A Deep Learning-based System for Classification and Quality Evaluation of Seeds

Farmers and agro-industries must automate seed segregation because it is a time-consuming and labor-intensive task when performed manually. Deep learning (DL) and Machine Learning (ML) based algorithms have exhibited promising results in object recognition, classification, and pattern recognition. Despite this, no research has been conducted on the classification of seed types for crops grown with several crop kinds and varying quality criteria. The objective of this research is to develop a system based on deep learning dubbed the Mixed Cropping Seed Classifier and Quality Tester (McscQT). This system will be capable of analyzing the quality of seeds based on their shape, color, and texture. The system is trained with labelled images of healthy and damaged pearl millet and maize seeds, and it has a recall and accuracy of 98.9%. The MCSCQT is an essential piece of equipment in the food sector since it can distinguish between healthy and damaged maize seeds.