Synthetic Image Classification Using ML
Nupoor M. Yawale, Neeraj Sahu, Nikkoo N. Khalsa · 2021 IEEE International Conference on Mobile Networks and Wireless Communications (ICMNWC) · 2021
Machine learning approaches typically need large volumes of training data and often demand costly manual etiquette to achieve their maximum ability. The value of images produced solely by visual means is on the increase. The color histogram is commonly used for the image classification challenge as a significant color characteristic that indicates the material. Since synthetic representations are an essential means through which visual knowledge is collected and presented. It is increasingly critical that these images are accurately classified in sub-categories – such as symbols, diagrams, figures and tables, logos, etc. The proposed research work aims at classifying synthetic images into sub-categories. The essential features of images shall be extensively analyzed and processed. When the web begins, photographs are used to convey content and beautify, shape, and align. In particular, the work is directed to image recognition techniques to identify pictures that are of great interest. This work tries to classify synthetic images from different images using a machine learning approach.