Thursday 28th of March 2024
 

Improving 2D Boosted Classifiers Using Depth LDA Classifier for Robust Face Detection


Mahmood Rahat, Masoom Nazari, Akram Bafandehkar and Saeed Shiry Ghidary

Face detection plays an important role in Human Robot Interaction. Many of services provided by robots depend on face detection. This paper presents a novel face detection algorithm which uses depth data to improve the efficiency of a boosted classifier on 2D data for reduction of false positive alarms. The proposed method uses two levels of cascade classifiers. The classifiers of the first level deal with 2D data and classifiers of the second level use depth data captured by a stereo camera. The first level employs conventional cascade of boosted classifiers which eliminates many of nonface sub windows. The remaining sub windows are used as input to the second level. After calculating the corresponding depth model of the sub windows, a heuristic classifier along with a Linear Discriminant analysis (LDA) classifier is applied on the depth data to reject remaining non face sub windows. The experimental results of the proposed method using a Bumblebee-2 stereo vision system on a mobile platform for real time detection of human faces in natural cluttered environments reveal significantly reduction of false positive alarms of 2D face detector.

Keywords: Face Detection, Human Machine Interaction, Stereo Vision, False Positive Error Reduction.

Download Full-Text


ABOUT THE AUTHORS

Mahmood Rahat
Mahmoud Rahat was born in Tehran, Iran, in July 1985. He received the BS degree in Computer science from Tehran University, Tehran, Iran and the MS degree in computer engineering from Amirkabir University, Tehran, Iran, in 2008 and 2010. His research interests include human and machine vision, neural networks, and pattern recognition.

Masoom Nazari
Masoom Nazari was born in 1985. He received the B.Sc. degree in electronics engineering from Sabzevar University, sabzevar, Iran, in 2007, the M.Sc. degree in electronics engineering from the Electrical Engineering Department of Shahid Rajaee University, in 2010. His research interests include human and machine vision, neural networks, and pattern recognition.

Akram Bafandehkar
Akram Bafandehkar was born in 1986. She received the B.Sc. degree in computer engineering from Azad University, Dezful, Iran, in 2008, the M.Sc. degree in Artificial Intelligence from the University of Shiraz, in 2011. She is a member of young research club since 2008 till now. Her research interests are machine vision, digital image processing, pattern recognition and machine learning.

Saeed Shiry Ghidary
Saeed Shiry Ghidary was born in 1967. He is Assist Prof. of Computer Engineering & Information Technology Department. He received the B.Sc. degree in electrical engineering from Electronic engineering, Amirkabir University, Tehran, Iran, in 1990, the M.Sc. degree in Computer architecture, Amirkabir University, Tehran, Iran, in 1994 and the Ph.D degree in Robotics, Kobe University, Kobe, Japan, in 2002. His research interests include fuzzy modeling, machine learning robotics and fuzzy modeling.


IJCSI Published Papers Indexed By:

 

 

 

 
+++
About IJCSI

IJCSI is a refereed open access international journal for scientific papers dealing in all areas of computer science research...

Learn more »
Join Us
FAQs

Read the most frequently asked questions about IJCSI.

Frequently Asked Questions (FAQs) »
Get in touch

Phone: +230 911 5482
Email: info@ijcsi.org

More contact details »