臺灣博碩士論文加值系統

菇類採收方式分成採摘(harvesting by picking)或切割(harvesting by cutting)兩種。傳統的人工採收方法存在勞動強度大、效率低、成本高等問題，且易受人為因素影響，導致採收品質不穩定。本研究提出AI採收高度辨識分法及改善文獻[1][2]機構之缺點，以最適裁切高度進行杏鮑菇與秀珍菇之完株採收研究，將採收機構優化並結合影像辨識系統來提升袋栽太空包完株採收品質。透過實際觀察人工採收過程之採摘與切割方式，提出以子實體與廢料之交接面為其最佳裁切位置，並透過YOLO神經網路學習，將辨識之裁切高度位置傳輸至刀具高度移動機構，使其位移至裁切高度進行裁切。此外，採收模式以單次一包方式進行採收。本研究所設計之改良版採收機構與影像辨識功能結合後，經實驗研究結果可提升菇類採收良率達77.94%。未來在智慧採收技術成熟後，將能使菇類產業朝智慧化更邁進一步。

There are two types of mushroom harvesting: harvesting by picking or harvesting by cutting. Traditional manual harvesting methods have problems such as high labor intensity, low efficiency, high cost, and are susceptible to human factors, resulting in unstable harvesting quality. In this study, we proposed an AI harvesting height identification method and improved the shortcomings of the literature [1][2] mechanism, and used the optimal cutting height to study the harvesting of Pleurotus eryngii and Pleurotus ostreatus, and optimized the harvesting mechanism and combined with the image recognition system to improve the harvesting quality of bagged space wrapping plants. Through the actual observation of the picking and cutting methods in the manual harvesting process, it is proposed that the interface between the fruiting body and the waste is the best cutting position, and through the YOLO neural network learning, the identified cutting height position is transmitted to the tool height moving mechanism, so that it can be displaced to the cutting height for cutting. In addition, the harvest mode is carried out in a single package. After combining the improved harvesting mechanism and image recognition function designed in this study, the experimental results showed that the yield of mushroom harvesting could be improved by 77.94%. In the future, after the intelligent harvesting technology matures, the mushroom industry will be able to take a step forward towards intelligence.

摘要......i
Abstract......ii
誌謝......iii
目錄......iv
表目錄......vi
圖目錄......vii
第一章 緒論......1
1.1 前言......1
1.2 研究動機......2
1.3 研究目的......2
1.4 論文架構......2
第二章 文獻回顧......3
2.1 傳統菇類採收方法簡介......3
2.2 智慧化技術的應用......4
2.3 影像辨識與YOLO演算法......5
2.4 智慧採收方法與專利分析......9
第三章 研究方法......16
3.1 改良設計......18
3.1.1 菇包定位機構......19
3.1.2 夾爪設計......20
3.1.3 菇包傳遞推送桿機構......21
3.1.4 菇包固定座-活動端與固定端......22
3.1.5 刀具高度調整機構......23
3.2 影像辨識功能......24
3.2.1 資料收集與標註......25
3.2.2 訓練模型環境與相關參數設置......27
3.2.3 高度計算與網頁介面設計......30
3.3 系統整合與功能測試......32
第四章 結果與討論......33
4.1 改良設計......34
4.1.1 菇包定位機構與夾爪設計......34
4.1.2 菇包固定-活動端......36
4.2 影像辨識功能......38
4.2.1 杏鮑菇......38
4.2.2 秀珍菇......39
4.3 完株採收實驗......40
4.3.1 名詞解釋......43
4.3.2 A場域之太空包採收結果分析......44
4.3.3 B場域太空包之採收分析結果......45
4.3.4 農試所場域(木屑) 太空包之採收結果分析......47
4.3.5 農試所場域(禾本科) 太空包之採收結果分析......48
第五章 結論與未來展望......50
5.1 結論......50
5.2 未來展望......51
參考文獻......52
附錄一 階段一A場域採收實驗紀錄......55
附錄二 階段一B場域採收實驗紀錄......60
附錄三 階段二農試所(木屑)採收實驗紀錄......65
