A review on remote sensing of weeds in agriculture

K. R. Thorp, L. F. Tian

Research output: Contribution to journalArticlepeer-review


In the effort of developing precision agriculture tools, remote sensing has been commonly considered as an effective technique for weed patch delineation, where weed infestations are detected based on variations in the plant canopy spectral response. Because the canopy spectral response is important for weed detection, discussions on the irradiative interaction of light in plant canopies and the effect of variable soil background on the canopy spectral response are presented in this review. Also, a presentation of the current techniques for removal of soil effects, including vegetation indices and spectral mixture analysis, shows that these techniques have not been adequately developed for use in remote sensing-based weed detection applications. Given the nature of light interaction in a plant canopy, this review proposes that the spectral response of a plant canopy depends on both the species and the biomass density. Remote detection of weeds from ground-, aircraft-, and satellite-based platforms has been accomplished on a wide scale, yet the use of these weed detection methods to make variable-rate herbicide applications has not occurred as often. By judging success based on variable-rate herbicide applications rather than precise weed localization, some of the current problems in weed sensing may be skirted.

Original languageEnglish (US)
Pages (from-to)477-508
Number of pages32
JournalPrecision Agriculture
Issue number5
StatePublished - Oct 2004


  • remote sensing
  • vegetation indices
  • vegetative reflectance
  • weed detection
  • weed distribution

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)

Fingerprint Dive into the research topics of 'A review on remote sensing of weeds in agriculture'. Together they form a unique fingerprint.

Cite this