Development of automatic weeding system in loan of golf course (1st report) - Investigation of the automatic weed detecting method

Yoshiyuki Miwa, Akitoshi Itoh, Hisao Mashita, Satoshi Shinozaki

Research output: Contribution to journalArticle

Abstract

According to the increase of the concern to the ecological problem, the use of agricultural chemicals for weeding in the golf course is made the social problem. Therefore, automatic machine weeding is expected This study, first, surveys what is the weeding work of golf course. Next, it investigates how to detect the weed in the loan. Sensors of several principles are trially made and examined comparatively at the "fairway" and "green" of golf course. Examined sensors are follows: (1) Detecting from the difference of the color (image processing by CCD camera, color sensor with band pass filter). (2) Detecting from the difference of the water percentage (microwave transmitter and receiver). (3) Detecting from the difference of the tactile feeling (tactile sensor made by urethane rubber with strain gauge). The result shows that any sensor can detect the weed in the loan. Sensor output, however, changes depending on the season, time and weather Therefore, to realize the automatic weed detecting system, it is very important to develop the "sensor fusion" technique by using the plural sensors of the different methods.

Original languageEnglish
Pages (from-to)1602-1606
Number of pages5
JournalSeimitsu Kogaku Kaishi/Journal of the Japan Society for Precision Engineering
Volume62
Issue number11
Publication statusPublished - 1996 Nov
Externally publishedYes

Keywords

  • Agricultural chemical
  • Automatic weed detection system
  • Ecological problem
  • Fairway
  • Golf course
  • Green
  • Loan
  • Sensor
  • Weed

ASJC Scopus subject areas

  • Mechanical Engineering

Fingerprint Dive into the research topics of 'Development of automatic weeding system in loan of golf course (1st report) - Investigation of the automatic weed detecting method'. Together they form a unique fingerprint.

  • Cite this