Identifying and Analysis of Scene Mining Methods Beased on Scenes Extracted Features
This work addresses the difficulty in comparing scene mining methods for applications like medicine and computer vision, but it is incremental as it focuses on classification and evaluation rather than new techniques.
The paper tackles the challenge of comparing diverse scene mining methods by introducing a framework based on extracted features to classify and evaluate them, analyzing and assessing these methods using the proposed framework.
Scene mining is a subset of image mining in which scenes are classified to a distinct set of classes based on analysis of their content. In other word in scene mining, a label is given to visual content of scene, for example, mountain, beach. Scene mining is used in applications such as medicine, movie, information retrieval, computer vision, recognition of traffic scene. Reviewing of represented methods shows there are various methods in scene mining. Scene mining applications extension and existence of various scenes, make comparison of methods hard. Scene mining can be followed by identifying scene mining components and representing a framework to analyzing and evaluating methods. In this paper, at first, components of scene mining are introduced, then a framework based on extracted features of scene is represented to classify scene mining methods. Finally, these methods are analyzed and evaluated via a proposal framework.