各位同行與學界朋友：
感謝大家對國際精準農業航空學會(International Society of Precision AgriculturalAviation,ISPAA)和《國際精準農業航空學報》（International Journal of Precision AgriculturalAviation, IJPAA）的支持,我們於7月在線發布了2020年第2期（Vol 3,No2 (2020): IJPAA），歡迎大家閱讀和引用！在此，感謝各位作者通過我們的期刊發表分享自己的研究成果。為了方便大家更好的了解，我們為每篇文章的題目和摘要等進行了中文翻譯，內容按第2期目錄的順序，依次展示：Overview of spray nozzles for plant protection from manned aircrafts: Present research and prospectiveHaibo Chen, Bradley K Fritz*, Yubin Lan*, ZhiyanZhou,Jingfu Zheng
Abstract: Aerial application is a critical component of modern agriculture, and it is crucial for aerial application ofpesticides to be environmentally protective and efficacious. The spray nozzles involved in the application process are a vital component in the precise and safe delivery of applied products. This paper reviews and summarizes the state-of-the-art in aviation nozzle technology and the physical processes of nozzle atomization on manned platforms. Highlights are two main aerial nozzle types along with their working principle, the factors that influence atomization performance and new technologies for reducing drift and enhancing application efficiency. Moving forward, the research mainly focused on the development and evaluation of drift-reducing and variable-rate technologies, enhanced atomization models, the impacts of aerial tank mix adjuvants, and non-conventional application technologies (such as electrostaticor pulse-width modulation systems) are likely to have the most significant impact on the aerial application industry. This review provides a summary of the history and advancements in nozzle technologies and encourages further development.Keywords: aerial application, nozzle, dropletsize, atomization performance, drift-reducingDOI: 10.33440/j.ijpaa.20200302.76作者：陳海波，Bradley K Fritz*，蘭玉彬*，周志豔，鄭景富
摘要：航空施藥作業是現代農業的重要組成部分，實現環保及高效的航空施藥至關重要。作業過程中涉及的噴嘴是精確和安全噴施農藥的一個重要的組成部分。本文綜述了航空噴嘴技術的最新進展和有人駕駛飛機平臺上噴嘴霧化的物理過程。重點介紹了兩種主要的空中噴嘴類型及其工作原理、影響霧化性能的因素以及減少漂移和提高應用效率的新技術。展望未來，研究主要集中在減漂技術、可變速率技術、優化增強的霧化模型、航空噴霧助劑的影響以及非常規應用技術（如靜電或脈衝寬度調製系統）的開發和評估上，這些技術可能對航空施藥行業產生最重大的影響。這篇綜述總結了噴嘴技術的研究歷史和最新進展，並鼓勵進一步的研究和探索。關鍵詞: 航空施藥；噴嘴；霧滴尺寸；霧化性能；減漂DOI: 10.33440/j.ijpaa.20200302.76Key technologies for testing and analyzing aerial spray deposition and drift: A comprehensive reviewRuirui Zhang, Liping Chen*, Yao Wen, Qing Tang, Longlong LiAbstract: The technologies for testing and analyzing aerial spray deposition and drift serve as the tools and foundational technologies for spray deposition and drift modeling, deposition and drift control, and the development of aerial spray equipment. These technologies can be categorized into four types by comprehensively considering their testing analysis methods,analysis objects, and application technologies: sampling, laboratory simulation, computer simulation modeling, and new analysis technologies. With regard to sampling analysis technologies, this study mainly analyzed the water-sensitive paper (WSP) sampling testing method, tracer testing method,combined WSP and tracer testing method, as well as the electronic information technologies that have been widely used and rapidly developed in recent years.With respect to the laboratory simulation analysis technologies, this paper elaborates on the applications of laser particle size measurement technology