)格出現(xiàn)拓?fù)湫畔㈠e(cuò)誤的情況。對(duì)于點(diǎn)云去噪,現(xiàn)階段還是以濾波為主。近年來(lái),有許多學(xué)者在研究點(diǎn)云的基礎(chǔ)上提出許多去噪方法,如對(duì)比高程均值離差去噪方法[65]、自適應(yīng)移動(dòng)盒子去噪算法[65]、基于法向量為特征向量的網(wǎng)格去噪方法[66]、基于小波閾值的沉陷模型去噪[67]、移動(dòng)最小二乘曲面算法[72]、各向異性消噪算法[68],以及統(tǒng)計(jì)濾波、半徑濾波和快速雙邊濾波結(jié)合的去噪算法[69]等。
利用多期DEM的疊加分析,不僅能得到數(shù)字高程變化模型,計(jì)算出礦區(qū)的地表移動(dòng)變形參數(shù),還能提取出礦區(qū)開(kāi)采沉陷邊界,得到礦區(qū)地表沉陷的影響范圍。Xu Xianlei等[70]介紹了一種確定采煤沉陷破壞耕地邊界的新方法和評(píng)價(jià)準(zhǔn)則,認(rèn)為邊界可以45 mm下沉線或大于0.5的附加坡度確定;侯恩科等[71]、高冠杰等[72]通過(guò)對(duì)比工作面采前和采后的地表高程差,進(jìn)行地表沉降量計(jì)算與成圖;Zhou Dawei等[73]也利用無(wú)人機(jī)攝影技術(shù)對(duì)開(kāi)采引起的地表動(dòng)態(tài)沉陷盆地進(jìn)行監(jiān)測(cè)并提出獲得短期開(kāi)采沉陷參數(shù)的方法,已經(jīng)成功應(yīng)用于內(nèi)蒙古王家莊煤礦。
綜上,無(wú)人機(jī)在監(jiān)測(cè)礦區(qū)災(zāi)害方面對(duì)不易到達(dá)和飛行困難地區(qū)的高分辨率影像獲取具有明顯優(yōu)勢(shì)。隨著無(wú)人機(jī)數(shù)據(jù)處理效率和精度的不斷提高,無(wú)人機(jī)監(jiān)測(cè)技術(shù)將成為礦山地表沉陷監(jiān)測(cè)的新型主流監(jiān)測(cè)技術(shù),具有很好的推廣價(jià)值。
采煤引起的地表塌陷是礦區(qū)地表移動(dòng)變化的一種外在表現(xiàn)形式,而地裂縫是煤礦區(qū)最常見(jiàn)、最直觀的一種地面塌陷形式之一。監(jiān)測(cè)地表裂縫的手段包括早期的人工野外裂縫填圖、布置觀測(cè)線、觀測(cè)點(diǎn)[74-75]、衛(wèi)星遙感[76]、三維激光掃描[77]、InSAR/SAR[78]、無(wú)人機(jī)監(jiān)測(cè)技術(shù)[79-80]等。相比于其他技術(shù),無(wú)人機(jī)遙感具有成本低、效率高、精度較高、機(jī)動(dòng)靈活等優(yōu)點(diǎn),已被多次應(yīng)用于地面塌陷調(diào)查中?,F(xiàn)階段,利用無(wú)人機(jī)監(jiān)測(cè)地裂縫的研究,主要集中在對(duì)地裂縫發(fā)育規(guī)律、分布特征的研究。侯恩科等[81]以羊場(chǎng)灣煤礦為研究區(qū),總結(jié)分析該區(qū)地表裂縫發(fā)育規(guī)律、特征及其與采礦地質(zhì)條件的關(guān)系;毛崔磊[82]采用谷歌遙感影像數(shù)據(jù)與無(wú)人機(jī)數(shù)據(jù)的結(jié)合,對(duì)平朔礦區(qū)歷年地裂縫的分布情況進(jìn)行提取,從時(shí)間維度分析采煤地裂縫的線密度和長(zhǎng)度,從空間維度利用分形理論對(duì)采煤地裂縫進(jìn)行分析,定量化描述地裂縫的發(fā)育情況。
基于無(wú)人機(jī)遙感影像提取地裂縫的處理流程為:影像預(yù)處理、地裂縫初提取、“非地裂縫”目標(biāo)去除、地裂縫結(jié)果精處理以及地裂縫提取結(jié)果精度評(píng)定。地裂縫初提取常用方法有邊緣檢測(cè)算法、灰度閾值分割算法、影像分類算法[83]。邊緣檢測(cè)算法是一種采用把邊緣檢測(cè)算子增強(qiáng)影像中邊緣信號(hào)的數(shù)字處理方法,可分為一階邊緣檢測(cè)算法和二階邊緣檢測(cè)算法[84]。典型的邊緣檢測(cè)算子包括Roberts算子、Sobel算子、Prewitt算子和Canny算子、拉普拉斯算子和高斯拉普拉斯算子等[85]。遙感影像分類是從光譜和紋理信息豐富,各個(gè)類別之間對(duì)比度存在較大差異的遙感影像中提取條帶狀線性目標(biāo)的一種方法。影像分類方法有最大似然分類法、最小距離分類法、決策樹(shù)分類法、均值分類法和ISODATA分類法等[86]。基于機(jī)器學(xué)習(xí)和神經(jīng)網(wǎng)絡(luò)等的遙感影像分類方法是近幾年的研究熱點(diǎn)。
對(duì)于有些非裂縫信息被誤提取的問(wèn)題,可采用針對(duì)遙感影像植被分類的算法如最鄰近距離法、決策樹(shù)法、隨機(jī)森林法(Random Forests,RF)等生成掩膜文件,消除植被影響,實(shí)現(xiàn)裂縫的精提取。2015年,F(xiàn)eng Quanlong等[87]證明了結(jié)合紋理信息的隨機(jī)森林算法能有效地對(duì)植被進(jìn)行分類,其結(jié)果比最鄰近距離、面向?qū)ο笠约癝VM等算法更優(yōu)。故大部分研究采用RF算法制成掩膜文件,去除非地表裂縫信息。由于地裂縫的灰度值并不隨地裂縫位置的變化而均勻變化,且地裂縫初提取和非地裂縫目標(biāo)去除過(guò)程以單個(gè)像元為操作對(duì)象,因此,地裂縫的最終提取結(jié)果會(huì)存在一定的斷續(xù)現(xiàn)象,除了明顯的地裂縫圖斑之外,結(jié)果中還含有一些面積遠(yuǎn)小于地裂縫的碎小圖斑,影響提取精度和地裂縫的判別,故應(yīng)對(duì)地裂縫提取結(jié)果進(jìn)行精處理。數(shù)學(xué)形態(tài)學(xué)中的閉運(yùn)算、擊中擊不中變換算法具有連接二值影像中斷裂目標(biāo)的功能[88-89],開(kāi)運(yùn)算和面積濾波算法可進(jìn)行碎小圖斑去除[90],可用于地裂縫精處理研究。
綜上,無(wú)人機(jī)能全面準(zhǔn)確地展現(xiàn)地表裂縫的分布特征與發(fā)育規(guī)律。但在后期的影像處理方面,地裂縫的提取受植被覆蓋度及地裂縫陰影的影響,斷續(xù)現(xiàn)象增加。為此,可結(jié)合采動(dòng)裂縫發(fā)育規(guī)律及裂縫圖像的灰度特征,構(gòu)建斷續(xù)裂縫的自動(dòng)生成算法。同時(shí),針對(duì)隱性地裂縫,其灰度差異性變小,可結(jié)合實(shí)地調(diào)查及這種裂縫的其他光譜特征進(jìn)行解譯,以實(shí)現(xiàn)采動(dòng)裂縫的高效監(jiān)測(cè)與自動(dòng)獲取。
