Parallel Equilibrium Difference Local Binary Pattern Using GPU

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In the realm of computer vision and image processing, local pattern descriptors play a crucial role in extracting discriminative features for various applications. The Equilibrium Difference Local Binary Pattern (ED-LBP) is a recently proposed method that combines color information with local binary patterns, enhancing its robustness and discriminative power. However, the computational demands of ED-LBP is significantly high due to additional mathematical constraints, especially when processing high-resolution images. In this paper, the implementation of ED-LBP is parallelized to improve computational efficiency and enable real-time processing of large-scale image datasets. Our approach leverages the power of Graphics Processing Units (GPUs), resulting in a significant reduction in feature extraction time without compromising the accuracy of the ED-LBP algorithm.

Original languageEnglish
Title of host publication5th Congress on Intelligent Systems, CIS 2024
EditorsSandeep Kumar, E.A. Mary Anita, Joong Hoon Kim, Atulya Nagar
PublisherSpringer Science and Business Media Deutschland GmbH
Pages83-92
Number of pages10
ISBN (Print)9789819626991
DOIs
Publication statusPublished - 2025
Event5th Congress on Intelligent Systems, CIS 2024 - Bengaluru, India
Duration: 04-09-202405-09-2024

Publication series

NameLecture Notes in Networks and Systems
Volume1277 LNNS
ISSN (Print)2367-3370
ISSN (Electronic)2367-3389

Conference

Conference5th Congress on Intelligent Systems, CIS 2024
Country/TerritoryIndia
CityBengaluru
Period04-09-2405-09-24

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Signal Processing
  • Computer Networks and Communications

Fingerprint

Dive into the research topics of 'Parallel Equilibrium Difference Local Binary Pattern Using GPU'. Together they form a unique fingerprint.

Cite this