蘭玉彬,林澤山,王林琳,鄧小玲
·農(nóng)業(yè)信息與電氣技術(shù)·
基于文獻(xiàn)計(jì)量學(xué)的智慧果園研究進(jìn)展與熱點(diǎn)分析
蘭玉彬1,2,3,4,林澤山1,2,王林琳1,2,鄧小玲1,2,3,4※
(1. 華南農(nóng)業(yè)大學(xué)電子工程學(xué)院、人工智能學(xué)院,廣州 510642; 2. 嶺南現(xiàn)代農(nóng)業(yè)科學(xué)與技術(shù)廣東省實(shí)驗(yàn)室,廣州 510642; 3.國家精準(zhǔn)農(nóng)業(yè)航空施藥技術(shù)國際聯(lián)合研究中心,廣州 510642;4.廣東省智慧農(nóng)業(yè)工程技術(shù)中心,廣州 510642)
為了宏觀掌握智慧果園在國內(nèi)外的研究動(dòng)態(tài)、前沿和熱點(diǎn),更好地推動(dòng)智慧果園乃至智慧農(nóng)業(yè)的發(fā)展,該研究采用文獻(xiàn)計(jì)量分析方法,以Web of science核心論文集為檢索平臺(tái)分析了智慧果園2002年1月1日—2022年8月累計(jì)20年的時(shí)空分布、主要研究?jī)?nèi)容以及前沿?zé)狳c(diǎn)。主要結(jié)論如下:智慧果園的研究自2014年起步入正軌,2018年起在人工智能技術(shù)推動(dòng)下發(fā)展迅猛,2018-2021年總發(fā)文量占比37.5%;總體而言,作者(Lan Yubin、Chen Chao、Tang Yu等)、機(jī)構(gòu)(華南農(nóng)業(yè)大學(xué)、中國農(nóng)業(yè)大學(xué)和佛羅里達(dá)大學(xué)等)、地域(中國、美國、西班牙等國)交流和合作均較為密切;中國、美國是開展智慧果園研究的主要國家,總發(fā)文量共占比58.2%;當(dāng)前主要研究集中在果樹長(zhǎng)勢(shì)和病蟲害識(shí)別和預(yù)警、無人化或智能化農(nóng)機(jī)作業(yè)。根據(jù)研究目的細(xì)分的技術(shù)主要包含人工智能模型/算法、傳感、物聯(lián)和精準(zhǔn)農(nóng)業(yè)等。自2007年以來,研究熱點(diǎn)由柑橘病害、產(chǎn)量預(yù)估等對(duì)象研究逐步過渡到技術(shù)研究上,深度學(xué)習(xí)、無人機(jī)、人工智能的研究是當(dāng)今智慧果園的發(fā)展前沿。智慧果園研究深受技術(shù)推動(dòng)尤其在當(dāng)前人工智能技術(shù)背景下方興未艾,而當(dāng)前的環(huán)境復(fù)雜度高、種植欠規(guī)范等問題依舊制約著其進(jìn)一步發(fā)展。星-空-地立體化果園感知、空-地協(xié)同無人化精準(zhǔn)作業(yè)、水果采摘、果品的可視化溯源等方面將是未來智慧果園主要研究方向。
智能化;自動(dòng)化;智慧果園;量化分析;Web of Science;Citespace;文獻(xiàn)計(jì)量分析
農(nóng)業(yè)在人類歷史上經(jīng)歷了從傳統(tǒng)農(nóng)業(yè)到現(xiàn)代化農(nóng)業(yè)的過程,現(xiàn)代化農(nóng)業(yè)必將是集機(jī)械化、數(shù)字化、信息化和智能化于一體的智慧農(nóng)業(yè)。智慧農(nóng)業(yè)是以信息和知識(shí)為核心要素,通過將互聯(lián)網(wǎng)、數(shù)據(jù)挖掘、云計(jì)算等現(xiàn)代信息技術(shù)與農(nóng)藝深度融合,實(shí)現(xiàn)信息智能感知、智能控制、定量決策、個(gè)性化服務(wù)的農(nóng)業(yè)生產(chǎn)模式,將成為農(nóng)業(yè)信息化農(nóng)業(yè)數(shù)字化的高級(jí)階段[1-2]。
智慧果園是智慧農(nóng)業(yè)的重要一環(huán)。智慧果園旨在通過宜機(jī)化樹形改造[3-4]、星-空-地立體化監(jiān)測(cè)[5-11]、果樹全生長(zhǎng)期精準(zhǔn)管控[12-14]、病蟲害綠色防控[15-18]、空-地協(xié)同無人化作業(yè)[19-23]、云-邊-端協(xié)同智能計(jì)算[24-27]、智慧倉庫管理、可視化溯源[28-30]等技術(shù)體系,實(shí)現(xiàn)穩(wěn)產(chǎn)優(yōu)質(zhì)化、調(diào)控精細(xì)化、省力高效化、綠色生態(tài)化、管理智能化、作業(yè)無人化的現(xiàn)代化農(nóng)業(yè)生產(chǎn)模式。文獻(xiàn)計(jì)量學(xué)已成為對(duì)各個(gè)研究領(lǐng)域進(jìn)行規(guī)律、熱點(diǎn)和趨勢(shì)分析的得力助手。為更好地對(duì)當(dāng)前智慧果園的研究現(xiàn)狀進(jìn)行分析,本文以Web of Science核心合集即累計(jì)20年來收錄的579篇智慧果園相關(guān)研究文獻(xiàn)作為研究對(duì)象,基于文獻(xiàn)計(jì)量學(xué)分析了智慧果園的研究信息上的時(shí)空分布,接著分析智慧果園的主要研究?jī)?nèi)容和前沿?zé)狳c(diǎn),最后依據(jù)分析結(jié)果進(jìn)行總結(jié)并做出展望。
本文以Web of science作為檢索平臺(tái)。智慧果園研究主要集中在龍眼、柑橘、荔枝和桃等重點(diǎn)作物上,為防止遺漏其他果樹作物在檢索詞中補(bǔ)充了“orchard”一詞。為提高檢索效率多次更換檢索詞,最終以“TI=(longan or citrus or litchi or peach or orchard) And AB=(growth or disease or growing or pest or insect or tree or fruit) AND AK=(drone or UAV or AI or intelligent or detection or segmentation or precision or spray or unmanned or robot or sensor or “deep learning” or “machine learning” or “Agricultural machinery”)”作為檢索項(xiàng),將時(shí)間范圍設(shè)置為2002年1月1日至2022年8月31日,共篩選出598篇文獻(xiàn);剔除無關(guān)文獻(xiàn)和經(jīng)去重處理后最終得到579篇文獻(xiàn)。
本文通過文獻(xiàn)計(jì)量方法分析數(shù)據(jù)。利用數(shù)據(jù)處理軟件Excel和文獻(xiàn)計(jì)量分析工具Citespace 對(duì)論文檢索數(shù)據(jù)進(jìn)行量化分析。先通過Excel和Web of science自帶數(shù)據(jù)庫得到2002—2022年的年度發(fā)文量,再利用Citespace和Excel統(tǒng)計(jì)分析工具得到核心作者、機(jī)構(gòu)、地域和關(guān)鍵詞的共線知識(shí)圖譜,對(duì)檢索數(shù)據(jù)進(jìn)行機(jī)構(gòu)合作關(guān)系網(wǎng)絡(luò)、文獻(xiàn)共引分析、高頻詞聚類、關(guān)鍵詞共現(xiàn)、關(guān)鍵詞突現(xiàn)分析[31],借助知識(shí)圖譜梳理、歸納智慧果園的發(fā)展脈絡(luò)及累計(jì)20年間研究地域、機(jī)構(gòu)等空間信息和研究技術(shù)及應(yīng)用熱點(diǎn)。所采用的的計(jì)量文獻(xiàn)分析工具Citespace通過直觀的知識(shí)圖譜方式,展現(xiàn)研究領(lǐng)域的熱點(diǎn)關(guān)鍵詞、研究進(jìn)展和前沿方向,其在科學(xué)和技術(shù)領(lǐng)域得到廣泛應(yīng)用[32-34]。
