International Agrophysics · 2010 · 22 citations · 16 references
AgricultureComputer ScienceFood ScienceColor TheoryComputer VisionMachine VisionAgricultural Machine VisionColor AnalysisColor QualityFruit QualityFruit Quality EvaluationImage ClassificationClassification AccuracyClassification Accuracy AnalysisFruit Color EvaluationHueImage Processing AlgorithmsImage Signal ProcessingMachine Vision SensingMachine Vision Systems
A b s t r a c t. A machine vision system was introduced for the evaluation and classification of the Iranian saffron peach. Physical features such as size and colour were measured to categorize peaches into three quality classes of red-yellow, yellow-red, and yellow. The HIS model hue, saturation, and luminance was used for colour processing of flawless samples and four boundaries were selected for the peach size image. The colours of peaches estimated by the system were well correlated with the colorimetric index values that are currently used as standards. Experimental results are promising and suggest that using the USDA standard, the size classification accuracy achieved is almost 96%, while the colour classification accuracy is around 90%, and that the spot detection algorithm performs well with correct detection levels of 97 and 85% for brown and white skin spots, respectively. K e y w o r d s: saffron peach, machine vision, image processing, size, colour
16