Conference paper

WIGLASZ Michal and SEKANINA Lukáš. Evolutionary Approximation of Gradient Orientation Module in HOG-based Human Detection System. In: 2017 IEEE Global Conference on Signal and Information Processing GlobalSIP 2017. Montreal: IEEE Signal Processing Society, 2017, pp. 1300-1304. ISBN 978-1-5090-5989-8.
Publication language:english
Original title:Evolutionary Approximation of Gradient Orientation Module in HOG-based Human Detection System
Title (cs):Evoluční aproximace výpočtu orientace gradientu v systému detekce osob založeném na HOG
Pages:1300-1304
Proceedings:2017 IEEE Global Conference on Signal and Information Processing GlobalSIP 2017
Conference:5th IEEE Global Conference on Signal and Information Processing
Place:Montreal, CA
Year:2017
ISBN:978-1-5090-5989-8
Publisher:IEEE Signal Processing Society
Files: 
+Type Name Title Size Last modified
icon0001300.pdf229 KB2017-11-20 19:34:43
^ Select all
With selected:
Keywords
Functional approximation, Cartesian genetic programming, Histogram of oriented gradients
Annotation

The histogram of oriented gradients (HOG) feature extraction is a computer vision method widely used in embedded systems for detection of objects such as pedestrians. We used Cartesian genetic programming (CGP) to exploit the error resilience in the HOG algorithm. We evolved new approximate implementations of the arctan function, which is typically employed to compute the gradient orientations. When the best evolved approximations are integrated into the SW implementation of the HOG algorithm, not only the execution time, but also the classification accuracy was improved in comparison with the accurate implementation and the state-of-the-art approximate implementations.
BibTeX:
@INPROCEEDINGS{
   author = {Michal Wiglasz and Luk{\'{a}}{\v{s}} Sekanina},
   title = {Evolutionary Approximation of Gradient Orientation
	Module in HOG-based Human Detection System},
   pages = {1300--1304},
   booktitle = {2017 IEEE Global Conference on Signal and Information
	Processing GlobalSIP 2017},
   year = {2017},
   location = {Montreal, CA},
   publisher = {IEEE Signal Processing Society},
   ISBN = {978-1-5090-5989-8},
   language = {english},
   url = {http://www.fit.vutbr.cz/research/view_pub.php?id=11441}
}

Your IPv4 address: 54.243.17.113
Switch to https

DNSSEC [dnssec]