對(duì)Web of science收集的2002年到2022年的579篇文獻(xiàn)進(jìn)行發(fā)文時(shí)間的統(tǒng)計(jì)(見圖1),統(tǒng)計(jì)數(shù)據(jù)顯示:1)2002年至2013年的每年論文發(fā)表量均不超過20篇,且發(fā)文量有所波動(dòng),12年間發(fā)文總量?jī)H占9.33%;2)2014至2021的8年間的論文發(fā)表量年增長(zhǎng)較大,其中2019年年度發(fā)文增長(zhǎng)量(19篇)和年增長(zhǎng)率均達(dá)到了最高(41.5%),而后年度發(fā)文量繼續(xù)增長(zhǎng)但增長(zhǎng)率有所下降;3)2018年—2021年發(fā)文總量達(dá)到216篇,占比37.3%,而2022截至8月31日為止,該年度的發(fā)文量為67篇。該統(tǒng)計(jì)數(shù)據(jù)體現(xiàn)了智慧果園的研究從2018年開始進(jìn)入了快速發(fā)展階段,這與人工智能第三次發(fā)展浪潮的時(shí)間節(jié)點(diǎn)高度吻合,意味著當(dāng)前的智慧果園的快速發(fā)展主要得益于人工智能技術(shù)的推動(dòng)[35-37]。
圖1 Web of science年度發(fā)文量統(tǒng)計(jì)
2.2.1 地域和機(jī)構(gòu)分布
采用Citespace對(duì)文獻(xiàn)發(fā)布的地域分布進(jìn)行分析,時(shí)間切片為1 a,將Node Types設(shè)置為“Country”,Pruning選擇“Pathfinder”和“Pruning sliced networks”,其余參數(shù)保持默認(rèn)值,得到節(jié)點(diǎn)數(shù)為64,連線數(shù)117,密度為0.058的國家共線知識(shí)圖譜,如圖2a所示。知識(shí)圖譜上的連線表示彼此之間存在交流與合作,由圖可知,節(jié)點(diǎn)之間連線較為緊密,基本上沒有孤立的節(jié)點(diǎn),可見中國、美國、巴西、西班牙、意大利等國聯(lián)系相對(duì)緊密,存在較多合作,而由節(jié)點(diǎn)的大小,發(fā)現(xiàn)超過一半的研究集中于中國(181篇)和美國(156篇),兩國的發(fā)文總量占比58.2%,其次主要在西班牙、意大利、印度,分別有61篇、41篇和28篇。
注:節(jié)點(diǎn)大小表示論文發(fā)表量,節(jié)點(diǎn)越大,發(fā)文量越多。
將節(jié)點(diǎn)設(shè)置為“Institution”,得到機(jī)構(gòu)的共線知識(shí)圖譜,如圖2b所示。通過知識(shí)圖譜不難發(fā)現(xiàn),盡管圖譜中均存在少數(shù)分散的節(jié)點(diǎn),但基本上形成了一個(gè)較為整體的網(wǎng)絡(luò),可見當(dāng)前的智慧果園中大部分機(jī)構(gòu)之間的交流與合作均較為密切,如華南農(nóng)業(yè)大學(xué)、中國農(nóng)業(yè)大學(xué)、佛羅里達(dá)大學(xué)之間存在交流與合作,華南農(nóng)業(yè)大學(xué)與嶺南現(xiàn)代農(nóng)業(yè)科學(xué)與技術(shù)廣東省實(shí)驗(yàn)室和仲愷農(nóng)業(yè)工程學(xué)院之間、佛羅里達(dá)大學(xué)與華中農(nóng)業(yè)大學(xué)、農(nóng)業(yè)農(nóng)村部和西北農(nóng)林科技大學(xué)之間的交流較為直接。
表1為論文發(fā)表量前15的機(jī)構(gòu),由表可知:發(fā)文量前15的機(jī)構(gòu)中,有8個(gè)機(jī)構(gòu)來自中國,6個(gè)來自美國,其總發(fā)文量在前15機(jī)構(gòu)中分別占比22.72%和36.15%。結(jié)合地域分析,該統(tǒng)計(jì)數(shù)據(jù)體現(xiàn)出美國在智慧果園方面的研究更多集中于排名前2的佛羅里達(dá)州立大學(xué)和佛羅里達(dá)大學(xué),分別為66篇和65篇,而中國除了排名第3的華南農(nóng)業(yè)大學(xué)研究較多(44篇)之外,在其他機(jī)構(gòu)的分布較為均衡。
表1 國際發(fā)表論文數(shù)量排名前15的機(jī)構(gòu)
2.2.2 核心作者分布
將節(jié)點(diǎn)設(shè)置為“Author”,其余參數(shù)保持默認(rèn)值,分別得到作者共線知識(shí)圖譜(見圖3所示)。節(jié)點(diǎn)和線條的顏色深度由淺到深表示時(shí)間年份的遠(yuǎn)近程度,越深則表示越近期。由圖可知:1)大部分作者之間的交流與合作較為密切,如:作者方面,Lan Y B、Chen C、Tang Y、Li J之間存在交流與合作,而Lan Y B、Deng X L,Zheng Z之間,Tang Y、Wang H、Zhuang J之間合作較為直接和緊密;2)Lan Y B、Tang Y、Li J、Wang X等在近年來的研究較為突出。
表2為論文發(fā)表量前15的作者,由表可知:1)論文發(fā)表量前3的分別為L(zhǎng)an Yubin(14篇)、Lee Won Suk(13篇)、Xiong Juntao(11篇);2)結(jié)合機(jī)構(gòu)分析,發(fā)文量前15的作者中,有14位所在機(jī)構(gòu)來自發(fā)文量前15的機(jī)構(gòu)中的4個(gè)機(jī)構(gòu):華南農(nóng)業(yè)大學(xué)、佛羅里達(dá)大學(xué)、華盛頓州立大學(xué)、仲愷農(nóng)業(yè)工程學(xué)院、北京農(nóng)林科學(xué)院;3)發(fā)文量前15的作者中,有9人單位所在地是中國,5人單位所在地是美國。
注:節(jié)點(diǎn)和線條的顏色深度由淺到深表示時(shí)間年份的遠(yuǎn)近程度,越深則表示越近期。
表2 國際發(fā)表論文數(shù)量排名前15的作者
2.2.3 出版期刊分布
基于Web of science統(tǒng)計(jì)文獻(xiàn)出版來源,發(fā)文量前15的分布情況如表3所示。由圖可得,累計(jì)20年來《Computers and Electronics in Agriculture》、《Sensors》、《Remote Sensing》、《Precision Agriculture》和《Frontiers in Plant Science》等期刊在智慧果園研究方面均有不少發(fā)文,而《Computers and Electronics in Agriculture》的發(fā)文量達(dá)到75篇(占據(jù)12.89%),排名第1并遙遙領(lǐng)先于其他出版期刊。
表3 刊發(fā)智慧果園主題論文數(shù)量前15名的期刊
將節(jié)點(diǎn)設(shè)置為“keyword”,其他參數(shù)不變,得到節(jié)點(diǎn)數(shù)為526,連線數(shù)1 387,密度為0.01的關(guān)鍵詞共線聚類圖譜,其模塊值和平均輪廓值分別為0.689和0.849(見圖4)。當(dāng)>0.3時(shí),表明圖譜結(jié)構(gòu)較為成型,而當(dāng)>0.7時(shí),聚類效果較為理想。顯示12個(gè)聚類區(qū)域,每個(gè)區(qū)域分別0~11的數(shù)字標(biāo)簽,數(shù)字標(biāo)簽越小則包含的關(guān)鍵詞越多,聚類中心區(qū)域?yàn)檠芯恐攸c(diǎn)[38]。依據(jù)標(biāo)簽聚類結(jié)果:1)研究?jī)?nèi)容從大體上可分為兩大類:研究對(duì)象(#0柑橘黃龍病,#1水果檢測(cè),#5葉面噴灑,#7油橄欖,#8天敵群落)和研究技術(shù)(#2 機(jī)器視覺,#3風(fēng)助式噴霧器,#4 精準(zhǔn)農(nóng)業(yè),#6重采樣驗(yàn)證,#9機(jī)器學(xué)習(xí),#10遙感,#11表面增強(qiáng)拉曼散射);2)機(jī)器學(xué)習(xí)、精準(zhǔn)農(nóng)業(yè)、機(jī)器視覺是當(dāng)前在智慧果園領(lǐng)域的重點(diǎn)研究技術(shù),而水果檢測(cè)、柑橘黃龍病是重點(diǎn)研究對(duì)象。
