Publication: Insect Detection on an Unmanned Ground Rover
1
Issued Date
2019-02-22
Resource Type
ISSN
21593450
21593442
21593442
Other identifier(s)
2-s2.0-85063186024
Rights
Mahidol University
Rights Holder(s)
SCOPUS
Bibliographic Citation
IEEE Region 10 Annual International Conference, Proceedings/TENCON. Vol.2018-October, (2019), 954-959
Suggested Citation
Ananta Srisuphab, Piyanuch Silapachote, Warittha Tantratorn, Pongsatorn Krakornkul, Prompiriya Darote Insect Detection on an Unmanned Ground Rover. IEEE Region 10 Annual International Conference, Proceedings/TENCON. Vol.2018-October, (2019), 954-959. doi:10.1109/TENCON.2018.8650312 Retrieved from: https://repository.li.mahidol.ac.th/handle/123456789/50647
Research Projects
Organizational Units
Authors
Journal Issue
Thesis
Title
Insect Detection on an Unmanned Ground Rover
Other Contributor(s)
Abstract
© 2018 IEEE. Crop losses and damages caused by insects and pests negatively impact nations' economies worldwide, particularly in agricultural producing countries. Farmers struggling to maintain both the product yield and profit as well as to make ends meet often resort to the apparently easy solution of using insecticides. A health hazard, exposure to chemical pesticides, may result in acute poisoning and long-term chronic toxicity. Applying information technology to contribute to agricultural innovation, we propose an approach to detect the presence of flying insects and to monitor insect populations in an open field. Simple yet effective computer vision and image processing techniques are employed. Moving objects in the scenes are detected, then classified whether or not they are flying insects. Instead of requiring an installation of either equipment and tools or electronic sensors in and around the field, our small-sized video-camera is mounted on a remotely controlled big-rock crawler. A user only needs to specify a search area of interest and point out locations over where the rover cannot go. The system then runs a path finding algorithm to lay out a mission, which the rover is programmed to execute unmanned. In our experiment, the rover successfully followed a desired mission. Its accuracy was evaluated and the effects of terrain conditions were analyzed. The fully-integrated system has achieved an insect detection accuracy of 89.43%.