Image Color Composition Detection Using Classification Supervised Learning to Support Object Detection in Machine Learning
Noach Nathanael Tjahjadi, Benny Hardjono, Sutrisno Sutrisno · 2022
Object detection is one of the most challenging problems in Computer Vision. It is difficult because there are many variations between images which have the same object category. Other factors include changes in perspective, scale, partial occlusion. The human visual system is more sensitive to color information than “gray levels” hence it is used for feature extraction in this study. The color interval value model has been chosen as a new Machine Learning Model. It is hoped that this model will be able to predict the colors in images. The study begins with preparation and preprocessing of the dataset, followed by Classifier Training to get a reference dataset, and finally the interval model. Evaluation has been conducted to determine the predictive performance of the created model. Initial results show that the macro average of precision is 60% and the recall is 23%. These results have shown that the model still cannot achieve color prediction of images with higher accuracy. (Abstract)