基于對(duì)數(shù)似然比計(jì)算(Log-likelihood Ratio,LLR)的關(guān)鍵詞聚類結(jié)果統(tǒng)計(jì)如表4所示。結(jié)合圖4和表4進(jìn)行分析可得,當(dāng)前研究對(duì)象可大致分為長(zhǎng)勢(shì)監(jiān)測(cè)、無人作業(yè)2類。長(zhǎng)勢(shì)監(jiān)測(cè)研究較多的有病蟲害研究(聚類#0、#2、#7、#10)、果實(shí)的檢測(cè)(聚類#2),病蟲害研究如柑橘黃龍病不僅包含了其表面特征的檢測(cè)[39-41],還有多光譜條件下的檢測(cè)和引發(fā)黃龍病的媒介的檢測(cè)[42-44],而果實(shí)的檢測(cè)需要考慮到識(shí)別、定位、高度等問題,主要用于產(chǎn)量預(yù)估[45-46]和后續(xù)的機(jī)械采摘[47-49];無人作業(yè)研究較多的有果園的施肥灌溉(聚類#5、#7、#10)、噴灑和施藥(聚類#3、#8)等,其中包含了果樹水分脅迫[50-51]、灌溉方式對(duì)果園的影響[52-53]、噴灑和施藥時(shí)的霧滴漂移和沉積分布[54-56]、噴霧器的設(shè)置及性能分析[57-58]等。
注:節(jié)點(diǎn)和連線分別代表對(duì)應(yīng)聚類區(qū)域的關(guān)鍵詞及其關(guān)聯(lián)。
表4 基于對(duì)數(shù)似然比計(jì)算的關(guān)鍵詞聚類結(jié)果
當(dāng)前智慧果園研究技術(shù)主要集中在算法(聚類#0、#1、#2、#4、#9、#10)、傳感(聚類#4、#10)、精準(zhǔn)農(nóng)業(yè)(聚類#3#4#7)、無損檢測(cè)(聚類#0、#2、#11)等多個(gè)方面,前三者的研究居多。算法方面的研究主要為基于雷達(dá)[59-60]、機(jī)器學(xué)習(xí)[61-62]、深度學(xué)習(xí)[63-65]、圖像處理等方面的識(shí)別和檢測(cè);傳感的研究主要為無線傳感器[66]、地理信息系統(tǒng)、遙感[67-68]、圖像分辨率等;而精準(zhǔn)農(nóng)業(yè)在人工智能的基礎(chǔ)上還結(jié)合了農(nóng)學(xué)、光學(xué)成像、作物科學(xué)等學(xué)科領(lǐng)域,如光譜、植被指數(shù)[69-70]、葉面積指數(shù)[71-72]等參數(shù);無損檢測(cè)的研究則更傾向于光學(xué)、化學(xué)、環(huán)境科學(xué)等學(xué)科領(lǐng)域。
本研究采用Citespace進(jìn)行共被引分析,以獲取智慧果園的研究脈絡(luò)和前言,再進(jìn)行關(guān)鍵詞突現(xiàn)分析某一段時(shí)間內(nèi)的研究熱點(diǎn)。通過Citespace進(jìn)行共被引分析獲取引用頻次圖譜結(jié)果如圖5所示。根據(jù)圖譜聚類結(jié)果以及作者發(fā)文時(shí)間總結(jié)如下:1)由標(biāo)簽和共被引時(shí)間得出,在2010年以前,由于該領(lǐng)域研究較少,技術(shù)也亟待發(fā)展,并沒有熱點(diǎn)出現(xiàn)。2010—2022年可大致分為3個(gè)階段,階段Ⅰ為2010—2014年(聚類#15、#16、#14、#19),該階段也出現(xiàn)圖像處理、精準(zhǔn)農(nóng)業(yè)、果園病蟲害(如柑橘病害)的研究;階段Ⅱ?yàn)?015—2017年(聚類#12、#13、#16、#14、#18),該階段同樣有圖像處理和精準(zhǔn)農(nóng)業(yè),而深度學(xué)習(xí)和無人機(jī)開始在果園中開展應(yīng)用;階段Ⅲ為2018—2022年(聚類#12、#13、#16、#17、#18),該階段還是有圖像處理和精準(zhǔn)農(nóng)業(yè),而該時(shí)期深度學(xué)習(xí)和無人機(jī)出現(xiàn)頻次更多,同時(shí)出現(xiàn)了“人工智能”方面的詞匯。2)由標(biāo)簽序號(hào)和圖譜位置得深度學(xué)習(xí)、精準(zhǔn)農(nóng)業(yè)、無人機(jī)是較為突出的研究熱點(diǎn),在時(shí)期Ⅲ的論文發(fā)表數(shù)量增量也遠(yuǎn)多于其他時(shí)期。
注:節(jié)點(diǎn)和連線分別代表對(duì)應(yīng)聚類區(qū)域的作者(年份)及其關(guān)聯(lián)。
進(jìn)一步進(jìn)行共被引突現(xiàn)分析,得到11篇高突現(xiàn)值的被引文獻(xiàn),如表5所示。結(jié)合圖5和表5,階段Ⅰ和階段Ⅱ一共只有2篇高被引文獻(xiàn)[73-74],均為柑橘黃龍病相關(guān)文獻(xiàn),其中引用強(qiáng)度最高的是Gottwald TR的《Current Epidemiological Understanding of Citrus Huanglongbing》,該文從地域來源、病媒種群、進(jìn)化的角度、和受感染后的影響對(duì)柑橘黃龍病的特征進(jìn)行分析;階段Ⅲ有9篇高被引文獻(xiàn)[75-83],均為算法類文章,涉及了深度學(xué)習(xí)、深度相機(jī)、無人機(jī)和多光譜圖像的研究與應(yīng)用,其中引用強(qiáng)度最高的2篇分別是He Kaiming的《Deep Residual Learning for Image Recognition》和Tian Yunong的《Apple detection during different growth stages in orchards using the improved YOLO-V3 model》,前者提出了殘差學(xué)習(xí)框架以簡(jiǎn)化更深的網(wǎng)絡(luò)的訓(xùn)練,分別在ImageNet測(cè)試集上和COCO對(duì)象檢測(cè)數(shù)據(jù)集上實(shí)現(xiàn)了3.57%的誤差和28%的改進(jìn),而后者提出了一種通過DenseNet方法處理的改進(jìn)的YOLO-V3網(wǎng)絡(luò),該網(wǎng)絡(luò)實(shí)現(xiàn)了對(duì)果園3個(gè)不同生長(zhǎng)階段、高分辨率圖像中以及遮擋和重疊條件下的蘋果實(shí)時(shí)檢測(cè)。
通過Citespace進(jìn)行關(guān)鍵詞突現(xiàn)分析,獲得8個(gè)聚類主題詞,如表6所示。依據(jù)關(guān)鍵詞突現(xiàn)分析可分為另外3個(gè)時(shí)期:2007—2014年的研究熱點(diǎn)是柑橘作物和果園噴霧器,其中該時(shí)期的柑橘作物研究為較為傳統(tǒng)的圖像處理和高光譜儀器在病蟲害上的檢測(cè)、產(chǎn)量預(yù)估的研究和柑橘果樹灌溉技術(shù)等方面的使用;2015—2017年的研究熱點(diǎn)是果園管理、病蟲害和圖像處理,該時(shí)期主要圍繞果園的病蟲害和長(zhǎng)勢(shì)進(jìn)行識(shí)別和管控;2018—2022年更多地利用無人機(jī)和機(jī)器學(xué)習(xí)特別是深度學(xué)習(xí)技術(shù),其引用強(qiáng)度最高(11),遠(yuǎn)高于其余7個(gè)主題詞的平均值(3.55)。
