Attribute Sparse Coding (ASC’s) for Content-Based Image Retrieval (CBIR)
Journal Title: International Journal of Research in Computer and Communication Technology - Year 2014, Vol 3, Issue 9
Abstract
Photographs with individuals are the foremost attention of users. Consequently with the exponentially increasing pictures, large-scale content-based face image retrieval is a facilitating knowledge for various developing applications. The main objective is to apply automatically spotted human characteristics that comprise semantic cues of the face photos to increase gratified based face retrieval by creating semantic code words for effectual large-scale face retrieval. By leveraging human characteristics in a scalable and methodical framework, we suggest two orthogonal methods named attribute-enhanced sparse coding and attribute embedded inverted indexing to develop the face retrieval in the offline and online stages. The results illustrate that the proposed methods can attain up to 43.5% qualified enhancement in MAP associated to the existing methods.
Authors and Affiliations
Satish Kumar Koduri, V. Chandra Kumar
Keywords
Related Articles
- EP ID EP27999
- DOI -
- Views 264
- Downloads 0
Ensuring Data Storage Security in a Cloud Computing Using ‘MONA’
Due to the frequent change of the membership, sharing data in a multi-owner manner is a major problem in cloud computing. Identity privacy and Privacy preserving from an entrusted cloud is still a challenging issue....
An improved contourlet transform technique for image watermarking
In this paper, contourlet transform technique is explored to hide the information confidentially from the unauthorized users. The image is decomposed into subbands by using contourlet transform. These subbands are us...
A Novel Technique For Efficient Data Access In MDTN
It is pleasing to offer differentiated access services such thatdata access policies are distinct over user attributes or roles,which are direct by the key authorities. We propose a securedata recovery scheme using C...
A Mustang Security User Authentication Scheme Based On Graphical Password
In all computer security contexts, User authentication is an essential thing. The most popular and simple type of user authentication method is to use the alphanumeric passwords. The reasons behind why we are using a...
Secured Bluetooth Authentication Using Artificial Neural Networks
Authentication in wireless networking is a mechanism to proof identities and avoid masking. The use of wireless devices and technology has made rapid growth in the market. In all such devices, user authentication is...