Computer-generated artificial classifiers that are intended to
mimic human decision making for grading and product quality
have recently been studied intensively. The online lentil color classification
using a flatbed scanner with neural classifier developed
by Shahin and Symons (2001) achieved an overall accuracy of more
than 90%. Leemans et al. (2002) developed an on-line fruit grading
system based on external quality features of two varieties of apples,
Golden Delicious and Jonagold, using quadratic discriminant
analysis and neural networks. The image grading was achieved in
six steps: image acquisition; ground color classification; defect
segmentation; calyx and stem recognition; defects characterization;
and finally the fruit classification into quality classes. Both
algorithms resulted in similar results (79% and 72%) for both varieties
studied. Blasco et al. (2003) combined machine vision techniques
with Bayesian discriminant analysis for online estimation
of the quality of oranges, peaches, and apples, and evaluated the
efficiency of these techniques regarding the following quality attributes:
size, color, stem location, and detection of external
blemishes.