and instrument based on the laser diffraction principle,particle image velocimetry technology and instrument, phase doppler interferometer based on laser scattering principle, and other spray measurement technologies. In case of computational modeling simulation analysis technologies, this paper mainly expounds the spray deposition model analysis and research methods based on the Gaussian plume, Lagrange, statistical, and computational fluid dynamics (CFD) models. Additionally, the paper describes the applications of LIDAR, thermal infrared imaging, and other technologies to the analysis of spray deposition. Electronic technology, computer technology, and other information technologies are being used more widely for analyzing aerial spray deposition, and have become a development trend in recent years. The instruments rapid measurement of spray deposition in the field and the real-time accurate prediction models for spray drift are in high demand. The instrument for rapid in-field measurement should be compact, exhibit good portability and convenience of usein the field, and guarantee high measurement accuracy. The spray deposition and drift mechanisms are relatively well clarified, and the use of advanced technologies to develop practical instrument is the main work of future research in this area.Keywords:aerial application, deposit measurement, drift monitor,plant protectionDOI: 10.33440/j.ijpaa.20200302.80作者：張瑞瑞, 陳立平*, 文瑤, 唐青, 李龍龍摘要：航空施藥噴霧沉積漂移檢測分析技術是施藥沉積漂移建模、沉積漂移控制和施藥器械研發的工具性和基礎性技術。綜合檢測分析方式、分析對象和應用技術，分為取樣式分析技術、實驗室模擬分析技術、計算建模仿真分析技術和新型分析技術四種檢測分析技術類型。取樣式分析技術主要分析了應用較廣而近年發展較快的水敏紙取樣檢測方式、示蹤劑檢測方式、水敏紙與示蹤劑組合檢測方式和電子信息技術在取樣式分析技術中的應用。實驗室模擬分析技術闡述了雷射衍射原理的雷射粒度測量技術和儀器、粒子圖像測速技術和儀器、雷射散射原理的相位都卜勒幹涉儀等噴霧測量技術的應用。計算建模仿真分析技術主要闡述了基於高斯煙羽模型、拉格朗日模型、統計模型和CFD模型的噴霧沉積模型分析研究方法。本文還闡述了雷射雷達、熱紅外成像等技術在噴霧沉積分析中應用。電子技術、計算機技術等信息技術在航空施藥噴霧沉積分析中的應用越來越廣泛，成為近年的發展趨勢。噴霧沉積的田間快速測量儀器和噴霧漂移的實時精準預測模型需求最為迫切。田間快速測量儀器既需要儀器較小、便攜方便田間使用，又需要有較高測量精度。噴霧漂移受環境、施藥機型、藥劑等因素交互影響，計算建模數據計算量巨大、精度難以保證。噴霧沉積漂移機理已經比較明確，利用先進技術研發實用化的設備是該領域未來的主要發展方向。DOI: 10.33440/j.ijpaa.20200302.80Comparison of water stresscoefficient using three alternative canopy temperature-based indicesHuihui Zhang*, Liyuan Zhang, Yaxiao Niu, Ming Han, Kevin YemotoAbstract: In this study three crop canopy temperature-based water stress indices, standard deviation of the distribution of canopy temperature(CTSD), the ratio of canopy temperature of non-stressed to stressed canopy (Tc–ratio) and Degrees Above Non-Stressed (DANS), were tested as the substitute of water stress coefficient (Ks) for maize crop water use estimation. Thermal imagery was taken from maize under various levels of deficit irrigation at different crop growth stages in 2015 and 2016 growing seasons. The Expectation-Maximization algorithm was used to estimate the canopy temperature distribution from thermal imagery under a range of crop coverage and water stress conditions. CTSD, Tc–ratioand DANS were calculated from the extract canopy temperature and converted towater stress coefficient denoted as Ks–CTSD, Tc–ratio,and Ks–DANS. Crop transpiration estimated using three water stress coefficients were compared with sap flow measurements in 2015. The results further confirmed that CTSD responded well to irrigation events (timing and depth) on crops with water stress and