Freshness Identification of Banana using Image Processing Techniques

Abstract

Agriculture is a very important culture which includes several practices for instance cultivating soil, producing crops and raising livestock. In Sri Lankan context, banana is widely consumed as it fits all the occasions and it has export value too. As a result, determining freshness of banana has major influence in defining its quality. The naked eye observation of experts is the main approach adopted for determining the freshness of banana in terms of days. We developed a method to identify the freshness using image processing techniques. For this experiment, images were captured using a professional camera. The fruit’s regions were segmented using K-Means clustering and the determination of freshness was done with Support vector machine by training with the selected features from the training set of images. The accuracy level of freshness determination was calculated separately for each category in terms of days from day one to ten. Association among the features as Contrast, Correlation, Energy, Homogeneity, Entropy, Mean, Standard deviation, Skew, and Kurtosis gave the optimum accuracy. This system with high accuracy motivates the other researchers to extend the system with added functionality, which will be a consumer friendly software solution.