Averaging Forest for Online Vision

In this study we consider vision as a binary classification problem, where an ensemble of decision-tree-based classifiers is trained on-line, new images are continuously added and the recognition decision is made without delay. Ensemble of decision trees is combined into a forest classifier using av...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Journal of advanced computational intelligence and intelligent informatics Ročník 13; číslo 4; s. 400 - 406
Hlavní autor: Osman, Hassab Elgawi
Médium: Journal Article
Jazyk:angličtina
Vydáno: 20.07.2009
ISSN:1343-0130, 1883-8014
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:In this study we consider vision as a binary classification problem, where an ensemble of decision-tree-based classifiers is trained on-line, new images are continuously added and the recognition decision is made without delay. Ensemble of decision trees is combined into a forest classifier using averaging , generate an on-line Random Forest (RF) classifier. First we employ object descriptor model based on a bag of covariance matrices, to represent an object features, then run our on-line RF learner to select object descriptors and to learn object classifiers. Validation of our proposal with empirical studies in the GRAZ02 dataset domain demonstrates its superior performance over histogram-based counterparts, yielding object recognition performance comparable to state-of-the-art standard RF, AdaBoost, and SVM classifiers, even when only 10% of the training examples are used.
ISSN:1343-0130
1883-8014
DOI:10.20965/jaciii.2009.p0400