大学院農学研究科
English
更新日：2024/12/17
准教授
ｲｽﾗﾑ　ｴﾑﾃﾞｲｰ ﾊﾟｰﾍﾞｽﾞ
ISLAM　MD PARVEZ

経歴

  1. 2005/04-2008/09Bangladesh Agricultural UniversityFarm Power and MachineryLecturer
  2. 2008/09-2015/09Bangladesh Agricultural UniversityFarm Power and MachineryAssistant Professor
  3. 2015/10-2018/03Irie Koken Co. Ltd.Design and DevelopmentEngineer
  4. 2019/04-2021/03NARO農業情報研究センター 農業AI研究推進室 制御チームResearcher
  5. 2021/04-2021/08NAROResearch Center for Agricultural RoboticsResearcher
  6. 2021/09-現在愛媛大学大学院農学研究科准教授　

学位

  1. PhDBiomechanical Systems愛媛大学2015/09

研究分野

  1. 環境・農学農業環境工学、農業情報工学
  2. 環境・農学農業環境工学、農業情報工学

研究キーワード

  1. Plant factory and greenhouse
  2. Aritificial Intelligence
  3. Deep learning
  4. Intelligent control system
  5. Renewable energy

書籍等出版物

  1. Artificial Intelligence Applications in Specialty CropsMd. Parvez Islam , Yuka Nakano, Unseok Lee , Keinichi Tokuda and Nobuo KochiChapter AuthorFrontiers Media SA.2022/02978-2-88974-557-9URL
詳細表示...

論文

  1. 人工知能支援トマト植物監視システム - ユニバーサルマルチブランチ汎用畳み込みニューラルネットワークに基づく実験的アプローチ2024/09M.P. Islam, K. HatouComputers and Electronics in Agriculture224/ 109201研究論文（学術雑誌）10.1016/j.compag.2024.109201Elsevier B.V.Real-time monitoring of tomato plants in plant factories is necessary to identify and classify diseases at early stages to prevent possible outbreaks. The proposed DeepD381v4plus network exhibits higher class-wise accuracy, sensitivity, specificity, precision, F1 score and Matthews correlation coefficient scores exceeding 0.96 for multi-varietal tomato leaf diseases. During the reproductive stage, bud formation, flower appearance, bite marks and fruit set also need to be monitored to confirm pollination. The detector DeepDet381v4 – YOLOv4M achieves the highest mean average precision (mAP) (0.90) and lowest mAP (0.68) in the TFl_Blooming class and the lowest mAP (0.68) in the TFl_Transforming class. However, in real-world simulations, DeepDet381v4 – YOLOv4M can detect and count ripe tomatoes at a distance of 40 cm with little to no error. Both networks used for classification and detection–counting tasks have small sizes with high classification and detection efficiency (>27 fps). Overall, the proposed experimental approach will help farmers prevent disease outbreaks, monitor flower shapes that can set fruits at the highest rate, timely detect and count ripened fruits or recognise damaged fruits due to surface cracks or diseases for harvesting at their optimal maturity stage. This will reduce the labour costs, improve cultivation management and ensure excellent quality of the harvested tomatoes.
  2. Four-Parameter Beta Mixed Models with Survey and Sentinel 2A Satellite Data for Predicting Paddy Productivity2024/08/09Dian Kusumaningrum, Hari Wijayanto, Anang Kurnia, Khairil Anwar Notodiputro, Muhlis Ardiansyah, Islam MD ParvezSmart Agricultural Technology100525研究論文（学術雑誌）10.1016/j.atech.2024.100525Elsevier B.V.Ensuring food security, fostering agricultural sustainability, and driving economic development. However, existing prediction models often overlook the unique characteristics of paddy productivity distribution, which varies between areas, skewed, and bounded within a certain minimum and maximum range, following a four-parameter beta distribution. Consequently, these models yield less accurate and potentially misleading predictions. Additionally, most approaches fail to capture the complex interrelationships among variables that often occur when we incorporate satellite data alongside survey data that has been recognized as a key approach for improving prediction accuracy and optimizing farming practices. To address these shortcomings, this study introduces a four-parameter beta Generalized Linear Mixed Model (GLMM) augmented within a four-parameter beta Generalized Mixed Effect Tree (GMET). The four-parameter beta GMET, an extension of the four-parameter beta GLMM model integrated with a regression tree, offers enhanced flexibility in modeling complex relationships. Application of this methodology to an empirical study in Central Kalimantan and Karawang reveals notable improvements over previous methods, as evidenced by substantially lower AIC and RRMSE values. Notably, the analysis identifies lagged values of band 4, band 8, and NDVI from Sentinel-2A satellite data as significant predictors of paddy productivity, overriding the importance of farmer survey variables. This underscores the potential of satellite data to be utilized in paddy productivity predictions, offering a more efficient and cost-effective alternative to farmer survey-based methods. By enhancing satellite technology, future efforts in paddy productivity prediction can achieve higher efficiency and accuracy, contributing to informed decision-making in agricultural management.
  3. TheLR531v1 – A deep learning multi-branch CNN architecture for day-night automatic segmentation of horticultural crops2022/12/19M.P. Islam, K. HatouComputers and Electronics in Agriculture204/ 107557研究論文（学術雑誌）10.1016/j.compag.2022.107557Elsevier B.V.There is an increasing need for crop management practices to improve the day-night segmentation efficiency of crops grown in the open-field or in the greenhouse, in soil or in hydroponic environments, or both. We propose an asymmetric TheLR531v1 network with a multi-branch structure with powerful feature representation to recover plants, plant leaf or canopy (horticultural crops) from the background in real-time. The multi-branch structure of the network successfully extracts various image features from the input images and classifies them into leaf and background classes. We evaluated the network performance by using over 55,992 augmented images (tomato, eggplant, and lettuce) and obtained 94.55% training and 95.89% validation accuracy. The total parameters of TheLR531v1 are only 4.5 M, which is significantly lower than other state-of-the-art models. But realise that in addition, the proposed network achieves an average BF score of 89.00%, an IoU of 87.00%, and a GA of 96.00% to distinguish plants or plant canopy pixels from background pixels. Overall, TheLR531v1 can replace the laborious manual image segmentation tasks, analyze variability in stress conditions (temperature), and visualize stress related changes in plants, leaves or canopy (pixel area), which can improve crop management practices and yield.
  4. TheLR531v2 — An asymmetric multi-branch dilated network for day-night autonomous image segmentation of horticultural crops2022/12Md. Parvez Islam Kenji HatouThe XX CIGR World Congress 2022（MISC）研究発表要旨（国際会議）
  5. ThelR547v1—An Asymmetric Dilated Convolutional Neural Network for Real-time Semantic Segmentation of Horticultural Crops2022/11/15Islam, M.P.; Hatou, K.; Aihara, T.; Kawahara, M.; Okamoto, S.; Senoo, S.; Sumire, K.Sensors22/ 8807研究論文（学術雑誌）10.3390/s22228807
詳細表示...