表5 高突現(xiàn)值的被引文獻(xiàn)
表6 高突現(xiàn)值的關(guān)鍵詞
綜合共被引分析和關(guān)鍵詞突現(xiàn)分析可知,在2007年之前并沒有熱點(diǎn)出現(xiàn),2007年之后的16年間的可分為2007—2014年、2015—2017年、2018—2022年3個(gè)階段,在這3個(gè)階段中隨著時(shí)代發(fā)展和技術(shù)革新其研究熱點(diǎn)由對(duì)象研究(柑橘病害、產(chǎn)量預(yù)估)逐步過渡到技術(shù)研究上,其中精準(zhǔn)農(nóng)業(yè)、圖像處理是貫穿了3個(gè)階段的研究熱點(diǎn),深度學(xué)習(xí)、無人機(jī)是該研究的進(jìn)一步深入,人工智能的第三次浪潮推動(dòng)了智慧果園研究的快速發(fā)展。
本文通過Web of science核心合集檢索2002年1月1日-2022年8月31日累積20年來智慧果園領(lǐng)域的文獻(xiàn),基于Citespace文獻(xiàn)計(jì)量分析軟件和Excel軟件統(tǒng)計(jì)智慧果園的研究信息,分析主要研究?jī)?nèi)容和前言熱點(diǎn),主要結(jié)論如下:
1)智慧果園的研究自2014年起步,在2018年后高速發(fā)展,至2021年的4年間發(fā)文總量占比為37.5%,其中2019年增長(zhǎng)量和增長(zhǎng)率均達(dá)到最高值,而后保持增勢(shì)但有所放緩;
2)整體而言,作者(Lan Yubin、Chen Chao、Tang Yu、Li Jun等)、機(jī)構(gòu)(華南農(nóng)業(yè)大學(xué)、中國農(nóng)業(yè)大學(xué)和佛羅里達(dá)大學(xué))、地域(中國、美國、西班牙意大利等國)均存在較為密切的交流和合作,其中中美的總發(fā)文量超過了其他國家和地區(qū)的總和;
3)發(fā)文量前15的機(jī)構(gòu)中,分別有8個(gè)來自中國,6個(gè)來自美國,美國的佛羅里達(dá)州立大學(xué)和佛羅里達(dá)大學(xué)排名前2,中國的華南農(nóng)業(yè)大學(xué)排名第3;發(fā)文量前15的作者中,分別有9位來自中國、5位來自美國,其中Lan Yubin、Lee Won Suk、Xiong Juntao發(fā)文量分列前3;
4)《Computers and Electronics in Agriculture》、《Sensors》、《Remote Sensing》等出版期刊均有較多發(fā)文,其中《Computers and Electronics in Agriculture》發(fā)文量最多,占比12.89%;
5)當(dāng)前主要研究可分為研究對(duì)象和研究技術(shù)兩大類。研究對(duì)象中以長(zhǎng)勢(shì)監(jiān)測(cè)和無人作業(yè)為主,長(zhǎng)勢(shì)監(jiān)測(cè)以柑橘黃龍病和果實(shí)檢測(cè)居多,無人作業(yè)以施肥灌溉、噴藥居多;研究技術(shù)主要有算法、傳感、精準(zhǔn)化作業(yè)和無損檢測(cè)等,前三者居多;
6)自2007年以來,可大致分為3個(gè)階段,在此過程中研究熱點(diǎn)由柑橘病害、產(chǎn)量預(yù)估等對(duì)象研究逐步過渡到技術(shù)研究,其中精準(zhǔn)農(nóng)業(yè)、圖像處理是持續(xù)性研究熱點(diǎn),深度學(xué)習(xí)、無人機(jī)、人工智能的研究是2018年至今智慧果園的發(fā)展前沿。
隨著人工智能、大數(shù)據(jù)、無人機(jī)、無人車等前沿技術(shù)的成熟與發(fā)展,未來多學(xué)科交叉融合的研究熱點(diǎn)將更加突出。展望未來,智慧果園技術(shù)的研究將更注重農(nóng)事生產(chǎn)實(shí)際需要并服務(wù)于農(nóng)業(yè)生產(chǎn)過程。該領(lǐng)域的發(fā)文質(zhì)量將取決于與農(nóng)事生產(chǎn)環(huán)節(jié)的結(jié)合程度和實(shí)際應(yīng)用效益。
從研究熱點(diǎn)上展望智慧果園的主要方向如下:
1)星空地立體化果園感知技術(shù)
衛(wèi)星遙感技術(shù)可實(shí)現(xiàn)區(qū)域級(jí)果樹種植分布的分析,有利于政府對(duì)水果產(chǎn)業(yè)的扶持調(diào)控以及果品市場(chǎng)價(jià)格調(diào)控等;無人機(jī)遙感方式可替代人工巡園,大大提高巡園效率,減少勞動(dòng)力的投入;近地觀測(cè)通過多種固定或移動(dòng)攝像頭、氣象站、土壤墑情傳感器等,局部掌握果樹長(zhǎng)勢(shì)和病蟲害情況,長(zhǎng)期不間斷的監(jiān)測(cè),有利于果樹的精準(zhǔn)植保和按需作業(yè),是實(shí)現(xiàn)智慧果園精準(zhǔn)管控的主要依據(jù)。此外,多源異構(gòu)數(shù)據(jù)的融合,將提高多源遙感系統(tǒng)感知和決策的快速性和準(zhǔn)確性。該方向的發(fā)文量預(yù)計(jì)呈爆發(fā)性增長(zhǎng)。
2)空-地協(xié)同無人化精準(zhǔn)作業(yè)技術(shù)
地面智能農(nóng)機(jī)和農(nóng)用無人機(jī)的協(xié)同作業(yè),將是未來智慧果園中代替人工勞動(dòng)的主要農(nóng)事作業(yè)方式。其中,農(nóng)用無人機(jī)可以在智慧果園中從事施藥、施肥、吹花、授粉、采摘、剪枝等多種農(nóng)事操作。智慧果園的精準(zhǔn)管控技術(shù),有利于實(shí)現(xiàn)農(nóng)事操作的按需和精準(zhǔn)作業(yè)。利用遙感技術(shù)生成作業(yè)處方圖信息,結(jié)合變量控制技術(shù)和精準(zhǔn)導(dǎo)航技術(shù),可實(shí)現(xiàn)果樹的對(duì)靶和精準(zhǔn)作業(yè)。該方向研究的熱點(diǎn)主要以農(nóng)用無人機(jī)為研究對(duì)象。
3)水果采摘技術(shù)
自動(dòng)采摘機(jī)器人是智慧果園無人化程度的最佳體現(xiàn),對(duì)于樹體較高的果樹,未來期待無人機(jī)采摘方式在智慧果園的廣泛應(yīng)用。集成計(jì)算機(jī)視覺、導(dǎo)航、平衡、操縱、感知和機(jī)械技術(shù)的無人機(jī)采摘方式,目前在蘋果、梨等果實(shí)采摘中已取得了突破性進(jìn)展,但對(duì)于樹體較高且枝條較硬的嶺南水果如荔枝、龍眼等果樹,無人化的采摘難度大??偟膩碚f,無人機(jī)采摘技術(shù)是趨勢(shì),但目前尚處于研究階段,離大面積應(yīng)用尚有一段較長(zhǎng)的距離。未來在該方向上的發(fā)文量有望呈上升趨勢(shì)。
4)果品的可視化溯源技術(shù)
消費(fèi)者對(duì)水果質(zhì)量和安全日益關(guān)注,對(duì)果品的生產(chǎn)環(huán)節(jié)的可視化、可溯源性提出了較高的需求。區(qū)塊鏈、二維碼等溯源技術(shù)是鏈接消費(fèi)者和生產(chǎn)環(huán)節(jié)的紐帶,未來將作為智慧果園的重要技術(shù)環(huán)節(jié),但該方向未來研究發(fā)文量仍較少,但其技術(shù)將愈加成熟,應(yīng)用將愈加廣泛。
[1] 趙春江. 智慧農(nóng)業(yè)發(fā)展現(xiàn)狀及戰(zhàn)略目標(biāo)研究[J]. 智慧農(nóng)業(yè),2019,1(1):1-7.
