Abstract—Accurate and precise pedestrian detection play a major role in analyzing and understanding its behavior. Pedes-trian recognition is a challenging task due to body deformation, weather, and lighting conditions variations. Various techniques combine feature extraction with support vector machine. How-ever, deep Convolutional Neural Networks (i.e. CNNs) achieved promising results in various recognition tasks. Although, CNNs require large databases and expensive computations, it can outperform other algorithms more accurately. In this paper, we proposed a deep learning technique for automatic pedestri-an recognition based on image normalization and CNN archi-tecture. The proposed architecture learns pedestrian represen-tation adaptively to achieve efficient recognition with higher accuracy and lower pre-processing time. The experimental results show that the proposed technique out performs conven-tional methods superiorly.
Index Terms—Neural network, machine learning, pedestrian recognition, deep learning.
The authors are with the Al-Balqa'Applied University, Jordan (e-mail: drobaidam@yahoo.com, masmadi@bau.edu.jo).
Cite: Obaida M. Al-Hazaimeh and Ma'moun Al-Smadi, "Automated Pedestrian Recognition Based on Deep Convolutional Neural Networks," International Journal of Machine Learning and Computing vol. 9, no. 5, pp. 662-667, 2019.
Copyright © 2019 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).