• Priya Singla
• Rajesh Mehra

The scene reocognition provides the visual data from the level of the objects and the relation between the objects. The aim of the scene recognition is to reduce the semantic gap between the computer system and social life. In the form of the example,one can determine the contexts of an input image and classify them in to scenes such as forest, buildings seashores. Some of the issues of the scene recognition is forest, coast,tall buildings highways and street. The applications of the scene recognition are object detection , object recognition,text and video detection, matching human perception,robotics and graphics.This paper surveys on different techniques of scene recognition from single image to multi-million images. Scene recognition performs a task for context for the object recognition. The availability of the large datasets like Image Net leads to multi-million-item dataset initiatives. Scene Recognition is one of the successful methods of the recognition of the images.The variations in color, size of the image are often affected by complex backgrounds, different lighting conditions, shadows and reflections of the image. So solution to this problem is to enhance the accuracy of scene image recognition. This leads to a variety of applications such as understanding the scene, automatic localisation of the image.

ImageNet, multimillion images, Scene recognition

"Recognize Various Scenes using Classification methods in Million Images Dataset: A Review", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.240-246, June-2019, Available :http://www.jetir.org/papers/JETIR1908B34.pdf

"Recognize Various Scenes using Classification methods in Million Images Dataset: A Review", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp240-246, June-2019, Available at : http://www.jetir.org/papers/JETIR1908B34.pdf