Morphology based Quality Estimation of Cocoa Beans using Digital Imaging

Abstract

The objective of this research is to determine the quality of cocoa beans through morphology of their digital images. Samples of cocoa beans were scattered on a bright white paper under a controlled lighting condition. A compact digital camera was used to capture the images. The images were then processed to extract their morphological parameters. Some of the parameters for extracted features are Area, Perimeter, Major Axis Length, Minor Axis Length, Aspect Ratio, Circularity, Roundness, and Ferret Diameter etc. Then feature optimization is implemented to both reduce the computational cost of modeling and, to improve the performance of the model. The cocoa beans are classified into 4 groups, i.e. Large beans, Medium Beans, Small Beans, and Fragmented or Broken Beans. The model of classification used in this paper is the Hierarchy-based Decision Tree Model, a proposed improvement model for normal Decision Tree in which single class will be determined at single step. Five classification approaches were applied ie LDA, QDA, NaiveBayes, Decision Tree and hierarchy-based Decision Tree and the last one gives the maximum accuracy. The result of our proposed model showed that the proposed classification model with morphological feature parameters can accurately classify 93% of beans into four classes.