was significantly correlated to leaf water potential and soil water deficit, especially when stress level was above moderate. Ks–CTSD was more sensitive to soil water deficit than Tc–ratio and Ks–DANS. Crop transpiration estimated using Ks–CTSD preformed the best among all methods when compared with sap flow measurements (R2_adj=0.58, relative absolute error =0.63, and root mean square error =0.87 mm day-1). Nash-Sutcliffe coefficient of 0.61 indicates the performance of the prediction model is sufficient and satisfactory. The canopy temperature-based index, CTSD, is easy to acquire from high resolution thermal imagery from remote sensing platforms, such as ground and unmanned aerial vehicles. It has a strong application potential to improve crop water stress detection and crop water use estimation for irrigation scheduling.Keywords:canopy temperature,CTSD, maize, water stress, thermal, soil water deficitDOI: 10.33440/j.ijpaa.20200302.78作者：張匯慧*，張立元, 牛亞曉, 韓明, KevinYemoto摘要：在這項研究中，基於三種作物冠層溫度的水分脅迫指數，冠層溫度分布的標準偏差（CTSD），無水分脅迫冠層溫度與脅迫冠層溫度的比率（Tc–ratio）以及無水分脅迫冠層溫度與脅迫冠層溫度差（DANS），被用作替代水分脅迫係數（Ks）來估算玉米作物用水。本研究在2015和2016兩期生長季節，在不同作物生長階段，不同水平的虧水灌溉條件下獲取玉米冠層的熱紅外圖像。期望最大化算法用於提取冠層溫度分布。根據提取的冠層溫度計算出CTSD，Tc–ratio和DANS，並將其轉換為水分脅迫係數（Ks–CTSD、Tc–ratio,和Ks–DANS）。2015年，本研究對使用三個水分脅迫係數估算的作物蒸騰量與液流的測量結果進行了比較。這項研究結果進一步證實，CTSD對作物的灌溉事件（時間和深度）反應良好，並且與葉片水勢和土壤水分虧缺顯著相關，尤其是在水分脅迫水平高於中等水平時。Ks_CTSD比Tc–ratio和Ks–DANS對土壤水分虧缺更為敏感。與莖液流量相比，使用Ks–CTSD估算的作物蒸騰性能是所有方法中最好的（R2_adj = 0.58，相對絕對誤差= 0.63，均方根誤差=0.87 mm 每天）。Nash-Sutcliffe係數為0.61表示預測模型的性能的滿意度。冠層溫度指數CTSD易於從遙感平臺（如地面和無人機等）的高解析度熱圖像中獲取，所以具有很大的應用潛力，可以提高灌溉計劃中作物水分脅的監測和作物用水量的估算。關鍵詞：冠層溫度；冠層溫度分布的標準偏差CTSD；玉米；水分脅迫；熱紅外；土壤水分虧缺DOI: 10.33440/j.ijpaa.20200302.78Control efficacy of different pesticide formulations and fan-nozzle model on wheat aphids by UAVsXiaojing Yan*, Huizhu Yuan*,XiaoxinZhou, Ming Wang, Xin Shi, YixuanChen, Daibin YangAbstract: The use of unmanned aerial vehicles (UAVs) has been significantly increased in wheat pest control in China. The formulation of pesticide and the type of nozzle are the important parameters in UAVs spraying application. The droplet density, deposition and control efficacy on wheat aphids of four formulation of imidacloprid (20% imidacloprid SL, 25% imidacloprid WP, 5% imidacloprid EC, 70% imidacloprid WDG) and three models of fan-nozzle (teejet11001vs, teejet110015vs ,teejet11002vs) applied by UAVs were determined in this paper. The droplet density and deposition of four imidacloprid formulation was notsignificantly different. However, the control efficacy of different formulation was 5% imidacloprid EC (7DAT=97.7%) > 20% imidacloprid SL (7DAT=95.7%) > 70% imidacloprid WDG (7DAT=93.2%) > 25% imidacloprid WP(7DAT=85.1%). With regard to the factor of fan-nozzle, teejet11001vs treatment performed better in droplet density and deposition than the other two nozzles,and the control efficacy of teejet11001vs was higher than teejet110015vs, teejet11002vs with the margin of 10.7%, 9.9% respectively. The deposition uniformity on wheat plant fromtop to bottom of three nozzles was teejet11001vs (CV=22.8%)>teejet11002vs (CV=27.1%)>teejet110015vs (CV=57.4%).Keywords: unmanned aerial vehicles, pesticide formulation, fan nozzlemodel, wheat aphids, droplet density, deposition, control efficacyDOI: 10.33440/j.ijpaa.20200302.74農藥劑型及扇形噴頭型號對植保無人飛機噴施防治小麥蚜蟲防治效果的影響作者：閆曉靜*，袁會珠*，周曉欣，王明，石鑫，陳奕璇，楊代斌摘要：植保無人飛機在我國小麥病蟲害防治中的應用日趨廣泛。