露天礦開(kāi)采過(guò)程中,礦山極易發(fā)生邊坡失穩(wěn),導(dǎo)致滑坡、崩塌、泥石流等地質(zhì)災(zāi)害,其形成原因主要是位于不穩(wěn)定邊坡上的巖土體在降雨、采煤擾動(dòng)等條件下,受自身重力及其他應(yīng)力的綜合影響下沿著傾斜面發(fā)生滑動(dòng)或崩塌。由于其分布廣泛、發(fā)生頻繁、形成原因以及變形機(jī)理復(fù)雜、很難進(jìn)行人為調(diào)查,現(xiàn)常使用無(wú)人機(jī)監(jiān)測(cè)邊坡的穩(wěn)定性,了解滑坡情況[91-92]。從工程塌方和山嶺地區(qū)滑坡的防災(zāi)減災(zāi)角度,邊坡的監(jiān)測(cè)技術(shù)可以大致分為邊坡的位移監(jiān)測(cè)、加固體的支擋物監(jiān)測(cè)、巖體破裂監(jiān)測(cè)、水的監(jiān)測(cè)和巡檢5個(gè)主要類型。
國(guó)內(nèi)外利用無(wú)人機(jī)對(duì)邊坡穩(wěn)定性的研究主要集中在邊坡的位移監(jiān)測(cè)和巖體破裂監(jiān)測(cè)上?;谟?jì)算機(jī)視覺(jué)和圖像處理技術(shù)對(duì)無(wú)人機(jī)獲取的圖像進(jìn)行處理,探討邊坡斜體邊緣的提取方法,可實(shí)現(xiàn)邊坡穩(wěn)定的可視化,從而評(píng)估邊坡的位移情況[93]。利用無(wú)人機(jī)攝影測(cè)量技術(shù)獲得邊坡表面地形和裂隙圖像,通過(guò)基于面向?qū)ο蠓治龅臑V波算法及濾波后圖像的后處理技術(shù),可對(duì)邊坡裂縫發(fā)育和失穩(wěn)趨勢(shì)進(jìn)行預(yù)測(cè)[89]。宋誠(chéng)[94]通過(guò)無(wú)人機(jī)攝影測(cè)量技術(shù)實(shí)現(xiàn)對(duì)邊坡地形高精度測(cè)量,并將其成功應(yīng)用于邊坡變形監(jiān)測(cè)中,為邊坡治理提供了依據(jù)。無(wú)人機(jī)監(jiān)測(cè)主要是利用無(wú)人機(jī)攝影測(cè)量技術(shù),獲取礦區(qū)數(shù)據(jù),利用三維點(diǎn)云處理軟件對(duì)模型進(jìn)行裁剪、去噪、平滑等處理,其次利用三維建模軟件實(shí)現(xiàn)可視化,進(jìn)而分析邊坡穩(wěn)定性。
目前,邊坡穩(wěn)定性分析的確定性分析常用方法主要有極限平衡法和數(shù)值模擬法2種,極限平衡法是以Mohr-Coulomb的抗剪強(qiáng)度理論為基礎(chǔ),根據(jù)斜坡上滑體的力學(xué)平衡原理分析斜坡的受力狀態(tài),以及斜坡抗滑力與下滑力之間的關(guān)系來(lái)評(píng)價(jià)斜坡的穩(wěn)定性[95]。靳遠(yuǎn)成等[96]采用此方法從無(wú)人機(jī)飛行數(shù)據(jù)三維建模結(jié)果截取多個(gè)剖面后進(jìn)行數(shù)值分析,結(jié)果表明:將無(wú)人機(jī)影像的邊坡精細(xì)化建模技術(shù)應(yīng)用于邊坡穩(wěn)定性分析,可以更快速地反映邊坡穩(wěn)定性情況。數(shù)值模擬方法主要是通過(guò)無(wú)人機(jī)影像后期處理獲得的三維數(shù)據(jù)通過(guò)有限元軟件ANSYS、ABAQUS、NASTRAN、FLAC3D等進(jìn)行計(jì)算[97]。金愛(ài)兵等[98]利用無(wú)人機(jī)攝影測(cè)量技術(shù),獲得邊坡測(cè)量數(shù)據(jù),將DEM與FLAC數(shù)值模型結(jié)合,達(dá)到高精度高效率分析邊坡穩(wěn)定性的目的。
無(wú)人機(jī)攝影測(cè)量技術(shù)具有成本低、操作簡(jiǎn)便、精度高等優(yōu)點(diǎn),可作為邊坡穩(wěn)定性監(jiān)測(cè)手段在礦區(qū)監(jiān)測(cè)中進(jìn)行推廣使用。但目前所使用的位移測(cè)量方法,在精度的穩(wěn)定性上仍存在不足,原因是無(wú)人機(jī)影像受風(fēng)速影響,每次所拍攝的相片質(zhì)量存在差異,因此,如何保證獲取圖像質(zhì)量的穩(wěn)定性是亟待解決的問(wèn)題。
a.無(wú)人機(jī)因可操作性強(qiáng),低成本和采集數(shù)據(jù)速度快,可以在人員難以進(jìn)入的地區(qū)獲取數(shù)據(jù)等優(yōu)點(diǎn)在礦業(yè)領(lǐng)域有極大的研究潛力和應(yīng)用前景。目前無(wú)人機(jī)在礦區(qū)生態(tài)環(huán)境中的監(jiān)測(cè)應(yīng)用仍處于初步階段,設(shè)計(jì)符合監(jiān)測(cè)需求的方案,選擇合適的傳感器,得到高分辨率圖像,根據(jù)數(shù)據(jù)處理結(jié)果作出評(píng)價(jià)仍是未來(lái)研究的熱點(diǎn)話題。
b.監(jiān)測(cè)地表土壤特征時(shí),雖然建模精度較高,但隨著地表類型、植被種類、復(fù)墾管理方式等的不同,反演精度會(huì)發(fā)生較大變化,容易受到多種因素的綜合影響,單一條件下反演預(yù)測(cè)仍有局限性。另外,上述模型大多基于平原、丘陵或縣域等自然狀態(tài)土壤進(jìn)行模型開(kāi)發(fā),針對(duì)采煤礦區(qū)土壤質(zhì)量演化趨勢(shì)模型方面的研究較少,未來(lái)有待進(jìn)一步研究。
c.在礦區(qū)地質(zhì)災(zāi)害研究方面,由于無(wú)人機(jī)能高精度、高頻率、多角度地持續(xù)動(dòng)態(tài)監(jiān)測(cè)地質(zhì)災(zāi)害情況,被廣泛應(yīng)用于研究地表塌陷、地裂縫的形成機(jī)理與發(fā)育規(guī)律以及采礦擾動(dòng)對(duì)土壤質(zhì)量變化的影響規(guī)律等方面。但目前少有研究利用無(wú)人機(jī)監(jiān)測(cè)不同程度地質(zhì)災(zāi)害對(duì)土壤質(zhì)量的影響,也未將這些變化與采礦參數(shù)、地表移動(dòng)變形和地表裂縫分布聯(lián)系起來(lái),今后可從這方面尋求突破。
d.未來(lái)可通過(guò)集成野外時(shí)序跟蹤調(diào)查、高精度土壤質(zhì)量監(jiān)測(cè)技術(shù)、高空間分辨率無(wú)人機(jī)監(jiān)測(cè)技術(shù)、數(shù)字模擬手段和典型工作面的試驗(yàn)監(jiān)測(cè)與分析,研究工作面自切眼至終采線動(dòng)態(tài)推進(jìn)中地質(zhì)災(zāi)害與土壤質(zhì)量演化耦合關(guān)系,構(gòu)建采煤沉陷區(qū)土壤質(zhì)量演化預(yù)測(cè)理論體系和時(shí)序演變模型,進(jìn)一步探討基于無(wú)人機(jī)高光譜影像的表土元素反演,提出減緩、控制及提升礦區(qū)土壤質(zhì)量的措施,為我國(guó)煤炭生產(chǎn)基地煤炭資源開(kāi)采與生態(tài)環(huán)境的協(xié)調(diào)可持續(xù)發(fā)展提供技術(shù)支撐。
[1] 胡振琪,肖武. 關(guān)于煤炭工業(yè)綠色發(fā)展戰(zhàn)略的若干思考:基于生態(tài)修復(fù)視角[J]. 煤炭科學(xué)技術(shù),2020,48(4):35–42.