Zhao Chunjiang. State-of-the-art and recommended developmental strategic objectivs of smart agriculture[J]. Smart Agriculture, 2019, 1(1): 1-7. (in Chinese with English abstract)
[2] 李道亮,李震. 無人農(nóng)場(chǎng)系統(tǒng)分析與發(fā)展展望[J]. 農(nóng)業(yè)機(jī)械學(xué)報(bào),2020,51(7):1-12.
Li Daoliang, Li Zhen. System analysis and development prospect of unmanned farming[J]. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(7): 1-12. (in Chinese with English abstract)
[3] 牛自勉,蔚露,李志強(qiáng),等. 蘋果樹形結(jié)構(gòu)變化對(duì)果園藥肥減施的影響[J]. 果樹資源學(xué)報(bào),2021,2(2):15-20.
Niu Zimian, Wei Lu, Li Zhiqiang, et al. Effects of apple tree structure on the reductions in pesticide and fertilizer in the orchard[J]. Journal of Fruit Resources, 2021, 2(2): 15-20. (in Chinese with English abstract)
[4] 張?zhí)煊?,常乃軍,叢培建,? 榮成市蘋果高光效樹形改造及其配套技術(shù)總結(jié)[J]. 煙臺(tái)果樹,2021(2):31-32.
[5] 劉羽. 基于衛(wèi)星遙感的蘋果果園分類和估產(chǎn)模型研究[D].西安:西安科技大學(xué),2021.
Liu Yu. Research on Apple Orchard Classification and Yield Estimation Model Based on Satellite Remote Sensing[D]. Xi'an:Xi'an University of Technology, 2021. (in Chinese with English abstract)
[6] 萬祖毅. 基于無人機(jī)遙感的柑橘果樹信息提取及應(yīng)用研究[D]. 重慶:西南大學(xué),2020.
Wan Zuyi. Extraction and Application of Citrus Fruit Tree Information Based on UAV Remote Sensing[D]. Chongqing:Southwest University, 2020. (in Chinese with English abstract)
[7] Tian H, Fang X, Lan Y, et al. Extraction of citrus trees from UAV remote sensing imagery using YOLOv5s and coordinate transformation[J]. Remote Sensing, 2022, 14(17): 4208.
[8] Zhu Y, Zhou J, Yang Y, et al. Rapid target detection of fruit trees using UAV imaging and improved light YOLOv4 algorithm[J]. Remote Sensing, 2022, 14(17): 4324.
[9] 殷獻(xiàn)博,鄧小玲,蘭玉彬,等. 基于改進(jìn)YOLOX-Nano算法的柑橘梢期長(zhǎng)勢(shì)智能識(shí)別[J/OL]. [2022-04-27]. 華南農(nóng)業(yè)大學(xué)學(xué)報(bào),2023(1):1-21.
Yin Xianbo, Deng Xiaoling, Lan Yubin, et al. Intelligent recognition of citrus shoot growth based on improved YOLOX-Nano algorithm[J/OL]. [2022-04-27]. Journal of South China Agricultural University, 2023(1): 1-21. https://kns.cnki.net/kcms/detail/detail.aspx?dbcode=CAPJ&dbname=CAPJLAST&filename=HNNB20220602003&uniplatform=NZKPT&v=lgXkAcVDmjQYjaifNS05GA30XcwR25Y9wouGKrQHTJURvb2g6QdEIcBua6qaK5_M. (in Chinese with English abstract)
[10] Hobart M, Pflanz M, Weltzien C, et al. Growth height determination of tree walls for precise monitoring in apple fruit production using UAV photogrammetry[J]. Remote Sensing, 2020, 12(10): 1656.
[11] Tariq S, Hakim A, Siddiqi A A, et al. An image dataset of fruitfly species (Bactrocera zonata and bactrocera dorsalis) and automated species classification through object detection[J]. Data in Brief, 2022(43): 108366
[12] Sabzi S, Nadimi M, Abbaspour-Gilandeh Y, et al. Non-destructive estimation of physicochemical properties and detection of ripeness level of Apples using machine vision[J]. International Journal of Fruit Science, 2022, 22(1): 628-645.
[13] Juillion P, Lopez G, Fumey D, et al. Shading apple trees with an agrivoltaic system: Impact on water relations, leaf morphophysiological characteristics and yield determinants[J]. Scientia Horticulturae, 2022(306): 111434.
[14] Xie B, Chen Y, Cheng C, et al. Long-term soil management practices influence the rhizosphere microbial community structure and bacterial function of hilly apple orchard soil[J]. Applied Soil Ecology, 2022, 180: 104627.
[15] 焦梓洲. 生物防治技術(shù)在農(nóng)業(yè)病蟲害防治中的應(yīng)用分析[J]. 種子科技,2020,38(3):85-86.
[16] 郝宇佳,李廣宇,徐鋒. 果園保護(hù)性耕作技術(shù)對(duì)比傳統(tǒng)種植探究[J]. 廣東蠶業(yè),2020,54(4):55-56.