農藥劑型和噴嘴類型是無人機田間噴施的重要參數。本文測定了四種吡蟲啉製劑(20%吡蟲啉SL、25%吡蟲啉WP、5%吡蟲啉EC、70%吡蟲啉WDG)及三種噴頭(teejet11001vs、teejet110015vs、teejet11002vs)對植保無人飛機田間噴施防治小麥蚜蟲的霧滴密度、沉積量及防治效果的影響。結果表明，四種吡蟲啉製劑的霧滴密度及沉降量無顯著差異。不同劑型的防治效果為5%吡蟲啉EC (7DAT=97.7%)>20%吡蟲啉SL (7DAT=95.7%)>70%吡蟲啉WDG (7DAT=93.2%)>25%吡蟲啉WP (7DAT=85.1%)。對於噴頭類型這一參數，teejet11001vs處理的霧滴密度及沉積效果優於teejet110015vs和teejet11002vs，其防治效果分別比其他兩種噴頭高10.7%和9.9%。3個噴頭在小麥植株自上而下的沉積量均勻性分別為teejet11001vs (CV=22.8%)>teejet11002vs (CV=27.1%)>teejet110015vs (CV=57.4%)。關鍵詞:無人機；農藥製劑；扇形噴頭類型；小麥蚜蟲；霧滴密度；沉積量；防治效果DOI: 10.33440/j.ijpaa.20200302.74Field evaluation of a Tri-Set spray nozzle for aerial application and discussion on release of biological control agentsYanbo Huang*, Steven J. ThomsonAbstract: Efficiency and efficacy of aerial application of liquid formations has been studied extensively in terms of optimal nozzle/atomizer setups, adjuvant mixes, weather conditions, and aerial spray release heights. In past studies, we have evaluated CP-11 flat-fan nozzles to assess impact factors of in-swath and downwind deposition from aerial spray application, and a solid stream radial Accu-Flo nozzle to determine penetration of the spray within crop canopy. When applying liquid tank mixes from aerial platforms, there are numerous nozzle types available with differing spray characteristics. More information is needed, however, on the ability of aerial delivery systems to effectively apply biological agents. The release ofnon-toxigenic A.flavus into corn fields has shown promise as a biological control agent for aflatoxin producing strains of the fungus. However, the application of a coarse granuleto mature, two-meter-tall corn is a challenge. Thus, there would be substantial advantages to a liquid formulation with necessary identification of appropriate adjuvants to disperse the highly hydrophobic spores of A. flavus. This paper presents the experiment and preliminary data analysis of testing and evaluating Davidon tri-set nozzles under various nozzle configurations, and discusses what we need to know for effective use of different nozzles for potential application of biological control agents, especially Afla-Guard®, a commercially available product containing non-toxigenic A.flavus as a biological control agent, and related products into corn fields.Keywords: aerial application, crop protection, nozzle evaluation,biological control agentsDOI: 10.33440/j.ijpaa.20200302.75摘要：在最佳噴嘴/霧化器設置、助劑混合、天氣條件和空中噴霧釋放高度等方面，人們已經廣泛研究了空中施用液體配方的效率和功效。在過去的研究中，我們評估了CP-11平扇形噴嘴在航空噴霧施加的吹入和順風沉積的影響因素，並評估了固態流徑Accu-Flo噴嘴，以確定噴霧在農作物冠層中的滲透。當從高空作業平臺上施加液體農藥混合物時，可使用多種具有不同噴霧特性的噴嘴。但是，把粗顆粒的農藥釋放到兩米高的成熟玉米上仍然是個挑戰。由此，對於液體製劑而言，配有適當的助劑來擴散黃麴黴的高疏水性孢子具有特別的優勢。本文介紹了在各種噴嘴配置下測試和評估Davidon三聯噴嘴的實驗和初步數據分析結果，並討論了如何可能有效地使用不同噴嘴使用各種生物防治治劑（尤其是Afla-Guard®，是一種已經商業化的非毒素黃麴黴作為生物防治治劑）以及其它有關產品進行玉米田間的的生物防治。