EXEMPLIFYING DIGITAL IMAGE PROCESSING USING EXPERIMENTS

The increasing global demand for energy necessitates actions ranging from the discovery of alternative energy sources to the advancement of technologies for more energy-efficient machinery and equipment. The area of digital image processing relates to the use of a digital machine to view digital photographs. It's important to remember that a digital picture is made up of a finite number of components, each of which has its own position and meaning. Picture elements, image elements, pels, and pixels are all terms used to describe these things. Pixel is the most often used concept to describe the components of a digital picture. We take a more formal look at these concepts. Since vision is the most sophisticated of our senses, it's no surprise that photographs are the most significant factor in human perception. Non-grain based electrical steels are commonly used in the construction of rotors and stators that form the centre of electric motors, and their microstructures are closely connected to their electromagnetic output, unlike humans, who are restricted to the visual band of the electromagnetic (EM) continuum. This paper uses photo micrographic analysis to present a modern, quick, and powerful method for classifying non-grain focused electrical steel microstructural states and their electromagnetic efficiency. The experiment was conducted on non-grain centred electrical steel samples containing 1.28 percent silicon, which were cold-rolled with reductions between 50 and 70 percent, annealed in box at 730°C for 12 hours, and then heat treated for grain development at 620°C, 730°C, 840°C, and 900°C for 1, 10, 100, and 1000 minutes at each temperature. A database of 192 photographs was generated using 32 samples in all. Our method fused extractor features (GLCM, LBP, and moments) with classifiers (Bayes, K-NN, K-means, MLP, and SVM), as well as two data partitioning and the keep out and leave one out techniques. The accuracy score for KNN with one neighbour using the GLCM extractor was 97.44 percent, with values greater than 96.0 percent for the other validation methods. The test only took 15.4 milliseconds to complete. The findings of this proposed method produce a new method for evaluating the electromagnetic efficiency of non-grain centred electrical steel.