Identification of Resistor Types Using Extreme Learning Machine Algorithms and Morphological Operation

Abstract

Electronic components are the basic elements to form a series of electronic devices that are usually used in everyday life. For someone who studies the field of electricity, knowledge of electrical components is an important thing. One of the components whose use is most often found in electronic circuits is a resistor. However, some people do not know about these types of resistors. Especially for someone or a student who will learn about electronic components. This study aims to develop an image processing system that can identify transistor type images using the Extreme Learning Machine (ELM) algorithm. This algorithm performs integrated learning through a special form of feedforward perceptron which has one hidden layer. In order for the ELM algorithm to work properly, information about the features contained in the object to be identified is needed. So, in this study the ELM algorithm is combined with morphological characteristics through parameters such as area, perimeter, eccentricity, major axis length, and minor axis length. Based on these parameters, features will be obtained which will be input in the identification process. At the evaluation stage, the precision value was 87%, recall was 84.47% and accuracy was 85.5%.