Defect Mango Identification and Classification

This is a project to identify and classify mango defects using machine learning. This project is a part of the final year project of the Department of Computer Science, Eastern University of Sri Lanka.

The assessment of mango quality plays a crucial role in maintaining consumer trust and ensuring market competitiveness. However, traditional methods rely on subjective human judgment, leading to inconsistencies and inaccuracies in quality detection of mangoes.

The primary goal of this project is to develop an innovative methodology that utilizes image processing analysis to identify and classify the quality of mangoes automatically using mango characteristics such as size, shape, color, and surface conditions.

Ultimately, goal is to provide the agricultural industry with that enhances supply chain management, reduces post-harvest losses, and ensures consistent quality standards in mango distribution.

Description

Mangoes, renowned for their delectable taste and nutritional value, hold substantial economic importance in various markets. The ability to accurately evaluate mango quality through automated methods can significantly enhance supply chain management and reduce post-harvest losses.

The proposed methodology involves capturing high-resolution images of mangoes and subjecting them to image processing algorithms.

These algorithms will extract essential visual features such as size, shape, color, and surface characteristics. Subsequently, machine learning techniques, encompassing both traditional methods will be employed to analyze these features and categorize mangoes into distinct quality level.