Detecting Deterioration Level Approach of Cassava Root Using Thermal Imaging with Multivariate Analysis

Abstract

Freshness is an important parameter that is indexed in the quality assessment of commercial cassava tubers. Cassava tubers that are not fresh have reduced starch content. Therefore, in this study, we aimed to develop a new approach to detecting cassava root deterioration level using thermal imaging with multivariate analysis. An underlying assumption was that the aging cassava root may have fermentation inside the cassava root that causes a difference of the inner temperature of the tuber. This creates the opportunity for the deterioration level to be measured using thermal imaging. The features (intensity and temperature) that were extracted from the region of interest (ROI) in the form of tuber thermal images were analyzed with multivariate analysis. Linear discriminant analysis (LDA), k-nearest neighbor (kNN), support vector machine (SVM), decision tree and ensemble classifiers were applied for the establishment of the optimal classification modelling algorithms. The highest accuracy model was developed from thermal images obtained in a dark room and at a control temperature of 25°C in the measurement chamber. The LDA, SVM and Ensemble classifiers gave the best overall performance for the discrimination of cassava root deterioration level with an accuracy of 86.7%. Interestingly, under uncontrolled environmental conditions, the combination of thermal imaging plus multivariate analysis gave results that were of lower accuracy but nevertheless acceptable. Then, our work revealed that thermal imaging coupled with multivariate methods was a promising method for the nondestructive evaluation of cassava root deterioration levels.

Keywords: cassava root, thermal imaging, multivariate analysis