HU Zhenqi,XIAO Wu. Some thoughts on green development strategy of coal industry:From aspects of ecological restoration[J]. Coal Science and Technology,2020,48(4):35–42.
[2] 文博杰,陳毓川,王高尚,等. 2035年中國(guó)能源與礦產(chǎn)資源需求展望[J]. 中國(guó)工程科學(xué),2019,21(1):68–73.
WEN Bojie,CHEN Yuchuan,WANG Gaoshang,et al. China’s demand for energy and mineral resources by 2035[J]. Engineering Sciences,2019,21(1):68–73.
[3] 王丹爽. 無(wú)人機(jī)技術(shù)在施工現(xiàn)場(chǎng)安全評(píng)價(jià)中的應(yīng)用研究[D]. 哈爾濱:哈爾濱工業(yè)大學(xué),2017.
WANG Danshuang. Application of UAV technology in construction site safety evaluation[D]. Harbin:Harbin Institute of Technology,2017.
[4] 張凱,李全生,戴華陽(yáng),等. 礦區(qū)地表移動(dòng)“空天地”一體化監(jiān)測(cè)技術(shù)研究[J]. 煤炭科學(xué)技術(shù),2020,48(2):207–213.
ZHANG Kai,LI Quansheng,DAI Huayang,et al. Research on integrated monitoring technology and practice of “space-sky-ground” on surface movement in mining area[J]. Coal Science and Technology,2020,48(2):207–213.
[5] 姚一飛. 基于北斗增強(qiáng)的實(shí)時(shí)PPP及在礦山地表監(jiān)測(cè)的研究[D].徐州:中國(guó)礦業(yè)大學(xué),2017.
YAO Yifei. Research on real-time precise point positioning based on Beidou augmentation and its application in mining surface monitoring[D]. Xuzhou:China University of Mining and Technology,2017.
[6] 仝云霄. InSAR礦區(qū)地表形變監(jiān)測(cè)及大氣延遲校正研究[D]. 徐州:中國(guó)礦業(yè)大學(xué),2020.
TONG Yunxiao. Study on surface deformation monitoring in mining area and atmospheric delay correction of InSAR[D]. Xuzhou:China University of Mining and Technology,2020.
[7] 廉旭剛,蔡音飛,胡海峰. 我國(guó)礦山測(cè)量領(lǐng)域三維激光掃描技術(shù)的應(yīng)用現(xiàn)狀及存在問(wèn)題[J]. 金屬礦山,2019(3):35–40.
LIAN Xugang,CAI Yinfei,HU Haifeng. Application status and existing problems of 3D laser scanning technique in mine surveying in China[J]. Metal Mine,2019(3):35–40.
[8] 徐坤,屈瑩,王寶山. 傾斜攝影測(cè)量技術(shù)在礦產(chǎn)資源監(jiān)測(cè)中的應(yīng)用[J]. 測(cè)繪工程,2020,29(4):38–43.
XU Kun,QU Ying,WANG Baoshan. Application of oblique photogrammetry technology to the dynamic monitoring of mineral resources[J]. Engineering of Surveying and Mapping,2020,29(4):38–43.
[9] 王昆,楊鵬,呂文生,等. 無(wú)人機(jī)遙感在礦業(yè)領(lǐng)域應(yīng)用現(xiàn)狀及發(fā)展態(tài)勢(shì)[J]. 工程科學(xué)學(xué)報(bào),2020,42(9):1085–1095.
WANG Kun,YANG Peng,LYU Wensheng,et al. Current status and development trend of UAV remote sensing applications in the mining industry[J]. Chinese Journal of Engineering,2020,42(9):1085–1095.
[10] 白洋,康會(huì)濤,張文超,等. 無(wú)人機(jī)在露天礦山監(jiān)測(cè)中的應(yīng)用[J]. 測(cè)繪通報(bào),2020(9):85–88.
BAI Yang,KANG Huitao,ZHANG Wenchao,et al. Application of UAV in open-pit mine monitoring[J]. Bulletin of Surveying and Mapping,2020(9):85–88.
[11] 許吉仁,董霽紅,楊源譞,等. 基于支持向量機(jī)的礦區(qū)復(fù)墾農(nóng)田土壤–小麥鎘含量高光譜估算[J]. 光子學(xué)報(bào),2014,43(5):1–8.
XU Jiren,DONG Jihong,YANG Yuanxuan,et al. Support vector machine model for predicting the cadmium concentration of soil-wheat system in mine reclamation farmland using hyperspectral data[J]. Acta Photonica Sinica,2014,43(5):1–8.
[12] LIU Meiling,LIU Xiangnan,WU Ling,et al. Wavelet-based detection of crop zinc stress assessment using hyperspectral reflectance[J]. Computers and Geosciences,2011,37(9):1254–1263.