關鍵字：航空施藥；作物保護；噴頭評估；生物防治治劑DOI: 10.33440/j.ijpaa.20200302.75Domain centralization and cross-modal reinforcement learning for vision-based robotic manipulationKai Yang, Zhanpeng Zhang, Hui Cheng*,Huadong Wu, Ziying Guo Abstract: Vision-based robotic manipulation with deep learning method has achieved substantial advances in the field of automatic agriculture, which can be deployed and applied in the picking,sorting and transporting of agricultural products and so on. Deep reinforcement learning (DRL) is one of the learning-methods that help the robot learn the policy itself by exploration and exploitation. Training real robots with DRL would take a great price that limits its application scope. Some approaches train the DRL policy in simulation and deploy the model to real robot by transferring the images in simulator to that of the real world. However, this method requires pre-collected images as training data for each real scene. In this paper, a domain centralized approach is proposed as the sim-to-real perception module tocapture the task-specific characteristics of the vision regardless of the reality gap between simulation and the real environments. Another challenge for vision-based manipulation is the learning difficulty caused by the high-dimensional vision input. Here we propose a cross-modal reinforcement learning scheme by leveraging the full system states to provide additional guidance. The experimental results show that the proposed method can perform a real robot grasping task without real-world data and outperforms current methods with the same experimental settings.Keywords: sim-to-real, robotic manipulation, agriculturalapplication, domain centralization, cross-modal learningDOI: 10.33440/j.ijpaa.20200302.77基於領域集中化和跨模態強化學習的視覺輸入機器人操作方法摘要：基於深度學習的視覺輸入機器人操作方法在自動農業領域取得了堅實的進步，在農產品揀選、分揀和運輸等領域實現了廣泛的推廣和應用。深度強化學習（DRL）是一種使機器人自主探索和利用環境以學習行為策略的深度學習方法。但是，使用深度強化學習算法讓真實世界裡的機器人直接進行訓練會付出很大的成本，這樣昂貴的訓練成本限制了強化學習在機器人上的應用範圍。目前一些方法通過轉化仿真圖像，使其看起來更加接近於真實世界圖像來實現在仿真器中訓練智能體的深度強化學習策略，並將模型部署到現實的機器人中的目的。但是這樣的方法通常需要預先收集一批真實環境的圖像作為每個真實場景的訓練數據，同樣增加了機器人的訓練成本。本文提出了一種領域集中化方法實現了從仿真圖像到現實圖像的感知策略，用來捕獲與任務相關的特定特徵，從而減輕模擬環境和真實環境之間的差異對後續深度強化學習訓練策略造成的影響。強化學習訓練視覺輸入機器人的另一個挑戰是由於高維視覺輸入引起的訓練困難問題。這裡我們提出了一種跨模態的深度強化學習方法，利用整個仿真系統中可知的狀態來提供額外的訓練梯度指導以解決上述問題。實驗結果表明，本文所提出的方法可以在沒有額外現實數據的前提下使強化學習方法完成對現實機器人抓取任務的訓練，並且在相同的實驗條件設置下優於目前的方法。關鍵字：仿真到真實；機器人操作方法；農業應用；領域集中化；跨模態強化學習DOI: 10.33440/j.ijpaa.20200302.77Method for establishing the UAV-rice vortex 3D model and extracting spatial parametersJiyu Li, Han Wu, Xiaodan Hu, Gangao Fan,Yifan Li, Bo Long, Xu Wei, Yubin Lan* Abstract: With the deepening research on the rotor wind field of UAV operation, it has become a mainstream to quantify the UAV operation effect and study the distribution law of rotor wind field via the spatial parameters ofthe UAV-rice interaction wind field vortex. At present, the point cloud segmentation algorithms involved in most wind field vortex spatial parameter extraction methods cannot adapt to the instantaneous changes and indistinct boundary of the vortex. As a result, there are problems such as inaccurate three-dimensional (3D) shape and boundary contour of the wind field vortex as well as large errors in the vortex’s spatial parameters. To this end, this paper proposes an accurate method for establishing the UAV-rice interaction vortex 3D model and extracting vortex spatial parameters. Firstly, the original point cloud data of the wind filed vortex were collected in the image acquisition area. Secondly, DDC-UL processed the original point cloud data to develop the 3D point cloud image of the wind field vortex. Thirdly, the 3D curved surface was reconstructed and spatial parameters were then extracted. Finally, the