SUPER-PIXEL BASED SALIENCY IN 3D-IMAGE OBJECT DETECTION USING CONTENT BASED IMAGE RETRIEVAL

Visual attention is an important factor in the Human Visual System (HVS) based method for the process of Information processing in visual media. Lots of visual information is provided by the HVS system, from which it would particularly process the wanted object or simply salient object by adding fil...

Full description

Saved in:
Bibliographic Details
Published in:Journal of Theoretical and Applied Information Technology 2017-02, Vol.95 (3), p.468-468
Main Authors: Rakesh, Y, Krishna, K Sri Rama
Format: Article
Language:English
Subjects:
Filtering
Image detection
Image management
Information technology
Object recognition
Retrieval
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Visual attention is an important factor in the Human Visual System (HVS) based method for the process of Information processing in visual media. Lots of visual information is provided by the HVS system, from which it would particularly process the wanted object or simply salient object by adding filtering techniques and also it lowers the level of complexity in the process of scene analysis. Content-based image retrieval is one of the major processes in visual processing technique. CBIR is used to identify the most similar images that are visually seen in response to the given set of query image from the huge contents of image database. Many techniques have been introduced so far in saliency detection of 3D images that are used by various multimedia and CBIR processing systems. In this paper, an efficient image retrieval algorithm is designed based on the Super-pixel based saliency detection. This aims at providing a successful method of saliency detection in 3D objects, which are the major process in CBIR applications. The Experimental results and verification of object tracking from this saliency detection method is visibly efficient than the all other methods as it provides super pixel generation in the objects and its contents.
ISSN:1817-3195