A Survey on Efficient Object Localization and Image Classification for Thumbnail Browsing
Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2015, Vol 3, Issue 3
Abstract
Nowadays, with an increasing demand of advanced intelligent systems, computer vision applications dealing with foreground objects are becoming more challenging. The common assumption taken into consideration in most of the cases is that an image contains more than one object which produces undesired results when noticeable objects do not appear in the image. Our visual attention model originates from a well-known property of the human visual system that the human visual perception is highly adaptive and sensitive to structural information in images rather than nonstructural information. In this paper, we not only address the problem of ascertaining the existence of objects in an image but also discuss various methods for object detection. The input image is divided into non-overlapping patches, then the patches are categorized into different classes, like natural, man-made, and object to estimate the existence of the object. This paper provides a systematic review of algorithms and performance measures and assesses their effectiveness via metrics.
Authors and Affiliations
P. S. Korhale, P. S. Deshpande
Keywords
Related Articles
- EP ID EP19702
- DOI -
- Views 295
- Downloads 5
slugLink Lifetime Prediction in Position-Based VANET Routing Protocol
Vehicular ad hoc network enables the vehicles to wirelessly network with other vehicles nearby without the aid of a permanent infrastructure. One of the main design constraints of vehicular ad hoc netwo...
Application of ANN for the Hydrological Modeling
Global climate change has been a burning problem for environmentalists since the middle of the 20th century. The application of artificial neural network (ANN) methodology for modeling climate change impact on river Jhe...
Secret Image Sharing Using Quick-Response Code Generation Technique
Patent safety and verification have become increasingly more important in regular life. The digital watermark is one of the methods invented to handle this problem. In this paper, a digitally invisible watermark is inse...
A Novel Method of Power Flow Control for the Compensation of Harmonics and Unbalanced Currents in Electric Railway Systems
This paper presents a novel method of power flow compensation of broad filtering and unbalance return system for electric traction systems using fuzzy logic controller. For a balanced three-phase three- wire system, the...
Prediction of Data Sets by using Fuzzy Mining Association Rule
This paper is proposed to reinforce fuzzy association rules, Association rule mining is one of the fundamental tasks of data mining. The conventional association rule mining algorithms, using crisp set, are meant for ha...