Abstract
Coal classification is an essential process in the mining industry, which involves identifying the quality and type of coal extracted from the earth. Traditional methods of coal classification rely on manual inspection and analysis, which can be time-consuming and prone to errors. With the advent of machine learning techniques, it is now possible to automate this process and achieve higher accuracy and speed in coal classification. The first effort in learning about coal is observing coal features. This project developed a coal search system that allows users to do a search even when they do not know the coal name simply by observing coal characteristics. At present, coal classification uses machine vision to extract and analyze color, size, shape, and surface texture. Still, the new extraction margin method can be carried out roughly yet there is still a difference between the margin of extracted polygon, shape and the margin of the shape of original image. The project aims in finding the gangue in the coal. Total gangue percent in the coal data is then calculated and displayed which is based on pixels count of gangue colors. This assists in evaluating the coal quality. If future researchers were to expand to other features, coal gangue, etc., even those that are hard to quantify, can also be quantified. Artificial Neural Network is used for classifying the coal dataset. The project is designed using Python as frontend environment. The coding language used is the Python 3.7.