[13] ROMAN M B,SERGEY V S. Variable selection in near-infrared spectroscopy:Benchmarking of feature selection methods on biodiesel data[J]. Analytica Chimica Acta,2011,692(1):63–72.
[14] 葉元元. 多金屬礦區(qū)土壤重金屬的高光譜定量估算研究[D]. 徐州:中國(guó)礦業(yè)大學(xué),2014.
YE Yuanyuan. Quantitative estimating of soil heavy metals with hyper-spectrum in polymetallic mining areas[D]. Xuzhou:China University of Mining and Technology,2014.
[15] 王惠敏. 基于光譜吸收特征的土壤有機(jī)質(zhì)與重金屬含量估算研究[D]. 徐州:中國(guó)礦業(yè)大學(xué),2019.
WANG Huimin. Study on the estimation of soil organic matter and heavy metal contents based on spectral absorption characteristics[D]. Xuzhou:China University of Mining and Technology,2019.
[16] 譚琨,葉元元,杜培軍,等. 礦區(qū)復(fù)墾農(nóng)田土壤重金屬含量的高光譜反演分析(英文)[J]. 光譜學(xué)與光譜分析,2014,34(12):3317–3322.
TAN Kun,YE Yuanyuan,DU Peijun,et al. Estimation of heavy metal concentrations in reclaimed mining soils using reflectance spectroscopy(English)[J]. Spectroscopy and Spectral Analysis,2014,34(12):3317–3322.
[17] ZHANG Shiwen,SHEN Qiang,NIE Chaojia,et al. Hyperspectral inversion of heavy metal content in reclaimed soil from a mining wasteland based on different spectral transformation and modeling methods[J]. Spectrochimica Acta Part A:Molecular and Biomolecular Spectroscopy,2019,211:393–400.
[18] 鄔登巍,吳昀昭,馬宏瑞. 基于中紅外漫反射光譜的土壤重金屬元素含量預(yù)測(cè)研究[J]. 光譜學(xué)與光譜分析,2010,30(6):1498–1502.
WU Dengwei,WU Yunzhao,MA Hongrui. Study on the prediction of soil heavy metal elements content based on mid-infrared diffuse reflectance spectra[J]. Spectroscopy and Spectral Analysis,2010,30(6):1498–1502.
[19] 左玲. 黑土區(qū)土壤重金屬高光譜遙感監(jiān)測(cè)方法探究[D]. 北京:中國(guó)地質(zhì)大學(xué)(北京),2020.
ZUO Ling. Research on hyperspectral remote sensing monitoring method of heavy metals in black soil region[D]. Beijing:China University of Geosciences(Beijing),2020.
[20] ARAúJO S R,DEMATTê J A M,VICENTE S. Soil contaminated with chromium by tannery sludge and identified by vis-NIR-mid spectroscopy techniques[J]. International Journal of Remote Sensing,2014,35(10):3579–3593.
[21] URSZULA S,LAKHMI C J. Feature selection for data and pattern recognition[M]. Heidelberg:Springer,2015.
[22] JAIN A,ZONGKER D. Feature selection:Evaluation,application,and small sample performance[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1997,19(2):153–158.
[23] YU Lei,LIU Huan. Efficient feature selection via analysis of relevance and redundancy[J]. The Journal of Machine Learning Research,2004:1205–1224.
[24] 劉英. 半干旱煤礦區(qū)受損植被引導(dǎo)型恢復(fù)研究[D]. 徐州:中國(guó)礦業(yè)大學(xué),2020.
LIU Ying. Study on guided restoration of damaged vegetation in semi-arid coal mine area[D]. Xuzhou:China University of Mining and Technology,2020.
[25] 胡忠正. 機(jī)載高光譜影像反演土壤重金屬含量的經(jīng)驗(yàn)?zāi)P瓦x擇與特征提取[D]. 北京:中國(guó)地質(zhì)大學(xué)(北京),2019.
HU Zhongzheng. Empirical model selection and feature extraction for retrieving soil heavy metal concentration from airborne hyperspectral imagery[D]. Beijing:China University of Geosciences(Beijing),2019.
[26] BRUCE L M,KOGER C H,LI Jiang. Dimensionality reduction of hyperspectral data using discrete wavelet transform feature extraction[J]. IEEE Transactions on Geoscience and Remote Sensing,2002,40(10):2331–2338.
[27] 宋婷婷. 土壤重金屬含量反演與礦區(qū)環(huán)境遙感監(jiān)測(cè)與評(píng)價(jià)研究[D]. 北京:北京化工大學(xué),2017.
SONG Tingting. Study on inversion of heavy metal content in soil and remote sensing monitoring and evaluation of mine environment[D]. Beijing:Beijing University of Chemical Technology,2017.
[28] 徐良驥,李青青,朱小美,等. 煤矸石充填復(fù)墾重構(gòu)土壤重金屬含量高光譜反演[J]. 光譜學(xué)與光譜分析,2017,37(12):3839–3844.
XU Liangji,LI Qingqing,ZHU Xiaomei,et al. Hyperspectral inversion of heavy metal content in coal gangue filling reclamation land[J]. Spectroscopy and Spectral Analysis,2017,37(12):3839–3844.
[29] 宋練,簡(jiǎn)季,譚德軍,等. 萬(wàn)盛采礦區(qū)土壤As,Cd,Zn重金屬含量光譜測(cè)量與分析[J]. 光譜學(xué)與光譜分析,2014,34(3):812–817.
SONG Lian,JIAN Ji,TAN Dejun,et al. Estimation of soil’s heavy metal concentrations(As,Cd and Zn)in Wansheng mining area with geochemistry and field spectroscopy[J]. Spectroscopy and Spectral Analysis,2014,34(3):812–817.
[30] 徐明星,吳紹華,周生路,等. 重金屬含量的高光譜建模反演:考古土壤中的應(yīng)用[J]. 紅外與毫米波學(xué)報(bào),2011,30(2):109–114.
XU Mingxing,WU Shaohua,ZHOU Shenglu,et al. Hyperspectral reflectance models for retrieving heavy metal content:Application in the archaeological soil[J]. Journal of Infrared and Millimeter Waves,2011,30(2):109–114.
[31] FILZMOSER P,TODOROV V. Review of robust multivariate statistical methods in high dimension[J]. Analytica Chimica Acta,2011,705(1/2):2–14.
[32] ZHANG Xia,HUANG Changping,LIU Bo,et al. Inversion of soil Cu concentration based on band selection of hyperspetral data[C]//2010 IEEE International Geoscience and Remote Sensing Symposium. New York:IEEE,2010:3680–3683.
[33] THISSEN U,PEPERS M,üSTüN B,et al. Comparing support vector machines to PLS for spectral regression applications[J]. Chemometrics and Intelligent Laboratory Systems,2004,73(2):169–179.
[34] 馬偉波,譚琨,李海東,等. 基于超限學(xué)習(xí)機(jī)的礦區(qū)土壤重金屬高光譜反演[J]. 生態(tài)與農(nóng)村環(huán)境學(xué)報(bào),2016,32(2):213–218.
