ISHS Acta Horticulturae 406: II IFAC/ISHS Workshop :Mathematical & Control Applications in Agriculture & Horticulture
FEATURE-EXTRACTION FROM DIGITIZED IMAGES FOR OBJECTIVE QUALITY GRADING: CHICORY GRADING BASED ON SIZE, SHAPE AND COLOR |
| Authors: | E. Schrevens, R. De Busscher, I. Paulus |
| Abstract:
Decisions about picking time, quality sorting and consumers' behavior with regards to horticultural products, are mainly based on a subjective, visual interpretation of external characteristics like size, shape, color and color distribution, texture and structure of the surface, defects, contamination's, … . All the essential information to quantify these characteristics is embedded in one or more digitized video images of the product. Feature-extraction from images leads to a methodology for objective quality characterization. As a model-object, chicory heads are used in this paper. The emphasis lays on the methods to quantify the above parameters. In addition to classical feature extraction, color quantification is carried out using both discrete, classification approaches (Projection Pursuit Classification) and continuous procedures (Principal Components). The set of quality descriptors is then used to investigate objective quality grading procedures. The objective quality grading (based on feature-extraction) is compared with existing auction classes (based on visual inspection and expert knowledge), through Tree Based Modeling. | |
Download Adobe Acrobat Reader (free software to read PDF files) | |
