Msc thesis subject: Detecting, phenotyping and mapping of turfgrass vegetation

The risk of herbicides resistance, health risks associated with herbicide exposure and environmental concerns lead to a restriction of pesticides in the European Union. A ban of herbicides was identified as the most problematic aspect for turfgrass managers, because weeds establish opportunistically and are capable to invade large areas in public green space, golf courses, football pitches and so forth. The presence of weed species in turfgrasses reduces the playing quality, the aesthetic value and the usability.

The green deal restricts any use of herbicides in turfgrass areas. Therefore, turfgrass managers require tools to detect, phenotype and map out weeds. Weeds can then be spot sprayed or mechanically removed. This research project involves taking RBG & multispectral images of two field trials over time, creating an ortho image, classifying vegetation in GIS as well as extracting and analysing data. Furthermore, fieldwork is required to take ground truth data, which will be compared with the digital image analysis approach.


  • Classifying vegetation composition in turfgrass areas over time (grass, weeds, soil)
  • Comparing ground truth data with digital image data
  • Comparing methodologies (multispectral images VS RGB VS ground truth data)


  • Hutto, K. C., Shaw, D. R., Byrd, J. D., & King, R. L. (2006). Differentiation of turfgrass and common weed species using hyperspectral radiometry. Weed Science, 54(2), 335-339.
  • Stiegler, J., Bell, G., Maness, N., & Smith, M. (2005). Spectral detection of pigment concentrations in creeping bentgrass golf greens. Int. Turf. Soc. Res. J, 10, 818-825.


  • Field work is required at Heelsum golfcourse and Wolfheze (Barenburg research station)

Theme(s): Sensing & measuring, Modelling & visualisation