volume parameters and top surface area parameters of the UAV-rice interaction vortex were calculated and analyzed. The results show that the error rate of the 3D model of the UAV-rice interaction wind field vortex developed by the proposed method is kept within 2%, which is at least 13 percentage points lower than that of algorithms like PointNet. The average error rates of the volume parameters and the top surface area parameters extracted by the proposed method are 1.4% and 4.12%, respectively. This method provides 3D data for studying the mechanism of rotor wind field in the crop canopy through the 3D vortex model and its spatial parameters.Keywords: UAV, rice canopy, wind field, binocular vision, point cloudsegmentation, spatial parametersDOI: 10.33440/j.ijpaa.20200302.84作者：李繼宇, 巫瀚, 胡瀟丹, 樊港奧, 李一凡, 龍波, 魏旭, 蘭玉彬*摘要：隨著無人機作業旋翼風場研究的深入，利用無人機-水稻互作風場渦旋空間參數量化無人機作業效果及旋翼風場分布規律已成為主流；目前，大部分風場渦旋空間參數提取方法所涉及到的點雲分割算法都無法適應渦旋瞬時變化、邊界不明顯等特點，均存在採集風場渦旋三維形態及邊界輪廓不準確、風場渦旋空間參數提取誤差大等問題；為解決上述問題，本文提出了一種無人機-水稻互作風場渦旋三維模型建立與空間參數提取方法，該方法在實驗採集渦旋所在圖像採集區原始點雲數據的基礎上，採用本文開發的DDC-UL點雲分割算法處理圖像採集區原始點雲，獲取風場渦旋三維點雲圖並進行三維重建及空間參數提取，解算出風場渦旋體積參數和頂面面積參數；通過實驗建立無人機-水稻互作風場渦旋三維模型並提取空間參數，結果表明：該方法建立風場渦旋三維模型誤差率維持2%以內，比PointNet等點雲分割算法提高了至少13個百分點；該方法提取的體積參數和頂表面積參數的平均誤差率分別為1.4％和4.12％。本文所述無人機-水稻互作風場渦旋三維模型建立與空間參數提取方法通過建立的渦旋三維模型和提取的渦旋空間參數為研究無人機旋翼風場在作物冠層的作用機理提供三維信息數據。關鍵詞：無人機；大米冠層；風場；雙目視覺；點雲分割；空間參數DOI:10.33440/j.ijpaa.20200302.84Deposition and biological efficacy of UAV-based low-volume application in rice fieldsGuobin Wang, Xuan Li, John Andaloro,Pengchao Chen, Cancan Song,Changfeng Shan, Shengde Chen, Yubin Lan*Abstract:Efficient and accurate application of pesticides can improve biological efficacy, reduce insecticide resistance,and protect the environment. The rapid development of unmanned aerial vehicles(UAV) technology as a new methodof pesticide application using low spray volumes demands scientific evaluation compared to conventional practices. The objective of this research was to analyze the effects of spray volume and tank-mix adjuvants on droplet deposition, canopy penetration, and control ofrice blast disease and rice leaf roller when applied by UAV technology onrice. An electric backpack sprayer wasused as the standard method of application for comparison. Increasing the spray volume and adding anadjuvant significantly increased droplet density, percentage of spray coverage, and control of rice blast and rice leaf roller for the UAV application. The control efficacy of the UAV sprayer was basically equal to or slightly worse than the backpack sprayer. These data indicate that a UAV application made at a spray volume of 18 L/hm2with the addition of a methylated crop oil adjuvant at panicle initiation provided excellent blast and leaf roller control.Keywords: UAV, adjuvant,biological efficacy, deposition, low-volume spray, riceDOI: 10.33440/j.ijpaa.20200302.86作者：王國賓，Xuan Li, John Andaloro, 陳鵬超，宋燦燦，單常峰，陳盛德，蘭玉彬*摘要：農藥高效、精準噴施可以有效提高生物效果、降低害蟲抗性、提高對環境的保護。當前植保無人機作為新的農藥低容量噴施技術發展迅速，需要與傳統的噴施技術進行科學系統的對比評估。本文的試驗目的分析植保無人機噴液量及噴霧助劑對霧滴沉積、霧滴在冠層中的穿透性以及對水稻稻瘟病、稻縱卷葉螟的防治效果，並採用電動背負式噴霧器作為標準方法進行對比。試驗結果表明，增加噴液量及添加助劑能顯著增加霧滴密度、覆蓋度以及對水稻稻瘟病、稻縱卷葉螟的防治效果。植保無人機的防治效果與背負式噴霧器防效基本一致。試驗數據表明，無人機噴液量為18L/hm2並添加改性植物油類噴霧助劑能實現對水稻稻瘟病、稻縱卷葉螟的良好防治效果。