Advanced search
Start date
Betweenand


Tuned Support Vector Machine Classifier for Pedestrian Recognition in Urban Traffic

Full text
Author(s):
Roncancio, Henry ; Hernandes, Andre Carmona ; Archila, John Faber ; Becker, Marcelo
Total Authors: 4
Document type: Journal article
Source: INGENIERIA; v. 17, n. 2, p. 9-pg., 2012-01-01.
Abstract

The need for autonomy and intelligent decision-making in automobile flow is booming. For this purpose there are a number of interesting problems related to recognition of features in urban environments. One of the main relevant aspects in this subject is the recognition of pedestrians, a technology that is expected to save millions of lives avoiding or decreasing the rates of pedestrian run away. In this paper we propose the recognition of pedestrians in urban environments using a classifier based on a Support Vector Machine. We used up to 5000 images from the INRIA database to train the classifier and validate its accuracy through the cross-validation method. (AU)

FAPESP's process: 11/03986-9 - Object detection and classification in outdoor environments for autonomous passenger vehicle navigation based on data-fusion of an artificial vision system and laser sensor
Grantee:Henry Antonio Roncancio Velandia
Support Opportunities: Scholarships in Brazil - Master