MA Weibo,TAN Kun,LI Haidong,et al. Hyperspectral inversion of heavy metals in soil of a mining area using extreme learning machine[J]. Journal of Ecology and Rural Environment,2016,32(2):213–218.
[35] WANG Fenghe,GAO Jay,ZHA Yong. Hyperspectral sensing of heavy metals in soil and vegetation:Feasibility and challenges[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2018,136:73–84.
[36] MORTEZA S,SCOTT B J,WILLIAM D P. A linear physically-based model for remote sensing of soil moisture using short wave infrared bands[J]. Remote Sensing of Environment,2015,164:66–76.
[37] 鄭小坡,孫越君,秦其明,等. 基于可見(jiàn)光–短波紅外波譜反射率的裸土土壤含水量反演建模[J]. 光譜學(xué)與光譜分析,2015,35(8):2113–2118.
ZHENG Xiaopo,SUN Yuejun,QIN Qiming,et al. Bare soil moisture inversion model based on visible-shortwave infrared reflectance[J]. Spectroscopy and Spectral Analysis,2015,35(8):2113–2118.
[38] 郭輝,卜小東,黃可京,等. 基于熱紅外遙感影像的玉米田間土壤水分反演研究[J]. 中國(guó)農(nóng)機(jī)化學(xué)報(bào),2020,41(10):203–210.
GUO Hui,BU Xiaodong,HUANG Kejing,et al. Inversion of soil moisture in corn field based on thermal infrared remote sensing image[J]. Journal of Chinese Agricultural Mechanization,2020,41(10):203–210.
[39] 李伯祥,陳曉勇. 基于Sentinel多源遙感數(shù)據(jù)的河北省景縣農(nóng)田土壤水分協(xié)同反演[J]. 生態(tài)與農(nóng)村環(huán)境學(xué)報(bào),2020,36(6):752–761.
LI Boxiang,CHEN Xiaoyong. Synergic use of Sentinel-1 and Sentinel-2 images for soil moisture retrieval in vegetation covered agricultural areas of Jingxian County of Hebei Province[J]. Journal of Ecology and Rural Environment,2020,36(6):752–761.
[40] 夏軍. 準(zhǔn)東煤田土壤重金屬污染高光譜遙感監(jiān)測(cè)研究[D]. 烏魯木齊:新疆大學(xué),2014.
XIA Jun. Study on the monitoring of soil heavy metal pollution with hyperspectral remote sensing in the Eastern Junggar Coalfield[D]. Urumqi:Xinjiang University,2014.
[41] HOFFMANN H,JENSEN R,THOMSEN A,et al. Crop water stress maps for entire growing seasons from visible and thermal UAV imagery[J]. Biogeosciences Discussions,2016,13(24):316.
[42] 夏權(quán),夏萍,陳黎卿,等. 基于多光譜遙感的土壤含水量定量監(jiān)測(cè)與分析[J]. 安徽農(nóng)業(yè)大學(xué)學(xué)報(bào),2015,42(3):439–443.
XIA Quan,XIA Ping,CHEN Liqing,et al. Quantitative analysis of the soil moisture content based on multispectral remote sensing data[J]. Journal of Anhui Agricultural University,2015,42(3):439–443.
[43] 馮珊珊,梁雪映,樊風(fēng)雷,等. 基于無(wú)人機(jī)多光譜數(shù)據(jù)的農(nóng)田土壤水分遙感監(jiān)測(cè)[J]. 華南師范大學(xué)學(xué)報(bào)(自然科學(xué)版),2020,52(6):74–81.
FENG Shanshan,LIANG Xueying,F(xiàn)AN Fenglei,et al. Monitoring of farmland soil moisture based on unmanned aerial vehicle multispectral data[J]. Journal of South China Normal University(Natural Science Edition),2020,52(6):74–81.
[44] 張智韜,王海峰,韓文霆,等. 基于無(wú)人機(jī)多光譜遙感的土壤含水率反演研究[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2018,49(2):173–181.
ZHANG Zhitao,WANG Haifeng,HAN Wenting,et al. Inversion of soil moisture content based on multispectral remote sensing of UAVs[J]. Transactions of the Chinese Society of Agricultural Machinery,2018,49(2):173–181.
[45] 王海峰,張智韜,付秋萍,等. 低空無(wú)人機(jī)多光譜遙感數(shù)據(jù)的土壤含水率反演[J]. 節(jié)水灌溉,2018(1):90–94.
WANG Haifeng,ZHANG Zhitao,F(xiàn)U Qiuping,et al. Inversion of soil moisture content based on multispectral remote sensing data of low altitude UAV[J]. Water Saving Irrigation,2018(1):90–94.
[46] 王啟元,趙艷玲,房鑠東,等. 基于多光譜遙感的裸土土壤含水量反演研究[J]. 礦業(yè)科學(xué)學(xué)報(bào),2020,5(6):608–615.
WANG Qiyuan,ZHAO Yanling,F(xiàn)ANG Shuodong,et al. Inversion of soil moisture in bare soil based on multi-spectral remote sensing[J]. Journal of Mining Science and Technology,2020,5(6):608–615.
[47] 陳俊英,王新濤,張智韜,等. 基于無(wú)人機(jī)–衛(wèi)星遙感升尺度的土壤鹽漬化監(jiān)測(cè)方法[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2019,50(12):161–169.
CHEN Junying,WANG Xintao,ZHANG Zhitao,et al. Soil salinization monitoring method based on UAV-satellite remote sensing scale-up[J]. Transactions of the Chinese Society of Agricultural Machinery,2019,50(12):161–169.
[48] 馮文哲,王新濤,韓佳,等. 基于衛(wèi)星和無(wú)人機(jī)遙感數(shù)據(jù)尺度轉(zhuǎn)換的土壤鹽漬化監(jiān)測(cè)研究[J]. 節(jié)水灌溉,2020(11):87–93.
FENG Wenzhe,WANG Xintao,HAN Jia,et al. Research on soil salinization monitoring based on scale conversion of satellite and UAV remote sensing data[J]. Water Saving Irrigation,2020(11):87–93.
[49] 邊玲玲. 基于多源數(shù)據(jù)的黃河三角洲土壤鹽分提取方法研究[D]. 淄博:山東理工大學(xué),2020.
BIAN Lingling. Study on the methods of soil salt extraction in the Yellow River Delta based on multi-source data[D]. Zibo:Shandong University of Technology,2020.
[50] ZEWDU S,SURYABHAGAVAN K V,BALAKRISHNAN M. Geo-spatial approach for soil salinity mapping in Sego Irrigation Farm,South Ethiopia[J]. Journal of the Saudi Society of Agricultural Sciences,2017,16(1):16–24.
[51] 王飛,丁建麗,伍漫春. 基于NDVI–SI特征空間的土壤鹽漬化遙感模型[J]. 農(nóng)業(yè)工程學(xué)報(bào),2010,26(8):168–173.
WANG Fei,DING Jianli,WU Manchun. Remote sensing monitoring models of soil salinization based on NDVI-SI feature space[J]. Transactions of the Chinese Society of Agricultural Engineering,2010,26(8):168–173.