附錄四 階段二農試所(狼尾草)採收實驗紀錄......69
Extended Abstract......73
I. Introduction......74
II. Literature Review......74
III. Methodology......74
IV. The Experimental Results......75
V. Conclusion......76

[1]林明毅，2020，”杏鮑菇自動化採收機構之研究”，國立虎尾科技大學機械設計工程系碩士論文。
[2]顏靖峰，2022，”袋栽杏鮑菇完株採收機構之研究”，國立虎尾科技大學機械設計工程系碩士論文。
[3]石信德、呂昀陞、邱相文、周榮源、呂椿棠、徐武煥、陳美杏、李瑋崧、黃棨揚，2019，”菇類智慧化生產之現況與展望”， 菇類智慧化生產與農場經營管理研討會專刊，頁29-34，台中，行政院農業委員會農業試驗所。
[4]石信德，2022，”Smart Mushroom Cultivation in Taiwan”，智慧農業，檢索自：
https://www.intelligentagri.com.tw/en/xmdoc/cont?xsmsid=0J170506304287007902&sid=0L286340672991900264
[5]陳美杏、李瑋崧、呂昀陞，2017，”菇類基本特性與栽培原理（一）”，農業知識入口網，檢索自：
https://kmweb.moa.gov.tw/theme_data.php?theme=news&sub_theme=variety&id=54715
[6]陳美杏、李瑋崧、呂昀陞，2017，”菇類基本特性與栽培原理（一）”，農業知識入口網，檢索自：
https://kmweb.moa.gov.tw/theme_data.php?theme=news&sub_theme=variety&id=54729
[7]4AG.ai., 2023, “4AG.ai: AI solutions for businesses.”, 檢索自：https://4ag.ai/
[8]Chang, S.T., & Miles, P.G., 2004, “Mushrooms:Cultivation, Nutritional Value, Medicinal Effect,and Environmental Impact (2nd ed.)”, CRC Press.
[9]Zhang, Y., Geng, W., Shen, Y., Wang, Y., & Dai, Y.C., 2014, “Edible mushroom cultivation for food security and rural development in China: Bio-Innovation, Technological Dissemination and Marketing.”, Sustainability, 6(5):2961-2973.
[10]Kavaliauskas, Ž., Šajev, I., Gecevičius, G., & Čapas, V., 2022, “Intelligent Control of Mushroom Growing Conditions Using an Electronic System for Monitoring and Maintaining Environmental Parameters.”, Applied Sciences, 12(24):13040.
[11]Liao, R., Zhang, S., Zhang, X., Wang, M., Wu, H., & Zhangzhong, L., 2021, “Development of smart irrigation systems based on real-time soil moisture data in a greenhouse: Proof of concept”, Fungal Biology and Biotechnology, Volume 245, pp.45-55.
[12]Moysiadis, V., Kokkonis, G., Bibi, S., Moscholios, I., Maropoulos, N., & Sarigiannidis, P., 2023, “Monitoring Mushroom Growth with Machine Learning.”, Agriculture, 13(1):223.
[13]Zied, D.C., & Pardo-Giménez, A.(Eds.), 2017, “Edible and medicinal mushrooms: technology and applications.”, John Wiley & Sons.
[14]Mirabelli, G., & Solina, V., 2020, “Blockchain and agricultural supply chains traceability: Research trends and future challenges.”, Procedia Manufacturing, 42, pp.414-421.
[15]胡原凱，2021，”袋栽菇類製包流程智慧檢測系統之研究”，國立虎尾科技大學機械設計工程系碩士論文。
[16]黃文信，2019，”基於深度學習之杏鮑菇自動化分級系統之研究”， 國立虎尾科技大學機械設計工程系碩士論文。
[17]邱弘丞，2021，”自主移動式智慧物聯網模組在菇類栽培之研究”， 國立虎尾科技大學機械設計工程系碩士論文。
[18]紀錦川，2023，”應用於菇類栽培桿測與資料擷取之ROS自主移動機器人研發”，國立虎尾科技大學機械設計工程系碩士論文。
[19]許清龍，2023，”杏鮑菇子實體三維點雲取像與平台建置之研究”， 國立虎尾科技大學機械設計工程系碩士論文。
[20]余昱輝，2023，”機械手臂整合深度視覺於杏鮑菇太空包之最適取放系統之研究”， 國立虎尾科技大學機械設計工程系碩士論文。
[21]Yu, W., Pen, J., Li, J., Deng, Y., & Li, H., 2022, “Application Research of Image Feature Recognition Algorithm in Visual Image Recognition.”, 2022 IEEE Conference on Telecommunications, Optics and Computer Science (TOCS), pp.812-816.
