Vision-Based Human Detection Techniques: A Descriptive Review

Shahriar Shakir Sumit,Seyedali Mirjalili,Dayang Rohaya Awang Rambli

doi:10.1109/access.2021.3063028

Abstract

Cameras are being used everywhere for the safety and security of citizens in different countries. Using a machine to detect humans in a photo or a video frame is a very complicated and challenging task. Various techniques have been developed for this purpose, which mainly rely on Artificial Intelligence. This article aims to provide a comprehensive review and analysis of the literatures from a descriptive perspective, which is its main differentiator from the existing survey papers in this area. Firstly, the vision-based human detection techniques and classifiers are elucidated in conjunction with the variants of feature extraction techniques. Secondly, various pros and cons of such techniques are discussed. Then, an investigation has been conducted and reported based on the state-of-the-art human detection descriptors (e.g. Log-Average Miss Rate and accuracy). Although techniques such as Viola-Jones and Speeded-Up Robust Features can detect objects in real-time and overcome Scale-Invariant Feature Transform (SIFT) limitations, they are still sensitive to illuminated conditions. Other techniques such as SIFT, Bag of Words, Orthogonal Moments, and Histogram of oriented Gradients provide other interesting benefits which include insensitivity to occlusion and clutters, simplicity, low-order element construction and invariance to illuminated conditions; nevertheless, they are computationally expensive and sensitive to image rotation. A meticulous review along similar lines revealed that the Deformable Part-based Model performs relatively better due to its ability to deal with particular pose variations and multiple views, occlusion handling (partial) and is application-free while its counterparts focus on only a single aspect. This article highlights and provides a brief description of each available data-sets for human detection research. Various use-cases of human detection systems are also elaborated. Finally, various conclusions are derived based on the conducted review followed by recommendations for future directions and possibilities to further improve the speed and accuracy of human detection systems.

Highlights

A novel coronavirus (COVID-19) pandemic [1], [2] affecting the respiratory portion of the human system is currently ongoing [3] causing a high degree of mortality and morbidity globally [4]
The contents are as follow: explains the human detection methods and their variants with their successes and challenges, illustrates the classifiers with their pros and cons, human detection applications are described with advantages and limitations, available human detection datasets with brief description are discussed, stateof-the-art human detection methods’ results are reported, some suggestions to improve the existing descriptors are provided and open issues problems and future direction are given, and section VIII concludes the article
Orthogonal moments have a key role to play in image processing and other related applications

Summary

INTRODUCTION

A novel coronavirus (COVID-19) pandemic [1], [2] affecting the respiratory portion of the human system is currently ongoing [3] causing a high degree of mortality and morbidity globally [4]. We have investigated the existing vision-based human detection surveys, in which the authors reviewed human detection mostly for specific applications. The key objective of this article is to offer a comprehensive review on human detection utilizing various machine learning techniques. The contents are as follow: (section II) explains the human detection methods and their variants with their successes and challenges, (section III) illustrates the classifiers with their pros and cons, human detection applications are described with advantages and limitations (in section IV), available human detection datasets with brief description are discussed (in section V), stateof-the-art human detection methods’ results are reported (in section VI), (in section VII) some suggestions to improve the existing descriptors are provided and open issues problems and future direction are given, and section VIII concludes the article

FEATURE EXTRACTION TECHNIQUES

ORTHOGONAL MOMENTS

APPLICATION

RESULT

Findings

VIII. CONCLUSION