[52] LOBELL D B,LESCH S M,CORWIN D L,et al. Regional-scale assessment of soil salinity in the Red River Valley using multi-year MODIS EVI and NDVI[J]. Journal of Environmental Quality,2010,39(1):35–41.
[53] AMAL A,LALIT K,YOUSEF Y A. 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(230/231):1–8.
[54] 魏廣飛. 基于無(wú)人機(jī)多光譜遙感的土壤鹽漬化監(jiān)測(cè)模型研究[D].咸陽(yáng):西北農(nóng)林科技大學(xué),2020.
WEI Guangfei. Research on monitoring model of soil salinization based on multispectral remote sensing of UAV[D]. Xianyang:Northwest A&F University,2020.
[55] SANTOS I C D L,SANTOS A D,OUMAR Z,et al. Remote sensing to detect nests of the leaf-cutting Ant Atta sexdens(Hymenoptera:Formicidae) in teak plantations[J]. Remote Sensing,2019,11(14):1641.
[56] 王海峰. 基于高光譜遙感的土壤水鹽監(jiān)測(cè)模型研究:以河套灌區(qū)沙壕渠灌域沙壤土為例[D]. 咸陽(yáng):西北農(nóng)林科技大學(xué),2019.
WANG Haifeng. Hyperspectral remote sensing based models for soil moisture and salinity prediction:A case study from sandy loam soil in Shahaoqu district of Hetao irrigation area[D]. Xianyang:Northwest A&F University,2019.
[57] 張智韜,魏廣飛,姚志華,等. 基于無(wú)人機(jī)多光譜遙感的土壤含鹽量反演模型研究[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2019,50(12):151–160.
ZHANG Zhitao,WEI Guangfei,YAO Zhihua,et al. Soil salt inversion model based on UAV multispectral remote sensing[J]. Transactions of the Chinese Society of Agricultural Machinery,2019,50(12):151–160.
[58] 張曦. 基于無(wú)人機(jī)攝影測(cè)量技術(shù)的礦區(qū)沉陷監(jiān)測(cè)方法研究[D]. 邯鄲:河北工程大學(xué),2019.
ZHANG Xi. Study on the method of subsidence monitoring in mining areas based on UAV photogrammetry technology[D]. Handan:Hebei University of Engineering,2019.
[59] 蘆家欣. 基于機(jī)載LiDAR點(diǎn)云的榆神礦區(qū)采煤沉陷建模[D]. 西安:西安科技大學(xué),2020.
LU Jiaxin. Mining subsidence modeling based on airborne LIDAR point cloud in Yushen mining area[D]. Xi’an:Xi’an University of Science and Technology,2020.
[60] 楊秋麗,魏建新,鄭江華,等. 離散點(diǎn)云構(gòu)建數(shù)字高程模型的插值方法研究[J]. 測(cè)繪科學(xué),2019,44(7):16–23.
YANG Qiuli,WEI Jianxin,ZHENG Jianghua,et al. Comparison of interpolation methods of digital elevation model using discrete point cloud data[J]. Science of Surveying and Mapping,2019,44(7):16–23.
[61] MENG Xuelian,CURRIT N,ZHAO Kaiguang. Ground filtering algorithms for airborne LiDAR data:A review of critical issues[J]. Remote Sensing,2010,2(3):833–860.
[62] YU Jinze,WEI Mingqiang,QIN Jing,et al. Feature-preserving mesh denoising via normal guided quadric error metrics[J]. Optics and Lasers in Engineering,2014,62:57–68.
[63] 胡達(dá),黎永索,張可能,等. Kriging點(diǎn)云濾波改進(jìn)算法及監(jiān)測(cè)試驗(yàn)研究[J/OL]. 土木與環(huán)境工程學(xué)報(bào)(中英文),2021:1–12 [2021-03-12]. https: //kns.cnki.net/kcms/detail/50.1218.TU. 20210312.0940.002.html
HU Da,LI Yongsuo,ZHANG Keneng,et al. Improved Kriging point cloud filtering algorithm and monitoring experiment[J]. Journal of Civil and Environmental Engineering,2021:1–12[2021-03-12]. https: //kns.cnki.net/kcms/detail/50.1218.TU. 20210312.0940.002.html
[64] 汪文琪,李宗春,付永健,等. 一種多尺度自適應(yīng)點(diǎn)云坡度濾波算法[J/OL]. 武漢大學(xué)學(xué)報(bào)(信息科學(xué)版),2021:1–8[2021-04-09]. https: //doi.org/10.13203/j.whugis20200016
WANG Wenqi,LI Zongchun,F(xiàn)U Yongjian,et al. A multi-scale adaptive slope filtering algorithm of point cloud[J]. Geomatics and Information Science of Wuhan University,2021:1–8 [2021-04-09]. https: //doi.org/10.13203/j.whugis20200016
[65] 李煉. 機(jī)載LiDAR點(diǎn)云濾波及分類算法研究[D]. 成都:成都理工大學(xué),2014.
LI Lian. Researching the filtering and classification algorithms of airborne LiDAR point cloud[D]. Chengdu:Chengdu University of Technology,2014.
[66] 曲金博. 基于聚類算法的三維激光點(diǎn)云數(shù)據(jù)去噪方法研究[D]. 沈陽(yáng):沈陽(yáng)建筑大學(xué),2020.
QU Jinbo. Study on 3D laser point cloud data denoising method based on clustering algorithm[D]. Shenyang:Shenyang Jianzhu University,2020.
[67] 湯伏全,蘆家欣,韋書平,等. 基于無(wú)人機(jī)LiDAR的榆神礦區(qū)采煤沉陷建模方法改進(jìn)[J]. 煤炭學(xué)報(bào),2020,45(7):2655–2666.
TANG Fuquan,LU Jiaxin,WEI Shuping,et al. Improvement of mining subsidence modeling method based on UAV LiDAR in Yushen mining area[J]. Journal of China Coal Society,2020,45(7):2655–2666.
[68] 戴士杰,任永潮,張慧博. 各向異性擴(kuò)散濾波的三維散亂點(diǎn)云平滑去噪算法[J]. 計(jì)算機(jī)輔助設(shè)計(jì)與圖形學(xué)學(xué)報(bào),2018,30(10):1843–1849.
DAI Shijie,REN Yongchao,ZHANG Huibo. Study on smooth denoising of 3D scattered point clouds with anisotropic diffusion filtering[J]. Journal of Computer Aided Design and Computer Graphics,2018,30(10):1843–1849.
[69] 李鵬飛,吳海娥,景軍鋒,等. 點(diǎn)云模型的噪聲分類去噪算法[J]. 計(jì)算機(jī)工程與應(yīng)用,2016,52(20):188–192.
LI Pengfei,WU Hai’e,JING Junfeng,et al. Noise classification denoising algorithm for point cloud model[J]. Computer Engineering and Applications,2016,52(20):188–192.