關鍵詞：無人機；噴霧助劑；防治效果；霧滴沉積；低容量噴灑；水稻DOI:10.33440/j.ijpaa.20200302.86Nitrogen detection of Dendrobium nobile based on hyperspectral imagesTeng Long, Yongbing Long, Houcheng Liu, Hailin Liu, Zaihua Wang*, JingZhao* Abstract: Dendrobium nobile Lindl. (D. nobile), as an important traditional Chinese medicine and highly ornamental value plant, has attracted more and more people’s attention. In order to meet the needs of tracking and testing the growth status of D.nobile, the visible-near infrared hyperspectral imaging technology was proposed for nitrogen nutrients detectionin vivo in its different growth stages. Firstly, collecting the hyperspectral images of D.nobile in spectral range 400-1000nm. Secondly, extracting the region of interesting (ROI). The 2G-R-B algorithm was used to segment the background and plants, and then the RGB threshold method was used to separate the leaf sheath and stems. Removing noise by two masks』or-operations, and then the ROI area was finally extracted by selecting the largest area. After that, the reflectance spectrum of the ROI area was extracted, and then two kinds of feature extraction methods and two kinds of optimizing band selection methods were researched for dimension reduction of hyperspectral images. Finally, Support vector machine (SVM) model was established to classify the nitrogen level of D. nobile. The results showed that the LDA combined with the SVM algorithm had the highest classification accuracy. The classification accuracy of training sets in the three growth stages were 97.47%, 95.03%, and 95.97%, respectively, and the classification accuracy of test set reached 97.00%, 88.8%, 92.67%. The visible-near infrared hyperspectral imaging technology combining LDA-SVM classification model could effectively distinguish D. nobile cultivated by gradient nitrogen in each growth stage. It is a potential technology applied in decision-making of precise nutrition supply.Keywords: facility gardening, hyperspectral imaging technology, Dendrobium nobile Lindl.,nitrogendetection, SVMDOI: 10.33440/j.ijpaa.20200302.87作者：龍騰，龍擁兵，劉厚誠，劉海林，王再花*，趙靜*摘要：金釵石斛作為傳統貴重中藥和極富觀賞價值的植物，引起了人們越來越多的關注。為了滿足跟蹤檢測金釵石斛生長狀況的需要，本文提出了一種可見-近紅外高光譜成像技術，用於檢測不同生長階段的金釵石斛氮素營養水平。首先，採集400-1000nm範圍內的金釵石斛高光譜圖像。進而，提取感興趣區域（ROI）。利用2G-R-B算法對背景和植株進行分割，然後採用RGB閾值法對葉鞘和莖進行分離。通過兩步掩膜的或運算去除噪聲，並選擇最大區域，最終實現ROI區域提取。接著，提取ROI區域的反射光譜，研究主成分分析PCA、線性判別分析LDA等降維方法和遺傳算法GA、決策樹DT等特徵波段提取方法以降低高光譜數據維度。最後，建立支持向量機（SVM）模型，對梯度氮素供給的金釵石斛進行分類。結果表明，LDA-SVM模型具有最高的分類精度。三個生長階段訓練集的分類準確率分別為97.47％，95.03％和95.97％，測試集的分類準確率分別為97.00％，88.8％，92.67％。結合LDA-SVM分類模型的可見-近紅外高光譜成像技術可以有效地區分每個生長階段不同氮素栽培的金釵石斛，是一項有望應用於精準營養供給決策的檢測技術。關鍵詞：設施園藝；高光譜成像；金釵石斛；氮素檢測；支持向量機DOI: 10.33440/j.ijpaa.20200302.87Automatic difference vegetation index generator for spider mite-infested cotton detection using hyperspectral reflectanceHuasheng Huang, Jizhong Deng, Yubin Lan*, Aqing Yang*,Yan Jiang,Gaoyu Suo, Pengchao Chen
