Schematic for a Computer Vision Framework for Post-Harvest Quality Grading of Dry Red Chilies

Abstract

The automatic sorting of the agricultural and food products is the need of the current era as there is a higher emphasis on quality. Food and Agricultural Organization (FAO) statistics say that 2959 thousand tonnes of Chilli are raised in the world over an area of 1832 thousand hectares. Chilli's export accounts for 48% in terms of amount and 28% in terms of the value of India's complete spice export. India is the primary caterer of global demand for Dry Red Chillies. It exports powder, dried Chilli, Pickled Chillies and Chilli oleoresins. Countries such as The United States of America, The United Kingdom, Germany and Sweden use Chilli to produce large-scale oleoresins and extracts. Considering the commercial value of Chilli and results of the literature survey built a strong case to undertake the automation of Postharvest quality grading of Chilli. The study provides a framework for the application of Soft Computing-based Computer Vision System for Automatic Post Harvest Quality Grading of Dry Capsicum Species (Red Chilli).  Given the disadvantages of manual methods and optoelectronic methods, which process only a single dimension of data, this study has provided a novel approach by using Computer Vision for Chilli Quality Grading. Morphological Features, Moisture Content, Colour and Pungency are the four quality parameters based on which the Chillies can be sorted. The study presents Schematic for Intelligent Chilli Sorter that proposes to use Artificial Neural Network for the prediction of Capsanthin content determination (ASTA/IC Colour Units); use thermal images to determine the moisture content of Chillies; use the Girth of Chillies extracted from the images to determine the Pungency of the Chillies. This paper presents a Schematic of Computer Vision Framework for Post-Harvest Quality Grading of Dry Red Chilles that uses multistage decision-making.