Prediction of raw produce surface area from weight measurement

Abstract

Determination or estimation of the surface area of various foods can have a wide variety of applications for food producers and processors. The objective of this study was to develop a computer vision technique to rapidly determine the surface area of objects with irregular shape. The imaging system developed acquires and stores images of multiple projections of an object. Surface fitting and approximation of a 3-D wire-frame model were used to calculate object surface area for apples, cantaloupes, strawberries and tomatoes. Weight and surface area measurements of each fruit set were used to develop equations to predict fruit surface area (cm 2) from raw fruit weight. One application of these equations can be for food microbiologists to obtain more accurate determination of surface microbial populations. Analysts can gravimetrically sample whole produce to estimate product surface area and, therefore, report microbial concentrations per area.