This is Madokoro Lab's
official website.


Link to lecture video on Yumenavi site.

おしらせ / What's New

2023年10月02日 研究生配属。 New member assignment
2023年04月03日 学部生配属。 New member assignment
2023年03月16日 レイアウト変更。 Renew layout
2022年07月01日 夢ナビ講義動画掲載。 Lecture video on Yumenavi site
2022年04月01日 学部生配属。New member assignment
2021年12月22日 暫定版ホームページ開設。 Draft website open
研究室紹介 / Introduction

概要 / Outline

Our lab focuses on fundamentals and applications of machine learning.

研究紹介動画概要編(5分間)/ Research Digest Video (5 mins)

研究室構成員 / Lab Members

教員 / Teacher

- 教授 間所洋和 博士(工学) / Hirokazu Madokoro, Professor, Ph.D.

大学院生 / Postgraduate Students

- 院1年生 2名 / 2 students (M1)

研究生 / Research Students

- 1名 / 1 students

学部生 / Undergraduate Students

- 学部4年生 3名 / 3 students (B4)

- 学部3年生 3名 / 3 students (B3)

研究内容 / Research Content


Our lab conducts extensive research on deep learning, a fundamental technology at the core of artificial intelligence, spanning from foundational studies to practical applications. In our fundamental research, we delve into the inner workings of deep neural networks, continuously enhancing learning theories and mechanisms, with the goal of developing innovative models. In our applied research, we take inputs such as images and point clouds, progressively advancing learning based on processing objectives, aiming to create implementations that enhance accuracy and efficiency. The applications of deep learning are diverse, but our lab primarily focuses on areas such as smart agriculture, forestry, fisheries, human behavior understanding, and environmental sensing to contribute to the realization of a low-carbon society.

深層学習の研究 / Study on Deep Learning

深層学習の台頭により、機械学習の応用が急速に進んでいます。 現在の深層学習は、大量のデータが確保できない場合、精度が急激に低下することが問題として挙げれられています。 機械学習の中でも、非教示学習に基づく自己組織化写像と適応的進化に関してこれまで研究してきました。 独自開発の学習方式により、少量データから位相構造やスパース特徴を導出できる。 この方式を深層学習に組み込むことにより、特に画像処理における少量データから、 認識精度の向上に寄与する基礎研究に取り組んでいます。

The rapid advancement of deep learning has greatly accelerated the applications of machine learning. However, a significant challenge in current deep learning is the drastic decrease in accuracy when dealing with limited data availability. To address this issue, our research has focused on self-organizing maps based on unsupervised learning and adaptive evolution. By incorporating our proprietary learning methods, we are able to extract topological structures and sparse features from small datasets. This foundational research aims to contribute to improved recognition accuracy, particularly in image processing, using limited data.

スマート農林水産業の研究 Study on Smart Agriculture, Forestry, Fisheries

リモートワークはオフィス労働者だけでなく、農業従事者にとっても魅力的かつ未来的な働き方といえます。 スマート農業のひとつとして、リモート農業の技術が確立できれば、後継者不足の解消に加えて、食料自給率の改善、都市住民による農業参加、作物生育からの食育など、その恩恵や利益は計り知れません。 特に、AIと小型移動ロボットに着目して、具体的には、 「小型移動ロボットと隊列ドローンによる圃場モニタリング」 「深層学習と複合センサによる作物生育と病害の自動判定」 「音源方位の位相差推定による水禽類と猛禽類の即時検出」 などの研究課題に取り組んでいます。

Remote work is not only attractive and futuristic for office workers but also holds great potential for agricultural practitioners. Establishing the technology of remote agriculture as a part of smart agriculture brings numerous benefits, such as resolving the issue of labor shortage, improving self-sufficiency in food production, encouraging urban residents' participation in agriculture, and promoting food education through crop cultivation. With a focus on AI and small mobile robots, our specific research projects include "Field monitoring using small mobile robots and drone formations," "Automatic detection of crop growth and disease using deep learning and composite sensors," and "Immediate detection of waterfowl and raptors through phase difference estimation of sound sources."

ヒューマンセンシングの研究 / Study on Human Sensing

パターン認識の要素技術を用いて、人間を対象とした非接触センシングによる動作の分類と認識・推定に関する共同研究を行っています。 具体的には、 「ドライバの表情センシングによる漫然運転と注意散漫状態の検出」 「無電源非拘束センサによるベッドモニタリングシステムの研究」 などの研究課題に取り組んでいます。

We are also engaged in collaborative research using pattern recognition techniques for non-contact sensing, specifically in the classification, recognition, and estimation of human movements. Some of our research projects include "Detection of drowsy driving and inattentive states through driver's facial sensing" and "Development of non-powered unconstrained sensor for bed monitoring system."

低炭素社会実現に関わる研究 / Study on Low Carbon Emissions

地球温暖化は年々深刻度を増しています。 温室効果ガスの中でも3/4を占めるCO2は、温暖化に及ぼす影響が最も大きいことがわかっています。 大量生産や消費が前提の現代社会では、一足飛びに解決の難しい問題ですが、 低炭素社会の実現に向けて、センサ、AI、ドローンをコア技術に、 「CO2の鉛直プロファイルを現場観測するドローンの開発」 「CCS(炭素地下貯留)のための大規模露頭画像のセグメンテーション」 「マイクロフォンアレイによる洋上風力発電装置のリモート点検」 などの研究課題に取り組んでいます。

The severity of global warming continues to escalate. Among greenhouse gases, CO2, which accounts for three-quarters of them, is known to have the most significant impact on global warming. While addressing this issue is challenging in our current society, which relies heavily on mass production and consumption, we are dedicated to achieving a low-carbon society through core technologies such as sensors, AI, and drones. Our research projects encompass "Development of drones for onsite observation of vertical profiles of CO2," "Segmentation of large-scale outcrop images for CCS (Carbon Capture and Storage)," and "Remote inspection of offshore wind turbines using microphone arrays."

