Library Open Repository

Multiple Classifier Object Detection with Confidence Measures

Downloads

Downloads per month over past year

Horton, M and Cameron-Jones, RM and Williams, R (2007) Multiple Classifier Object Detection with Confidence Measures. 20th Australian Joint Conference on Artificial Intelligence (AI 2007), 1. pp. 559-568. ISSN 1611-3349

This is the latest version of this item.

[img] PDF
Horton-Multiple...pdf | Request a copy
Full text restricted
Available under University of Tasmania Standard License.

Abstract

This paper describes an extension to the Haar Classifier Cascade technique for object detection. Existing Haar Classifier Cascades are binary; the extension adds confidence measurement. This confidence measure was implemented and found to improve accuracy on two object detection problems: face detection and fish detection. For fish detection, the problem of selecting positive training-sample angle-ranges was also considered; results showed that large random variations that result in cascades covering overlapping ranges increases their accuracy.

Item Type: Article
Keywords: Object detection,Fish detection,Haar Classifier Cascades
Journal or Publication Title: 20th Australian Joint Conference on Artificial Intelligence (AI 2007)
Page Range: pp. 559-568
ISSN: 1611-3349
Identification Number - DOI: 10.1007/978-3-540-76928-6
Additional Information:

After this paper was submitted for publication, errors were found in the positive sample generation and in the OpenCV training algorithms. These errors made the results in section 4.2 invalid. This version of the paper contains the correct results, which differ from those in the conference proceedings.
The original publication is available at www.springerlink.com

The original publication is available at www.springerlink.com

Date Deposited: 14 Dec 2007 03:47
Last Modified: 18 Nov 2014 03:26
Item Statistics: View statistics for this item

Available Versions of this Item

Actions (login required)

Item Control Page Item Control Page