A Novel Method of Object Identification and Tagging Using Speeded-Up Robust Feature

摘要

The one thing that is gradually increasing in the world is data. They are pieces of information's and knowledge and are measured, collected, reported and analyzed. Data can be visualized using graphs, images or any other analytic tools. The visual complexity of image data is large when compared to the text data. Image data could be transformed into useful information's if people could analyze them properly. Image mining is a special category under the field of data mining where new patterns and relationships are extracted from the preexisting database. There exist different image mining techniques that are used to identify images that contain potential objects. Embedded details in the image could be extracted using low level features such as color, shape and texture that are invariant to any changes associated with that particular image. This paper aims at introducing a method to identify potential objects from images based on high level feature extraction. Key points are extracted from the images using Speeded-Up Robust Feature (SURF) and objects are identified and tagged. The proposed method aiming at widening the possibilities of object identification and tagging.