[70] XU Xianlei,ZHAO Yanling,HU Zhenqi,et al. Boundary demarcation of the damaged cultivated land caused by coal mining subsidence[J]. Bulletin of Engineering Geology and the Environment,2014,73(2):621–633.
[71] 侯恩科,首召貴,徐友寧,等. 無(wú)人機(jī)遙感技術(shù)在采煤地面塌陷監(jiān)測(cè)中的應(yīng)用[J]. 煤田地質(zhì)與勘探,2017,45(6):102–110.
HOU Enke,SHOU Zhaogui,XU Youning,et al. Application of UAV remote sensing technology in monitoring of coal mining-induced subsidence[J]. Coal Geology & Exploration,2017,45(6):102–110.
[72] 高冠杰,侯恩科,謝曉深,等. 基于四旋翼無(wú)人機(jī)的寧夏羊場(chǎng)灣煤礦采煤沉陷量監(jiān)測(cè)[J]. 地質(zhì)通報(bào),2018,37(12):2264–2269.
GAO Guanjie,HOU Enke,XIE Xiaoshen,et al. The monitoring of ground surface subsidence related to coal seams mining in Yangchangwan coal mine by means of unmanned aerial vehicle with quad-rotors[J]. Geological Bulletin of China,2018,37(12):2264–2269.
[73] ZHOU Dawei,QI Lizhuang,ZHANG Demin,et al. Unmanned Aerial Vehicle(UAV) photogrammetry technology for dynamic mining subsidence monitoring and parameter inversion:A case study in China[J]. IEEE Access,2020,8:16372–16386.
[74] 胡振琪,王新靜,賀安民. 風(fēng)積沙區(qū)采煤沉陷地裂縫分布特征與發(fā)生發(fā)育規(guī)律[J]. 煤炭學(xué)報(bào),2014,39(1):11–18.
HU Zhenqi,WANG Xinjing,HE Anmin. Distribution characteristic and development rules of ground fissures due to coal mining in windy and sandy region[J]. Journal of China Coal Society,2014,39(1):11–18.
[75] 胡青峰,崔希民,袁德寶,等. 厚煤層開(kāi)采地表裂縫形成機(jī)理與危害性分析[J]. 采礦與安全工程學(xué)報(bào),2012,29(6):864–869.
HU Qingfeng,CUI Ximin,YUAN Debao,et al. Formation mechanism of surface cracks caused by thick seam mining and hazard analysis[J]. Journal of Mining & Safety Engineering,2012,29(6):864–869.
[76] 范立民,張曉團(tuán),向茂西,等. 淺埋煤層高強(qiáng)度開(kāi)采區(qū)地裂縫發(fā)育特征:以陜西榆神府礦區(qū)為例[J]. 煤炭學(xué)報(bào),2015,40(6):1442–1447.
FAN Limin,ZHANG Xiaotuan,XIANG Maoxi,et al. Characteristics of ground fissure development in high intensity mining area of shallow seam in Yushenfu coal field[J]. Journal of China Coal Society,2015,40(6):1442–1447.
[77] 施展宇. 地面三維激光掃描技術(shù)在開(kāi)采沉陷應(yīng)用研究[D]. 西安:西安科技大學(xué),2014.
SHI Zhanyu. Terrestrial 3D laser scanning technology application in mining subsidence[D]. Xi’an:Xi’an University of Science and Technology,2014.
[78] 劉文濤. 基于時(shí)序InSAR技術(shù)的礦區(qū)地面沉降監(jiān)測(cè)與分析[D]. 西安:西安科技大學(xué),2020.
LIU Wentao. Monitoring and analysis of land subsidence in mining area based on time series InSAR technology[D]. Xi’an:Xi’an University of Science and Technology,2020.
[79] 謝曉深,侯恩科,高冠杰,等. 寧夏羊場(chǎng)灣煤礦淺埋煤層開(kāi)采地面塌陷發(fā)育規(guī)律及形成機(jī)理[J]. 地質(zhì)通報(bào),2018,37(12):2233–2240.
XIE Xiaoshen,HOU Enke,GAO Guanjie,et al. A study of the development regularity and formation mechanism of ground subsidence in shallow coal seam mining of Yangchangwan coal mine,Ningxia[J]. Geological Bulletin of China,2018,37(12):2233–2240.
[80] 侯恩科,張杰,謝曉深,等. 無(wú)人機(jī)遙感與衛(wèi)星遙感在采煤地表裂縫識(shí)別中的對(duì)比[J]. 地質(zhì)通報(bào),2019,38(2/3):443–448.
HOU Enke,ZHANG Jie,XIE Xiaoshen,et al. Contrast application of unmanned aerial vehicle remote sensing and satellite remote sensing technology relating to ground surface cracks recognition in coal mining area[J]. Geological Bulletin of China,2019,38(2/3):443–448.
[81] 侯恩科,謝曉深,徐友寧,等. 羊場(chǎng)灣煤礦采動(dòng)地裂縫發(fā)育特征及規(guī)律研究[J]. 采礦與巖層控制工程學(xué)報(bào),2020,2(3):037038.
HOU Enke,XIE Xiaoshen,XU Youning,et al. Prediction of ground cracks induced by coal mining in Yangchangwan coal mine[J]. Journal of Mining and Strata Control Engineering,2020,2(3):037038.
[82] 毛崔磊. 黃土丘陵區(qū)采煤塌陷地裂縫分布特征研究[D]. 北京:中國(guó)地質(zhì)大學(xué)(北京),2018.
MAO Cuilei. Study on fracture distribution characteristics of coal mining collapse in loess hilly area[D]. Beijing:China University of Geosciences(Beijing),2018.
[83] 韋博文. 基于改進(jìn)的MF–FDOG算法和無(wú)人機(jī)影像提取黃土地區(qū)地裂縫[D]. 成都:西南交通大學(xué),2018.
WEI Bowen. Extracting ground fissures in loess landform area using modified MF-FDOG algorithm and UAV images[D]. Chengdu:Southwest Jiaotong University,2018.
[84] JAIN R,KASTURI R,SCHUNCK B G. Machine vision[M]. New York:McGraw Hill,1995.
[85] 趙毅鑫,許多,孫波,等. 基于無(wú)人機(jī)紅外遙感和邊緣檢測(cè)技術(shù)的采動(dòng)地裂縫辨識(shí)[J]. 煤炭學(xué)報(bào),2021,46(2):624–637.
ZHAO Yixin,XU Duo,SUN Bo,et al. Investigation on ground fissure identification using UAV infrared remote sensing and edge detection technology[J]. Journal of China Coal Society,2021,46(2):624–637.
[86] SHRUTHI R B V,KERLE N,JETTEN V. Object-based gully feature extraction using high spatial resolution imagery[J]. Geomorphology,2011,134(3/4):260–268.
[87] FENG Quanlong,LIU Jiantao,GONG Jianhua. UAV remote sensing for urban vegetation mapping using random forest and texture analysis[J]. Remote Sensing,2015,7(1):1074–1094.
[88] HARALICK R M,STERNBERG S R,ZHUANG Xinhua. Image analysis using mathematical morphology[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1987,9(4):532–550.
