Sensor comparison for grass growth estimation

Md Shaha Nur Kabir, Sun Ok Chung, Yong Joo Kim, Geung Joo Lee, Seung Hwa Yu, Kyeong Hwan Lee, Takashi Okayasu, Eiji Inoue

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)

Abstract

Precision agriculture has been implemented in various cultivation operations for various crops. Recently, autonomous mower tractor with variable rate fertilization has been tried. Variable rate fertilization requires quantification of grass growth level. Objective of this study was to compare the performance of potential sensors under different growth levels and sensor operation conditions. A CCD camera, an ultrasonic module, and an optical reflectance sensor (i.e., Crop Circle) were tested at fields with different grass growth levels. Effects of sensor mounting height and angle were investigated. Effects of vehicle (i.e., mower tractor) vibration and traveling were also investigated: under static, vibration and no traveling, and vibration with traveling conditions. Results showed that sensor mounting height had a little effect on the sensor performance, and grass growth level was successfully quantified. Camera angle of 90° showed the best performance with less noise. The ultrasonic sensor could work with less deviation under static condition, whereas, CCD camera proved the applicability under all operating conditions. Operation condition of the optical reflectance sensor showed a little effect on the grass growth detection results. This study would provide information useful for automated grass growth mapping and variable fertilization recommendation.

Original languageEnglish
Pages (from-to)367-374
Number of pages8
JournalJournal of the Faculty of Agriculture, Kyushu University
Volume61
Issue number2
Publication statusPublished - Sept 2016

All Science Journal Classification (ASJC) codes

  • Biotechnology
  • Agronomy and Crop Science

Fingerprint

Dive into the research topics of 'Sensor comparison for grass growth estimation'. Together they form a unique fingerprint.

Cite this