6533b85ffe1ef96bd12c1199

RESEARCH PRODUCT

A Rapid and Efficient Method for Determination of Fruit Peel Color

subject

description

Peel color is a critical index of external fruit quality and consumer appreciation level. Traditional methods for determination of peel color are based on visual analysis or punctual measurements by colorimeter. In this study we present a method based on digital image analysis that integrates the accuracy of an interactive measurement and the efficacy of an image analysis that descibes entire sides of the fruit. A sample of apple, mandarin, grape, and peach fruit was photographed (each fruit on two opposite sides) with a digital camera for determination of peel color. Digital images were converted from RGB to CIE L*a*b* format, and color characteristics were indexed and quantified. The implemented method uses a reference color image cropped from the best fruit (interactively chosen) to calculate a color index for each fruit of the image set. The final index is the weighted sum of the number of pixels of the fruit, where pixels closer to the reference color (distance in the CIE L*a*b space) are considered more relevant. This color index gives integrated information on fruit color quantity (% of cover color) and quality (hue and saturation). The method represents a rapid and efficient way of determining color of the entire fruit surface and overcomes difficulties and approximation of traditional methods. A modification of the same method can be used to determine peel rugosity (mandarin) and average size and number of grape berries per bunch.