[89] ANDRé S,JEAN-PHILIPPE M,NORMAN K,et al. Image-based mapping of surface fissures for the investigation of landslide dynamics[J]. Geomorphology,2013,186:12–27.
[90] BRUCE G B. Machine Vision Handbook[M]. London:Springer,2012.
[91] FRANCIONI M,SALVINI R,STEAD D,et al. An integrated remote sensing-GIS approach for the analysis of an open pit in the Carrara marble district,Italy:Slope stability assessment through kinematic and numerical methods[J]. Computers and Geotechnics,2015,67:46–63.
[92] MA Guotao,HU Xiewen,YIN Yueping,et al. Failure mechanisms and development of catastrophic rockslides triggered by precipitation and open-pit mining in Emei,Sichuan,China[J]. Recent Landslides,2018,15(7):1401–1414.
[93] LI Qing,MIN Gaochen,CHEN Peng,et al. Computer vision-based techniques and path planning strategy in a slope monitoring system using unmanned aerial vehicle[J]. International Journal of Advanced Robotic Systems,2020,17(2):172988142090430.
[94] 宋誠(chéng). 數(shù)字?jǐn)z影測(cè)量法在邊坡監(jiān)測(cè)中的應(yīng)用研究[D]. 廣州:廣州大學(xué),2014.
SONG Cheng. The application research of digital photogrammetry method in slope monitoring[D]. Guangzhou:Guangzhou University,2014.
[95] 呂愛(ài)鐘,賈曉陽(yáng). 邊坡危險(xiǎn)滑動(dòng)面及穩(wěn)定安全系數(shù)的力學(xué)解析方法[J/OL]. 工程地質(zhì)學(xué)報(bào),2021:1–7[2021-05-13]. https: //doi.org/10.13544/j.cnki.jeg.2020-568
LYU Aizhong,JIA Xiaoyang. Mechanical analysis method for dangerous sliding surface and factor of safety[J]. Journal of Engineering Geology,2021:1–7[2021-05-13]. https: //doi.org/10. 13544/j.cnki.jeg.2020-568
[96] 靳遠(yuǎn)成,趙鵬輝,薄霧,等. 基于無(wú)人機(jī)影像的邊坡精細(xì)化建模及穩(wěn)定性分析[J]. 水利與建筑工程學(xué)報(bào),2020,18(6):178–183.
JIN Yuancheng,ZHAO Penghui,BO Wu,et al. Refined modeling and stability analysis of slope based on UAV images[J]. Journal of Water Resources and Architectural Engineering,2020,18(6):178–183.
[97] 岳西蒙,伍法權(quán),沙鵬,等. 基于三維點(diǎn)云建模的礦山邊坡穩(wěn)定性分析[J]. 中國(guó)礦業(yè),2021,30(4):89–95.
YUE Ximeng,WU Faquan,SHA Peng,et al. Stability analysis of mine slope based on 3D point cloud modeling[J]. China Mining Magazine,2021,30(4):89–95.
[98] 金愛(ài)兵,陳帥軍,趙安宇,等. 基于無(wú)人機(jī)攝影測(cè)量的露天礦邊坡數(shù)值模擬[J]. 巖土力學(xué),2021,42(1):255–264.
JIN Aibing,CHEN Shuaijun,ZHAO Anyu,et al. Numerical simulation of open-pit mine slope based on unmanned aerial vehicle photogrammetry[J]. Rock and Soil Mechanics,2021,42(1):255–264.
Application of unmanned aerial vehicle in surface soil characterization and geological disaster monitoring in mining areas
LONG Linli1,LIU Ying1,2,ZHANG Xuyang1,SU Yongdong1,CHEN Xiaoyang1,2
(1. School of Earth and Environment, Anhui University of Science and Technology, Huainan 232001, China; 2. Anhui Engineering Laboratory for Comprehensive Utilization of Water and Soil Resources and Ecological Protection in High Water Level Mining Areas, Huainan 232001, China)
With the emergence and development of UAVs and the improvement of the miniaturization and intelligence of various sensor, UAVs equipped with sensors have become an efficient tool for obtaining spatial data. Because UAVs are low cost, short revisit period, fast and efficient, light and flexible, simple operation, and high temporal and spatial accuracy of image acquisition, it is widely used in mining land damage monitoring.Using “UAV, Inversion, Soil Monitoring, Surface Collapse, Ground Fissure” as keywords, this paper summarizes the academic papers of the search system in the web of science, CNKI, and Google Scholar from January 2010 to May 2021. Through comparing and analyzing the differences between drone monitoring technology and other detection technologies, the drone monitoring of heavy metals, soil moisture content, and salt content in mining areas is reviewed. The general process and data processing methods of the measurement, surface subsidence, ground fissures and slope stability, and the application prospects of UAVs in surface soil characteristics and geological disaster monitoring in mining areas are summarized.It is believed that in the future, it is possible to integrate field time series tracking investigation, high-precision soil quality monitoring technology, high-spatial resolution drone monitoring technology, digital simulation methods, and test monitoring and analysis of typical working faces to study the coupling relationship between geohazards and soil quality evolution in the dynamic advancement of the working face from the open-off cut to the stop of mining. The coupled relationship is to construct a theoretical system and time series evolution model for the prediction of soil quality evolution in coal mining subsidence areas. This will further explore the relationship between soil quality in mining areas and geological disasters, and propose measures to mitigate, control and improve soil quality in mining areas, providing technical support for the coordinated and sustainable development of coal resource mining and ecological environment in China's coal production bases.
UAV; coal mining area; soil monitoring; geological disaster monitoring
語(yǔ)音講解
TD167
A
1001-1986(2021)06-0200-12
2021-05-28;
2021-08-05
國(guó)家自然科學(xué)基金項(xiàng)目(4157020161);安徽理工大學(xué)校級(jí)重點(diǎn)項(xiàng)目(自然科學(xué)類)(xjzd2020-04)
龍林麗,1998年生,女,四川自貢人,碩士研究生,從事礦山生態(tài)環(huán)境修復(fù)研究. E-mail:long6_6@163.com
陳孝楊,1976年生,男,安徽肥西人,博士,教授,博士生導(dǎo)師,從事礦山環(huán)境治理與場(chǎng)地污染控制研究. E-mail:chenxy@aust.edu.cn
龍林麗,劉英,張旭陽(yáng),等. 無(wú)人機(jī)在礦區(qū)表土特征及地質(zhì)災(zāi)害監(jiān)測(cè)中的應(yīng)用[J]. 煤田地質(zhì)與勘探,2021,49(6):200–211. doi: 10.3969/j.issn.1001-1986.2021.06.024
LONG Linli,LIU Ying,ZHANG Xuyang,et al. Application of unmanned aerial vehicle in surface soil characterization and geological disaster monitoring in mining areas[J]. Coal Geology & Exploration,2021,49(6):200–211. doi: 10.3969/j.issn.1001-1986.2021.06.024
移動(dòng)閱讀
(責(zé)任編輯 周建軍)