A Review on Computer Vision - Scene Classification Techniques

Abstract

In today's era, need for automatic response of machines on certain task has been prevalent. Humans want their life easier and automatic in every possible way. However, those tasks require better understanding by the machine to perform human like tasks. Tasks like classification, detection and localization are on high demand and dominant research area. These tasks fall into a domain called computer vision where computers by analyzing and understanding performs human like tasks. This domain provides the automatic inference by machines to make human life easier. In this paper, we focus on one of the difficult computer vision tasks called scene classification. Scene Classification deals with techniques that make machine intelligent and automated by processing given input say image. As machines are made automatic and intelligent to perform various tasks, Artificial Intelligence and Image processing comes into the picture. We study and analyze various approaches and methods by which such task can be handled easily and accurately. Furthermore, we compare all the approaches and find out the best approach to opt for this task.