A COMPREHENSIVE REVIEW OF CNN ARCHITECTURES FOR IMAGE RECOGNITION: ADVANCES AND OPEN CHALLENGES

Abstract

Convolutional Neural Networks (CNNs), considered revolutionary, have transformed the field of computer vision, aiding exceptional advancements in image recognition tasks. With their ability to automatically learn spatial hierarchies of features, CNNs have become the backbone of most innovative image recognition systems. From their inception with LeNet in the late 1990s to the latest revolutions like Vision Transformers (ViTs), CNN architectures have undergone significant evolution. This paper provides a comprehensive review of the key developments in CNN architectures, focusing on their impact on image recognition performance, the challenges they face, and the potential future directions.