Evaluating the Shape and Size Characteristics of Sweetpotatoes Using Digital Image Analysis

Abstract.

Consumers of fresh produce make their decision to purchase based primarily on the appearance of quality and shape is a significant quality factor. Current methods for evaluating sweetpotato size and shape are subjective and not scientifically rigorous. In addition to genetics, it is well known that soil, cultural practices and weather can have a profound effect on the average shape of the roots in a field. A Keyence LJ-V7300 high speed laser scanner is being used to measure individual roots. The unit rotates the sweetpotato through a full 360-degree scan. The data is transferred a spreadsheet file where the manipulation of the data yields surface data points in 3D space (x, y and z) with the axis of rotation acting as the datum line. The scan data for each sweetpotato is analyzed by a complex set of mathematical algorithms. From the manipulation of these data points, a range of characteristics may be derived that can be used to describe the shape and size of that individual sweetpotato. Further, the statistical analysis of the characteristics from a specific population may be used to characterize the population as a whole and thus optimize shape controlling factors.