EYES DETECTION IMPROVEMENT BY TRADITIONAL AND MODIFIED PULSE COUPLED NEURAL NETWORK

Authors

  • Rafidison Maminiaina Alphonse Telecommunication Automatic Signal Image Research Laboratory/Doctoral School in Science and Technology of Engineering and Innovation/ University of Antananarivo
  • Ramafiarisona Hajasoa Malalatiana Telecommunication Automatic Signal Image Research Laboratory/Doctoral School in Science and Technology of Engineering and Innovation/ University of Antananarivo

DOI:

https://doi.org/10.17605/OSF.IO/V5WN7

Keywords:

Traditional Pulse, Neural Network, Face detection

Abstract

This paper presents a fast method and robust for eyes detection, using Traditional and Modified Pulse Coupled Neural Networks (MPCNN). The functionality of MPCNN is not the same as Traditional Pulse Coupled Neural Networks (TPCNN) because we are taking care about human visual perception not only the neighbor pixel characteristic. Due of this feature, the algorithm response time is around two milliseconds. The approach has two components including: face area detection based on segmentation and eyes detection using edge. Segmentation operation is ensured by MPCNN and edge detection by TPCNN.
The biggest region which is constituted by pixel value one will be the human face area. The segmented face zone which will be the input of TPCNN for edge detection undergoes a vertical gradient operation. The two gravity’s center of close edge near the horizontal line which corresponds to the peak value of horizontal projection of vertical gradient image will be the eyes.

Downloads

Published

2021-05-16

Issue

Vol. 2 No. 05 (2021): wos

Section

Articles