Abstract
An automatic recognizing and evaluating method of double flowered stock seedlings was investigated as a basic research project to develop an automatic transplanting system. Three basic characteristics (area, complexity degree and saturation of the cotyledon) of the seedling image were measured by a color image processing system to fix the membership function of fuzzy sets. We have made the algorithm of the discrimination between double and single flowering forms in the seedling stage of stock based on the fuzzy evaluation theory. The accuracy of discrimination was 90.8%, and the availability of the seedling was 39.4% in stock plant 'Sourei'.
