Classification of mangoes by object features and contour modeling

Abstract

Object classification plays a vital role in computer vision applications such as decision making in industrial applications, satellite imagery analysis, medical image interpretation etc. Inter and intra classification of fruits in agro industries is one of the repeated and time consuming process. Intra-classification of fruits like mango, banana is being achieved by human experts who are proficient in the trade. Due to the significant growth in agro industries and lack of experts, an automation of fruits intra-classification is required. In order to address this issue, an image processing based solution is proposed for intra-classification of fruits especially mangoes based on shape and region features which are translation and rotation invariant. These features along with Object Contour Model drive Bayes classifier to classify mango varieties.