講演・口頭発表等

  1. Prediction of Biogas Production from Agriculture Waste Biomass Based on Backpropagation Neural NetworkThe 8th International Conference on Green and Agro-Industry and Bioeconomy (ICGAB)2024/10/03口頭発表（一般）URL
  2. Artificial intelligence assisted multi-task tomato plant monitoring system – A case study:International Conference on Statistics and Data Science 20242024/09/27口頭発表（基調）URL
  3. 空撮画像を用いた推定SPAD値計測のための画 像処理方法の検討日本生物環境工学会2024 年豊橋大会2024/09/20口頭発表（一般）
  4. Artificial intelligence assisted multi-task tomato plant monitoring system日本生物環境工学会2024 年豊橋大会2024/09/20口頭発表（一般）
  5. Optimizing Biogas Production from Agricultural Waste Biomass Using Artificial Neural Network and Particle Swarm Optimization日本生物環境工学会2024 年豊橋大会2024/09/20口頭発表（一般）
詳細表示...

産業財産権

  1. 特許権枝葉年齢推定システム、枝葉年齢推定方法、およびコンピュータプログラム特願2023-1461402023/09/08
  2. 特許権枝葉年齢推定システム、枝葉年齢推定方法、およびコンピュータプログラム特願2023-1461402023/09/08
  3. 特許権幹の水ポテンシャルの予想方法および幹の水ポテンシャルの予想システム特願2021-1557502021/09/242023-0469072023/04/05
  4. 特許権農業用ロボット特願2021-0554212021/03/31特開2022-1525932022/10/12
  5. 特許権計測システム、計測方法、計測用プログラムおよび計測装置特願2021-0025502021/01/12特開2022-1078842022/07/25
詳細表示...

受賞

  1. 2024/04/11Best teacher award
  2. 2019/11Best Poster Award赤外線画像システムの性能評価温室の最適な環境制御
  3. 2014/12Best research Award
  4. 2012/10Research innovation award
  5. 2009/12Appreciation Award Agricultural Innovation toward Biofuel Based Society
詳細表示...

担当授業科目

  1. 2024農学入門
  2. 2024植物工場特論
  3. 2024植物工場特論
  4. 2024農業情報工学特論
  5. 2024農業情報工学特論
詳細表示...

学術貢献活動

  1. Computers and electronics in agriculture査読2024/09-2024/12Elsevier, Inc. (New York, US)
  2. Future internet査読2024/06/05-2024/12MDPI (Basel, CH)
  3. Iot査読2024/05/20-2024/12MDPI (Basel, CH)
  4. Smart agricultural technology査読2024/01-2024/12Elsevier, Inc. (New York, US)
詳細表示...

メディア報道

  1. AI でトマトの生産を監視愛媛経済レポート愛媛経済レポート2024/12
  2. 「ＡＩでトマトの実や花の形を監視」愛媛大が新システム開発テレビ・ラジオ番組nhk愛媛 NEWS WEB2024/08/23
  3. トマト栽培AI監視開発新聞・雑誌日刊工業新聞2024/08/12
詳細表示...

所属学協会

  1. Japanese Society of Agricultural Informatics
  2. 日本生物環境工学会
詳細表示...

担当経験のある科目

  1. 農業情報工学農学研究科（修士課程) 国立大学法人　愛媛大学2021/10
  2. 人工知能入門農学研究科（修士課程) 国立大学法人　愛媛大学2022/04学部専門科目日本人工知能入門
  3. 農業におけるAI農学研究科（修士課程) 国立大学法人　愛媛大学学部専門科目日本農業におけるAI
  4. 農業情報工学特論 農学研究科（修士課程)農学研究科（修士課程) 国立大学法人　愛媛大学2022/04/01大学院専門科目日本修士課程
  5. Farm MechanicsBangladesh Agricultural University
詳細表示...