• Brassica napus;
  • digital analysis;
  • leaf area index;
  • number of plants;
  • soil cover;
  • vegetative developmental stages


In our experiment digital image processing is used to predict characteristics in a winter oilseed rape canopy. A large series of images was taken in 2002–2003 in close intervals from a measuring area of 1 m2. These images were automatically evaluated by a self-written computer program analysing the red/green/blue colour channels of each pixel. The number of determined green pixels was then related to the total pixel count of the image. Image evaluation helped to determine canopy structure by digital image analysis subjected to several applications, i.e. soil coverage, leaf area index (LAI), dynamics of plant number during vegetative developmental stages including entire winter season. Furthermore, number of plants per m2 and position of each plant were determined by image analysis. Results show that all parameters are dynamic during the vegetative developmental stages (germination–beginning of flowering) mainly depending on temperature. During the vegetative developmental stages number of plants varied. Emergence lasted 30 days resulting in large differences in growth and development of individual plants. During winter number of plants decreased due to longer phases of frost. Plant growth indicated by dynamics of LAI alternated with phases of cessation due to low temperatures above zero or frost. Reductions in soil coverage and LAI clearly started at daily mean temperatures below 5 °C. After the analysis, differences in LAI as well as changes in number of plants during the early phase crop development can serve (i) as input parameters to growth models, (ii) to improve canopy reflectance measurements by separating spectral signatures of soil and canopy and (iii) to determine and explain heterogeneities within the canopy.