Hao Yujia, Li Guangyu, Xu Feng. Conservation tillage technology in orchard compared with traditional planting[J]. Guangdong Sericulture, 2020, 54(4): 55-56. (in Chinese with English abstract)
[17] 百度百科. 林下經(jīng)濟(jì)[EB/OL]. (2022-01-01) [2022-04-27].https://baike.baidu.com/item/%E6%9E%97%E4%B8%8B%E7%BB%8F%E6%B5%8E?fromModule=lemma_search-box.
[18] 曹裕松,胡春琴,蘇啟陶,等. 生態(tài)學(xué)視野下的林下經(jīng)濟(jì)[J]. 廣西林業(yè)科學(xué),2022,51(1):142-146.
Cao Yusong, Hu Chunqin, Su Qihao, et al. Under-forest economy in ecological perspective[J]. Guangxi Forestry Science, 2022, 51(1): 142-146. (in Chinese with English abstract)
[19] Li S, Cui C. The effect of bait air broadcasting by unmanned aerial vehicles on the ant community diversity[J]. Journal of Applied Entomology, 2022:1-8.
[20] 王昌陵,何雄奎,曾愛軍,等. 基于仿真果園試驗(yàn)臺(tái)的植保無人機(jī)施藥霧滴飄移測(cè)試方法與試驗(yàn)[J]. 農(nóng)業(yè)工程學(xué)報(bào),2020,36(13):56-66.
Wang Changling, He Xiongkui, Zeng Aijun, et al. Measuring method and experiment on spray drift of chemicals applied by UAV sprayer based on an artificial orchard test bench[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2020, 36(13): 56-66. (in Chinese with English abstract)
[21] Fei Z, Vougioukas S. Row-sensing templates: A generic 3D sensor‐based approach to robot localization with respect to orchard row centerlines[J]. Journal of Field Robotics, 2022, 39: 712-738.
[22] Suo R, Gao F, Zhou Z, et al. Improved multi-classes kiwifruit detection in orchard to avoid collisions during robotic picking[J]. Computers and Electronics in Agriculture, 2021, 182: 106052.
[23] 蔡舒平,孫仲鳴,劉慧,等. 基于改進(jìn)型 YOLOv4 的果園障礙物實(shí)時(shí)檢測(cè)方法[J]. 農(nóng)業(yè)工程學(xué)報(bào),2021,37(2):36-43.
Cai Shuping,Sun Zhongming,Liu Hui,et al. Real-time detection methodology for obstacles in orchards using improved YOLOv4[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2021, 37(2): 36-43. (in Chinese with English abstract)
[24] 張文利,陳開臻,劉鈺昕,等. 基于邊緣設(shè)備的輕量化小目標(biāo)果實(shí)檢測(cè)模型[J]. 中國農(nóng)業(yè)信息,2021,33(1):28-36.
Zhang Wenli, Chen Kaizhen, Liu Yuxin, et al. Lightweight small target fruit detection model based on edge devices[J]. China Agricultural Informatics, 2021, 33(1): 28-36. (in Chinese with English abstract)
[25] 焦合軍,周萬春,施進(jìn)發(fā),等. 基于云計(jì)算的果園移動(dòng)機(jī)器人動(dòng)態(tài)路徑規(guī)劃[J]. 機(jī)械設(shè)計(jì)與研究,2021,37(2):45-49,65.
Jiao Hejun, Zhou Wanchun, Shi Jinfa, et al. Dynamic path planning of orchard mobile robot based on cloud computing[J]. Machine Design & Research, 2021, 37(2): 45-49, 65. (in Chinese with English abstract)
[26] Wang R, Li Z, Li J, et al. Research on orchard water and fertilizer automatic control system based on service composition[C]//International Conference on Image Processing and Intelligent Control (IPIC 2021). Lanzhou, 2021, 11928: 228-231.
[27] Kalyani Y, Collier R. A systematic survey on the role of cloud, fog, and edge computing combination in smart agriculture[J]. Sensors, 2021, 21(17): 5922.
[28] 溫晗秋子,鄭永強(qiáng),劉楊,等. 柑橘大數(shù)據(jù)研究與應(yīng)用[J]. 農(nóng)業(yè)大數(shù)據(jù)學(xué)報(bào),2021,3(1):33-44.
Wen-Han Qiuzi, Zheng Yongqiang, Liu Yang, et al. Research and application of citrus big data[J]. Journal of Agricultural Big Data, 2021, 3(1): 33-44. (in Chinese with English abstract)
[29] Ruggiero L, Amalfitano C, Di Vaio C, et al. Use of near-infrared spectroscopy combined with chemometrics for authentication and traceability of intact lemon fruits[J]. Food Chemistry, 2022, 375: 131822.
[30] Graci S, Olivieri F, Barone A. Phenotypic and genotypic characterization to authenticate and trace a high-quality yellow tomato ecotype through the processing chain[J]. Scientia Horticulturae, 2022, 306: 111449.
[31] 李繼宇,胡瀟丹,蘭玉彬,等. 基于文獻(xiàn)計(jì)量學(xué)的 2001-2020 全球農(nóng)用無人機(jī)研究進(jìn)展[J]. 農(nóng)業(yè)工程學(xué)報(bào), 2021,37(9):328-339.
Li Jiyu, Hu Xiaodan, Lan Yubin, et al. Research advance on worldwide agricultural UAVs in 2001-2020 based on bibliometrics[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2021, 37(9): 328-339. (in Chinese with English abstract)
[32] Azam A, Ahmed A, Wang H, et al. Knowledge structure and research progress in wind power generation (WPG) from 2005 to 2020 using CiteSpace based scientometric analysis[J]. Journal of Cleaner Production, 2021, 295: 126496.
[33] 袁媛,白中科. 基于文獻(xiàn)計(jì)量分析的生態(tài)系統(tǒng)恢復(fù)力研究進(jìn)展[J]. 農(nóng)業(yè)工程學(xué)報(bào),2020,36(3):271-278.
Yuan Yuan, Bai Zhongke. Research progress of ecosystem resilience based on bibliometric investigation[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2020, 36(3): 271-278. (in Chinese with English abstract)
[34] Wang X, Zhang Y, Zhang J, et al. Progress in urban metabolism research and hotspot analysis based on CiteSpace analysis[J]. Journal of Cleaner Production, 2021, 281: 125224.
[35] 蘭玉彬,王天偉,陳盛德,等. 農(nóng)業(yè)人工智能技術(shù):現(xiàn)代農(nóng)業(yè)科技的翅膀[J]. 華南農(nóng)業(yè)大學(xué)學(xué)報(bào),2020,41(6):1-13.
Lan Yubin, Wang Tianwei, Chen Shengde, et al. Agricultural artificial intelligence technology: Wings of modern agricultural science and technology[J]. Journal of South China Agricultural University, 2020, 41(6): 1-13. (in Chinese with English abstract)
[36] Thiebes S, Lins S, Sunyaev A. Trustworthy artificial intelligence[J]. Electronic Markets, 2021, 31(2): 447-464.
[37] 蔡靜穎. 計(jì)算機(jī)人工智能技術(shù)的應(yīng)用與發(fā)展[J]. 產(chǎn)業(yè)創(chuàng)新研究,2022(14):78-80.
