Taxonomy and Complete Overview from AI to Deep Convolutional Neural Networks in Computer Vision

Abstract

Homo sapiens are the only species that top among all the other species on the planet earth and also known for their three important qualities such as learning (patana), thinking (chinthana) and analysis (mamthana). Perhaps this is the only reason, many researchers try to mimic the human intelligence in machine which is still continued till date in the domain known as Artificial Intelligence. The paper focuses on Non-symbolic AI technique which is one of the two main important approaches such as symbolic and non-symbolic AI technique used in solving a problem or finding a solution. The Non-symbolic technique which is either principle-based or similar to that of biologically inspired way of solving problem through distributed and connected Neurons. In this way the paper narrow down from AI to Neural networks to Deep convolutional neural networks and its variations. The major contribution of this paper as it take journey from AI to Deep CNN, the understanding of the concept especially in the application domain such as computer vision is been captured in the form of taxonomy.