研究紹介動画詳細編(90分間)/ Detailed Research Video (90 mins)

研究紹介動画特別編(15分間)/ Special Research Video (15 mins)

講義 / Lectures

大学院 / Graduate School

  • サイエンスコニュニケーション / Science Communication

  • 機械知能学特論 / Advanced Machine Intelligence

    学部 / Undergraduate School

  • 科学技術史 / History of Science and Technology

  • 情報リテラシ / Information Literacy

  • ソフトウェア情報学概論 / Survey of Software Informatics

  • プロジェクト演習 / Project Seminar

  • 人工知能演習I&II / AI Seminar I&II
  • 研究実績 / Publications


  • A. Suetsugu, H. Madokoro, T. Nagayoshi, T. Kikuchi, S. Watanabe, M. Inoue, M. Yohsida, H. Osawa, N. Kurisawa, and O. Kiguchi, "Development and Field Testing of a Wireless Data Relay System for Multiple Amphibious Drones," Drones (Special Issue: Wireless Networks and UAV), vol.8, no.38, 2024. (doi:10.3390/drones8020038) PDF
  • H. Madokoro, K. Sato, S. Nix, S. Chiyonobu, T. Nagayoshi, and K. Sato, "OutcropHyBNet: Hybrid Backbone Networks with Data Augmentation for Accurate Stratum Semantic Segmentation of Monocular Outcrop Images in Carbon Capture and Storage Applications," Sensors (Special Issue: Machine Learning Based Remote Sensing Image Classification), vol.23, no.8809, 2023. (doi:10.3390/s23218809) PDF


  • H. Madokoro, S. Nix, and K. Sato, "Visualization and Semantic Labeling of Mood States Based on Gaze and Facial Expressions for Mental Health Self-Checking," Healthcare, vol.10, no.8, 2022. (doi:10.3390/healthcare10081493) PDF
  • H. Madokoro, K. Takahashi, S. Yamamoto, S. Nix, S. Chiyonobu, K. Saruta, T. K Saito, Y. Nishimura, and K. Sato, "Semantic Segmentation of Agricultural Images Based on Style Transfer Using Conditional and Unconditional Generative Adversarial Networks," Applied Sciences, vol.12, no.15, 2022. (doi:10.3390/app12157785) PDF
  • H. Madokoro, S. Nix, H. Woo, and K. Sato, "Mallard Detection using Microphone Arrays Combined with Delay-and-Sum Beamforming for Smart and Remote Rice-Duck Farming," Applied Sciences, vol.12, no.108, 2022. (doi:10.3390/app12010108) PDF


  • H. Madokoro, S. Nix, H. Woo, and K. Sato, "A Mini-Survey and Feasibility Study of Deep- Learning-Based Human Activity Recognition from Slight Feature Signals Obtained Using Privacy-Aware Environmental Sensors," Applied Sciences, vol.11, no.24, 2021. (doi:10.3390/app112411807) PDF
  • H. Madokoro, S. Yamamoto, K. Watanabe, M. Nishiguchi, S. Nix, H. Woo, K. Sato, "Prototype Development of Cross-Shaped Microphone Array System for Drone Localization Based on Delay-and-Sum Beamforming in GNSS-Denied Areas," Drones, vol.5, no.123, 2021. (doi:10.3390/drones5040123) PDF
  • H. Madokoro, O. Kiguchi, T. Nagayoshi, T.Chiba, M. Inoue, S. Chiyonobu, S. Nix, H. Woo, K. Sato, "Development of Drone-Mounted Multiple Sensing System with Advanced Mobility for In-Situ Atmospheric Measurement: A Case Study Focusing on PM2.5 Local Distribution," Sensors, vol.21, no.4881, 2021. (doi:10.3390/s21144881) PDF
  • H. Madokoro, S. Yamamoto, Y. Nishimura, S. Nix, H. Woo, and K. Sato, "Prototype Development of Small Mobile Robots for Mallard Navigation in Paddy Fields: Toward Realizing Remote Farming," Robotics, vol.10, no.2, 2021. (doi:10.3390/robotics10020063) PDF
  • S. Yamamoto, H. Madokoro, Y. Nishimura, and Y. Yaji, "Onion Bulb Counting in a Large-scale Field using a Drone with Real-Time Kinematic Global Navigation Satellite System," Engineering in Agriculture, Environment and Food, vol.13, no.1, pp.9-14, 2021. (doi:10.37221/eaef.13.1_9) PDF
  • H. Madokoro, S. Nix, and K. Sato, "Automatic Calibration of Piezoelectric Bed-Leaving Sensor Signals Using Genetic Network Programming," Algorithms, vol.14, no.4, 2021. (doi:10.3390/a14040117) PDF

  • 連絡先 / Contact

    岩手県立大学 ソフトウェア情報学部ソフトウェア情報学科 人工知能コース
    AI Course, Faculty of Software and Information Science, Iwate Prefectural University.