[38] 學(xué)術(shù)點(diǎn)滴. CiteSpace關(guān)鍵詞聚類圖譜含義詳細(xì)解析[EB/OL].(2020-02-21)[2022-04-27].https://blog.csdn.net/qq_39974284/article/details/104425964?ops_request_misc=%257B%2522request%255Fid%2522%253A%2522167145079716800186559380%2522%252C%2522scm%2522%253A%252220140713.130102334.pc%255Fblog.%2522%257D&request_id=167145079716800186559380&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2~blog~first_rank_ecpm_v1~rank_v31_ecpm-1-104425964-null-null.nonecase&utm_term=CiteSpace%E5%85%B3%E9%94%AE%E8%AF%8D%E8%81%9A%E7%B1%BB%E5%9B%BE%E8%B0%B1&spm=1018.2226.3001.4450.
[39] Lan Y, Huang Z, Deng X, et al. Comparison of machine learning methods for citrus greening detection on UAV multispectral images[J]. Computers and Electronics in Agriculture, 2020, 171: 105234.
[40] 蘭玉彬,朱梓豪,鄧小玲,等. 基于無人機(jī)高光譜遙感的柑橘黃龍病植株的監(jiān)測(cè)與分類[J]. 農(nóng)業(yè)工程學(xué)報(bào),2019,35(3):92-100.
Lan Yubin, Zhu Zihao, Deng Xiaoling, et al. Monitoring and classification of citrus Huanglongbing based on UAV hyperspectral remote sensing[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2019, 35(3): 92-100. (in Chinese with English abstract)
[41] Yang D, Wang F, Hu Y, et al. Citrus huanglongbing detection based on multi-modal feature fusion learning[J]. Frontiers in Plant Science, 2021, 12: 809506.
[42] Dai F, Wang F, Yang D, et al. Detection method of citrus psyllids with field high-definition camera based on improved cascade region-based convolution neural networks[J]. Frontiers in Plant Science, 2021, 12: 816272.
[43] Wheatley M S, Duan Y P, Yang Y. Highly sensitive and rapid detection of citrus Huanglongbing pathogen (‘Candidatus Liberibacter asiaticus’) using Cas12a-based methods[J]. Phytopathology, 2021, 111(12): 2375-2382.
[44] Bassanezi R B, Primiano I V. Huanglongbing and citrus variegated chlorosis integrated management based on favorable periods for vector population increase and symptom expression[J]. Plant Disease, 2021, 105(10): 3037-3047.
[45] Gao F, Fang W, Sun X, et al. A novel apple fruit detection and counting methodology based on deep learning and trunk tracking in modern orchard[J]. Computers and Electronics in Agriculture, 2022, 197: 107000.
[46] Wang L, Zhao Y, Xiong Z, et al. Fast and precise detection of litchi fruits for yield estimation based on the improved YOLOv5 model[J]. Frontiers in Plant Science, 2022, 13:1-16.
[47] Zhuang J, Hou C, Tang Y, et al. Computer vision-based localisation of picking points for automatic litchi harvesting applications towards natural scenarios[J]. Biosystems Engineering, 2019, 187: 1-20.
[48] Wang H, Lin Y, Xu X, et al. A study on long-Close distance coordination control strategy for litchi picking[J]. Agronomy, 2022, 12(7): 1520.
[49] Wang X, Kang H, Zhou H, et al. Geometry-aware fruit grasping estimation for robotic harvesting in apple orchards[J]. Computers and Electronics in Agriculture, 2022, 193: 106716.
[50] Park S, Ryu D, Fuentes S, et al. Adaptive estimation of crop water stress in nectarine and peach orchards using high-resolution imagery from an unmanned aerial vehicle (UAV)[J]. Remote Sensing, 2017, 9(8): 828.
[51] Egea G, Padilla-Díaz C M, Martinez-Guanter J, et al. Assessing a crop water stress index derived from aerial thermal imaging and infrared thermometry in super-high density olive orchards[J]. Agricultural Water Management, 2017, 187: 210-221.
[52] Provenzano G, Segovia-Cardozo D A. Detecting crop water requirements indicators in irrigated agro-ecosystems from soil water content profiles: An application for a citrus orchard[C]//EGU General Assembly Conference Abstracts, Gottingen, 2021: EGU21-9172.
[53] Hahn F, García J A. Mango stem response under different irrigation regimes[J]. International Journal of Fruit Science, 2022, 22(1): 35-56.
[54] Meng Y, Zhong W, Liu C, et al. UAV spraying on citrus crop: Impact of tank-mix adjuvant on the contact angle and droplet distribution[J]. PeerJ, 2022, 10: e13064.
[55] Guo S, Yao W, Xu T, et al. Assessing the application of spot spray in Nanguo pear orchards: Effect of nozzle type, spray volume rate and adjuvant[J]. Pest Management Science, 2022, 78(8): 3564-3575.
[56] Wang C, Liu Y, Zhang Z, et al. Spray performance evaluation of a six-rotor unmanned aerial vehicle sprayer for pesticide application using an orchard operation mode in apple orchards[J]. Pest Management Science, 2022, 78(6): 2449-2466.
[57] Sahni R K, Ranjan R, Hoheisel G A, et al. Pneumatic spray delivery-based solid set canopy delivery system for oblique banded leaf roller and codling moth control in a high-density modern apple orchard[J]. Pest Management Science, 2022, 78(11): 4793-4801.
[58] Godoy-Nieto A, Miranda-Fuentes A, Grella M, et al. Assessment of spray deposit and loss in traditional and intensive olive orchards with conventional and crop-adapted sprayers[J]. Agronomy, 2022, 12(8): 1764.
[59] Gu C, Zhao C, Zou W, et al. Innovative leaf area detection models for orchard tree thick canopy based on lidar point cloud data[J]. Agriculture, 2022, 12(8): 1241.
[60] Raman M G, Carlos E F, Sankaran S. Optimization and evaluation of sensor angles for precise assessment of architectural traits in peach trees[J]. Sensors, 2022, 22(12): 4619.
[61] Ljubobratovi? D, Vukovi? M, Brki? Bakari? M, et al. Assessment of various machine learning models for peach maturity prediction using non-destructive sensor data[J]. Sensors, 2022, 22(15): 5791.
[62] Minamikawa M F, Nonaka K, Hamada H, et al. Dissecting breeders’ sense via explainable machine learning approach: Application to fruit peelability and hardness in citrus[J]. Frontiers in Plant Science, 2022, 13: 832749.
[63] Mo J, Lan Y, Yang D, et al. Deep learning-based instance segmentation method of litchi canopy from UAV-Acquired images[J]. Remote Sensing, 2021, 13(19): 3919.
[64] Dhiman P, Kukreja V, Manoharan P, et al. A novel deep learning model for detection of severity level of the disease in citrus fruits[J]. Electronics, 2022, 11(3): 495.
[65] Li D, Sun X, Elkhouchlaa H, et al. Fast detection and location of longan fruits using UAV images[J]. Computers and Electronics in Agriculture, 2021, 190: 106465.
[66] Dogan H. A new empirical propagation model depending on volumetric density in citrus orchards for wireless sensor network applications at sub-6 GHz frequency region[J]. International Journal of RF and Microwave Computer-Aided Engineering, 2021, 31(9): e22778.
[67] Zhang Y, Ta N, Guo S, et al. Combining spectral and textural information from UAV RGB images for leaf area index monitoring in kiwifruit orchard[J]. Remote Sensing, 2022, 14(5): 1063.
[68] Caruso G, Palai G, Gucci R, et al. Remote and proximal sensing techniques for site-specific irrigation management in the olive orchard[J]. Applied Sciences, 2022, 12(3): 1309.
[69] Deng X, Zhu Z, Yang J, et al. Detection of citrus huanglongbing based on multi-input neural network model of UAV hyperspectral remote sensing[J]. Remote Sensing, 2020, 12(17): 2678.
