Paper Abstract and Keywords |
Presentation |
2023-03-16 14:25
Pear flower pollination path estimation using deep learning and combinatorial optimization Keita Endo (NIT), Tomotaka Kimura (Doshisha Univ.), Hiroyuki Shimizu (NIT), Tomohito Shimada, Akane Shibasaki (SATRC), Yoshihiro Takemura (Tottori Univ.), Takefumi Hiraguri (NIT) CQ2022-95 |
Abstract |
(in Japanese) |
(See Japanese page) |
(in English) |
Since fruit trees are grown outdoors, they require heavy and long hours working. In among fruit trees, pears needs artificial pollination, and labor saving is required. In this study, we propose the pollination systems that sprays pollen to pear flowers using a drone. In the proposed scheme, the drones takes an aerial image of the entire pear field from above, and by segmentation using deep learning, the range of the field and the space where the drone can fly are detected from the aerial image. Based on the detected results, image processing and combinatorial optimization are used to estimate the optimal route for autonomous flight that enables pollination work. |
Keyword |
(in Japanese) |
(See Japanese page) |
(in English) |
Smart agriculture / Deep learning / Drones / Image analysis / Segmentation / / / |
Reference Info. |
IEICE Tech. Rep., vol. 122, no. 438, CQ2022-95, pp. 74-77, March 2023. |
Paper # |
CQ2022-95 |
Date of Issue |
2023-03-08 (CQ) |
ISSN |
Online edition: ISSN 2432-6380 |
Copyright and reproduction |
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
Download PDF |
CQ2022-95 |
|