[22]He, K., Gkioxari, G., Dollár, P., & Girshick, R., 2017, “Mask R-CNN.”, 2017 IEEE International Conference on Computer Vision (ICCV), 2980-2988.
[23]LeCun, Y., Bottou, L., Bengio, Y., & Haffner, P., 1998, “Gradient-based learning applied to document recognition.”, Proceedings of the IEEE, vol, 86, no.11, pp.2278-2324.
[24]Redmon, J., Divvala, S., Girshick, R., & Farhadi, A., 2016, “You Only Look Once: Unified, real-time object detection.”, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, (CVPR), 779-788.
[25]Terven, J., Córdova-Esparza, D.M., & Romero-González, J.A., 2023, “A Comprehensive Review of YOLO Architectures in Computer Vision: From YOLOv1 to YOLOv8 and YOLO-NAS.”, ArXiv, abs/2304.00501.
[26]Hussain, M., & Khanam, R., 2024, “In-Depth Review of YOLOv1 to YOLOv10 Variants for Enhanced Photovoltaic Defect Detection.”, Solar, 4(3):351-386.
[27]Alif, M.A.R., & Hussain, M., 2024, “YOLOv1 to YOLOv10: A comprehensive review of YOLO variants and their application in the agricultural domain.”, ArXiv, abs/2406.10139.
[28]Wang, C.Y., Bochkovskiy, A., & Liao, H.Y.M., 2022, “YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors.”, ArXiv, abs/2207.02696.
[29]Wang, A., Chen, H., Liu, L., Chen, K., Lin, L., Han, J., & Ding, G., 2024, “YOLOv10: Real-Time End-to-End Object Detection.”, ArXiv, abs/2405.14458.
[30]Navas, E., Fernández, R., Sepúlveda, D., Armada, M., & Gonzalez-de-Santos, P., 2021, “Soft Grippers for Automatic Crop Harvesting: A Review.”, Sensors, 21(8):2689.
[31]Ji, W., Huang, X., Wang, S., & He, X., 2023, “A Comprehensive Review of the Research of the “Eye–Brain–Hand” Harvesting System in Smart Agriculture.”, Agronomy, 13(9):2237.
[32]Rong, J., Wang, P., Yang, Q., & Huang, F., 2021, “A Field-Tested Harvesting Robot for Oyster Mushroom in GreenhouseAgronomy.”, Agronomy, 11(6):1210.
[33]洪滉祐、楊朝旺、黃膺任、黃威仁、江一蘆、林金龍、王美蘭，2023，”用於鳳梨田之智慧農機”，中華民國發明專利第I827484號。
[34]周榮源，石信德，2023，”自主移動式人工智慧菇類栽培監控系統及方法”，中華民國發明專利第M431549號。
[35]游信強，吳承炎，游鈞翔，2023，”農業自動化採集設備”，中華民國發明專利第I776745號。
[36]周呈霙、江昭皚、王人正、林弘人、莊詠竣、賴怡穎、黃偉豪、張善程、陳博炤、闕振祐、李承駿、葉紹翔，2024，”作物採收機”，中華民國發明專利第I842416號。
[37]邱奕志、陳世銘、林家鋒、陳毓良、李柔靜、楊本源、潘姵如、蔡錦銘、張詠惠，2012，”蔬果自動採收系統”，中華民國發明專利第M431549號。
[38]周榮源，石信德，2021，”袋式栽培菇類自動化採收系統及其方法”，中華民國發明專利第I728544號。
[39]周榮源，石信德，2022，”菇類採收系統”，中華民國發明專利第I719925號。
[40]Roboflow., 2023, “Roboflow: The computer vision platform for developers.”, 檢索自:
https://roboflow.com/
[41]Wong, K.Y., 2023, “YOLOv7: A state-of-the-art object detection model.”, GitHub, 檢索自: https://github.com/WongKinYiu/yolov7.git
[42]OpenCV., 2024, “ArUco標記檢測教程”, OpenCV官方文檔上, 檢索自：https://docs.opencv.org/4.x/d5/dae/tutorial_aruco_detection.html
[43]Okalachev, A., 2023,Arucogen: A tool for automatic generation of AR markers., GitHub, https://github.com/okalachev/arucogen
[44]Flask., 2024, Wikipedia, 檢索自：https://zh.wikipedia.org/zh-tw/Flask