[70] Chang A, Yeom J, Jung J, et al. Comparison of canopy shape and vegetation indices of citrus trees derived from UAV multispectral images for characterization of citrus greening disease[J]. Remote Sensing, 2020, 12(24): 4122.
[71] Zhang Y, Ta N, Guo S, et al. Combining spectral and textural information from UAV RGB images for leaf area index monitoring in kiwifruit orchard[J]. Remote Sensing, 2022, 14(5): 1063.
[72] Liu Z, Guo P, Liu H, et al. Gradient boosting estimation of the leaf area index of apple orchards in UAV remote sensing[J]. Remote Sensing, 2021, 13(16): 3263.
[73] Qin J, Burks T F, Ritenour M A, et al. Detection of citrus canker using hyperspectral reflectance imaging with spectral information divergence[J]. Journal of Food Engineering, 2009, 93(2): 183-191.
[74] Gottwald T R. Current epidemiological understanding of citrus huanglongbing[J]. Phytopathol, 2010, 48: 119-139.
[75] Sa I, Ge Z, Dayoub F, et al. Deepfruits: A fruit detection system using deep neural networks[J]. Sensors, 2016, 16(8): 1222: 1-23.
[76] Gongal A, Amatya S, Karkee M, et al. Sensors and systems for fruit detection and localization: A review[J]. Computers and Electronics in Agriculture, 2015, 116: 8-19.
[77] Stein M, Bargoti S, Underwood J. Image based mango fruit detection, localisation and yield estimation using multiple view geometry[J]. Sensors, 2016, 16(11): 1915.
[78] Nguyen T T, Vandevoorde K, Wouters N, et al. Detection of red and bicoloured apples on tree with an RGB-D camera[J]. Biosystems Engineering, 2016, 146: 33-44.
[79] Csillik O, Cherbini J, Johnson R, et al. Identification of citrus trees from unmanned aerial vehicle imagery using convolutional neural networks[J]. Drones, 2018, 2(4): 39.
[80] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[EB/OL]. (2014-09-04) [2022-04-27]. https://arxiv.org/abs/1409.1556.
[81] Ampatzidis Y, Partel V. UAV-based high throughput phenotyping in citrus utilizing multispectral imaging and artificial intelligence[J]. Remote Sensing, 2019, 11(4): 410.
[82] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE conference on Computer Vision and Pattern Recognition. Las Vegas, 2016: 770-778.
[83] Tian Y, Yang G, Wang Z, et al. Apple detection during different growth stages in orchards using the improved YOLO-V3 model[J]. Computers and Electronics in Agriculture, 2019, 157: 417-426.
Research progress and hotspots of smart orchard based on bibliometrics
Lan Yubin1,2,3,4, Lin Zeshan1,2,Wang linlin1,2, Deng Xiaoling1,2,3,4※
(1.,510642,;2.510642,; 3.(),510642,; 4.510642,)
In order to better promote the development of smart agriculture, this review aims to analyze the research trends, frontiers, and hotspots of the smart orchard at home and abroad using bibliometric analysis. The commonly-used tool (Citespace) of quantitative literature analysis was adopted as the bibliometric analysis in the fields of science and technology. Web of science was selected as the retrieval platform to analyze the temporal and spatial distribution of research publications, main research contents, and frontier hotspots of smart orchards published from January 2002 to August 2022. Keywords of crop mainly included the longan, citrus, lychee, and peach. In addition, the keyword "orchard" was added for spare. 579 documents were finally obtained after screening and preprocessing using the following retrieval items: " TI=(longan or citrus or litchi or peach or orchard) And AB=(growth or disease or growing or pest or insect or tree or fruit) And AK=(drone or UAV or AI or intelligent or detection or segmentation or precision or spray or unmanned or robot or sensor or "deep learning" or "machine learning" or "agricultural machinery") . The retrieved data was used to conduct the following steps: The data processing software (Excel) and the bibliometric analysis tool (CiteSpace) were selected to conduct the quantitative analysis. The annual publication from 2002 to 2022 were counted using Excel and the built-in Web of science database. The collinear knowledge map of core authors, institutions, regions, and keywords was then obtained using Citespace and Excel statistical analysis tools. The analysis was also performed on the institutional cooperation network, literature co-citation, high-frequency word clustering, keyword co-occurrence, and keyword emergence. The development history, research regions, institutions, and spatial information, research technologies, and application hotspots of smart orchards were sorted out and summarized over the past 20 years using knowledge graphs. The main conclusions were as follows: The research on smart orchards was on the right track since 2014. There was the rapid development under the promotion of artificial intelligence technology since 2018. Reports published from 2018 to 2021 accounted for 37.5% of the total. In general, there were a relatively close exchange and cooperation between the authors (Lan YB, Chen C, and Tang Y), institutions (South China Agricultural University, China Agricultural University, and Univ of Florida), and regions (China, the United States, and Spain). China and the United States were the major countries in the smart orchard research, accounting for 58.2% of the total. The current research topics were focused mainly on fruit tree growth monitoring, pest identification, and early warning, unmanned or intelligent agricultural machinery operation. The technologies were adopted, including artificial intelligence models/algorithms, sensing, Internet of Things, and Precision control, according to the subdivision of research purposes. The research of deep learning, UAV, and artificial intelligence was the frontier of smart orchard development. The development of smart orchards was deeply promoted by advanced technology, especially artificial intelligence. However, the current limiting steps were determined by the high complexity of the environment and the lack of standard planting in further development. The research directions of smart orchards can be expected as the star-sky-ground three-dimensional orchard perception, air-ground collaborative unmanned precision operation, fruit picking, and visual traceability of fruit products in the future.
intelligent; automatation; smart orchard; quantitative analysis; web of science; citespace; bibliometric analysis
10.11975/j.issn.1002-6819.2022.21.016
S126
A
1002-6819(2022)-21-0127-10
蘭玉彬,林澤山,王林琳,等. 基于文獻(xiàn)計(jì)量學(xué)的智慧果園研究進(jìn)展與熱點(diǎn)分析[J]. 農(nóng)業(yè)工程學(xué)報(bào),2022,38(21):127-136.doi:10.11975/j.issn.1002-6819.2022.21.016 http://www.tcsae.org
Lan Yubin, Lin Zeshan, Wanglinlin, et al. Research progress and hotspots of smart orchard based on bibliometrics[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2022, 38(21): 127-136. (in Chinese with English abstract) doi:10.11975/j.issn.1002-6819.2022.21.016 http://www.tcsae.org
2022-04-27
2022-10-03
廣東省重點(diǎn)研發(fā)計(jì)劃項(xiàng)目(2019B020214003);嶺南現(xiàn)代農(nóng)業(yè)實(shí)驗(yàn)室科研項(xiàng)目(NT2021009);廣州市重點(diǎn)研發(fā)計(jì)劃項(xiàng)目(202103000090);廣東高校重點(diǎn)領(lǐng)域人工智能專項(xiàng)項(xiàng)目(2019KZDZX1012);高等學(xué)校學(xué)科創(chuàng)新引智計(jì)劃資助(D18019);國家自然科學(xué)基金面上項(xiàng)目(61675003)
蘭玉彬,博士,教授,研究方向?yàn)榫珳?zhǔn)農(nóng)業(yè)航空及航空應(yīng)用與遙感技術(shù)。Email:ylan@scau.edu.cn
鄧小玲,博士,副教授,研究方向?yàn)檗r(nóng)業(yè)航空遙感應(yīng)用。Email:dengxl@scau.edu.cn