Pyramidal Part-Based Model for Partial Occlusion Handling in Pedestrian Classification

M Thu,N Suvonvorn

doi:10.1155/2020/6153580

Abstract

Pedestrian detection and classification are of increased interest in the intelligent transportation system (ITS), and among the challenging issues, we can find limitations of tiny and occluded appearances, large variation of human pose, cluttered background, and complex environment. In fact, a partial occlusion handling is important in the case of detecting pedestrians, in order to avoid accidents between pedestrians and vehicles, since it is difficult to detect when pedestrians are suddenly crossing the road. To solve the partial occlusion problem, a pyramidal part-based model (PPM) is proposed to obtain a more accurate prediction based on the majority vote of the confidence score of the visible parts by cascading the pyramidal structure. The experimental results on the proposed scheme achieved 96.25% accuracy on the INRIA dataset and 81% accuracy on the PSU (Prince of Songkla University) dataset. Our proposed model can be applied in the real-world environment to classify the occluded part of pedestrians with the various information of part representation at each pyramid layer.

Highlights

Each year, there are about 1.3 million road-related fatalities worldwide, and crash rates are escalating in most urbanized countries [1,2,3]
Pyramidal Part-Based Model e challenging part of this research is to handle the partial occlusion when some parts of the pedestrian are invisible, and the inaccurate scores of the part classifier will affect the performance prediction. erefore, a pyramidal part-based model (PPM) is proposed to obtain a more accurate prediction based on the majority vote of the confidence scores of the visible parts by cascading the pyramid structure from top to down. ere are two classification stages in the proposed model: a part classifier (PC) and a pyramidal ensemble classifier (PEC)
We choose histogram of oriented gradients (HOG) + support vector machine (SVM) as the baseline classifier because of it being widely used in pedestrian detection and classification

Summary

Introduction

There are about 1.3 million road-related fatalities worldwide, and crash rates are escalating in most urbanized countries [1,2,3]. Since the turn of the century, many researchers have focused on employing the wide range of applications such as intelligent transport systems, driver assistance system, intelligent video surveillance system, and automotive safety system with an aim to mitigate and alleviate road crashes. Computer vision could increasingly resemble human vision and emulate sensed images and videos through applications of a wide range of machine learning technologies that can rectify erroneous human vision and create a safer environment in the daily drive [4]. To protect them and to reduce other potential risks, pedestrian detection and classification systems are widely employed. Occlusion handling under complex backgrounds in the real-world environment may involve further difficulties [8, 14, 19,20,21, 23]

Methods

Results

Conclusion