Abstract: Spider mites are one of the main pest stresses on cotton,which cause serious economic losses in cotton production in Xinjiang Region. This article explored the potential of ground based hyperspectral reflectance for mite-infestation detection. Also, the possibility to reduce bands and simplify analyzing was studied. In this regard, an automatic difference vegetable index was researched, which required only two bands and a simple subtraction operation. A multi-objective genetic algorithm was proposed for band selection, and its performance was compared with the main stream machine learning methods. Experimental results showed that the proposed approach outperformed others in accuracy with less complexity. All the results revealed that the proposed method has potential in mite-infestation detection in agricultural applications.Keywords: hyperspectral, spider mite, genetic algorithm, bandselectionDOI: 10.33440/j.ijpaa.20200302.88基於地面高光譜數據的棉花紅蜘蛛檢測的差異植被指數自動生成研究作者：黃華盛，鄧繼忠, 蘭玉彬*,楊阿慶*, 江龑,索高宇, 陳鵬超
摘要：紅蜘蛛是棉花的主要蟲害之一，對新疆地區的棉花生產造成嚴重的經濟損失。本文探討了地面高光譜數據在紅蜘蛛蟲害檢測中的應用潛力。此外，還研究了減少光譜波段和簡化分析方法的可能性。研究了一種差異植被指數的自動生成方法，該植被指數僅需要兩個波段和一個簡單的減法運算。本文提出了一種多目標遺傳算法進行波段選擇，並將其性能與主流機器學習方法進行了比較。實驗結果表明，該方法在精度上優於其他方法，且複雜度較低。所有結果表明，該方法在農業生產場景下的紅蜘蛛檢測具有應用潛力。DOI:10.33440/j.ijpaa.20200302.88Estimating the severity of sugarcane aphids infestation on sorghum with machine visionXiaoling Deng, J. Alex Thomasson, N. Ace Pugh, Junxi Chen, William L.Rooney, Michael J. Brewer, Yeyin Shi*Abstract: Sugarcane aphid (SCA), Melanaphis sacchari, is one of the most prominent insect pests of grain, forage and bio-energy sorghum in the southern US since 2013. The timing and dosage of a pesticide application for SCA depend on a close monitoring of its pressure or severity change in the field. To assist the field scouting, digital images were taken using a smartphone in proximity of infected leaves and corresponding image processing algorithms were developed later to estimate the infestation severity in this study. Image samples were grouped into four classes according to the infestation severity for aphid management considerations: no threat (0-10 SCA/leaf), insecticide use should be considered (11-125 SCA/leaf), insecticide should be used and yield loss likely (126-500 SCA/leaf), and plant death possible (more than 500 SCA/leaf). With 5-fold cross validation, results showed that the best average classification accuracy across the four SCA classes was 85.0% with the modified OVO-SVM algorithm. The SCA quantification accuracies achieved inthis study using the SVM algorithm showed the promise of using machine learning algorithms in this case of aphid density estimation on sorghum leaves. The methodology developed in this study can be modified with more sophisticated machine learning algorithms and more datain the future to be incorporated into a handheld or a mobile remote sensing system to assist growers and researchers with automatically quantifying SCA in a fast and objective manner.Keywords: IPM, machine vison, SVM, sugarcane aphid, severityestimationDOI:10.33440/j.ijpaa.20200302.89作者：鄧小玲, J. Alex Thomasson, N. Ace Pugh, 陳俊熹, William L. Rooney, Michael J. Brewer, Yeyin Shi*摘要:自2013年以來，甘蔗蚜蟲(SCA)是美國南部穀物、飼料和生物能源高粱最突出的害蟲之一。施用SCA農藥的時間和劑量取決於對田間壓力或嚴重程度變化的密切監測。為了協助野外偵察，研究者使用智慧型手機在受感染的樹葉附近拍攝了數字圖像，隨後開發了相應的圖像處理算法來估計本次研究中蟲害的嚴重程度。根據蚜蟲感染的嚴重程度，將圖像樣本分為4類，考慮蚜蟲管理:無威脅(0-10SCA/葉)、應考慮使用殺蟲劑(11-125SCA/葉)、應使用殺蟲劑、可能造成產量損失(126-500SCA/葉)以及可能導致植物死亡(超過500 SCA/葉)。通過5倍交叉驗證，結果表明，改進的OVO-支持向量機算法在四個SCA類中的平均分類準確率為85.0%。本研究中使用支持向量機算法實現的SCA量化精度，顯示了使用機器學習算法估計高粱葉片上蚜蟲密度的前景。在這項研究中開發的方法可以在未來通過更複雜的機器學習算法和更多的數據被整合到手持或移動感知系統中，以幫助種植者和研究人員以快速和客觀的方式自動量化SCA。關鍵詞:IPM；機器視覺；支持向量機；甘蔗蚜蟲；嚴重程度估計DOI:10.33440/j.ijpaa.20200302.89非常歡迎更多的專家和學者通過我們的平臺與同行和領域關注者們分享自己的研究成果，讓我們同心協力，共同推動精準農業航空學科發展、學術繁榮和技術進步，服務現代農業高質量發展。