A Review of Vehicle Accident Detection and Notification Systems Based on Machine Learning Techniques - Issue 2

Aug 4, 2026

Upload by Ezra

Vehicle accidents on roads are increasing in both human loss and the repair costs that follow, creating an urgent need to shorten the time between an incident and the arrival of medical help. This review collects prior research on detecting road traffic events using machine learning, especially deep learning, and compares the methods, experimental outcomes, and research gaps. A systematic search strategy was applied across major databases, resulting in a subset of studies aligned with the review focus.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages22

Reading Time55 min

AccessFree to download

Abstract

Introduction

Review Methodology and Search Strategy

Methods and Experimental Results

Research Gaps and Findings

Academic Science Journal A Review of Vehicle Accident Detection and Notification Systems Based on Machine Learning Techniques Duaa Hadi Nassar* and Jamal Mustafa Al-Tuwaijari Department of Computer science, College of Science, University of Diyala * [scicompms2209@uodiyala.edu.iq](scicompms2209@uodiyala.edu.iq) Received: 13 November 2022 Accepted: 17 March 2023 DOI: [https://dx.doi.org/10.24237/ASJ.02.02.717B](https://dx.doi.org/10.24237/ASJ.02.02.717B) Abstract The rabid growth of the vehicles accidents on our roads are increasing as well as the number of human lives lost and the costs of repairing damages due to these accidents in order to reduce the dangers associated with accidents. A lot of researchers and specialists turn to electronic systems based on machine learning, deep learning techniques algorithms or any other artificial intelligence methods to detect and generate an emergency signal can reduce the time gap between the accident happening and the arrival of medical help. The main purpose of this review article was to identify and collect all the studies that have been done previously to detect roads traffics based on using machine learning techniques particularly deep learning. This paper will present all the methods that have been used and the experimental results of these studies and the research gabs they contain. We are following a systematic search plan passed on selected the papers that have the most similar keyword to make sure that all paper are exactly matching this paper subject so we apply this plan by using Springer, Elsevier, Electrical and Electronics Engineers (IEEE) xplore ,Google scholar, Arxiv, Sciencedirect and Researchgate datasets. From the all previous mentioned datasets we found 37 papers only 26 of these papers are corresponded with the review subject. Keywords: Vehicles Accidents, Machine learning, Object detection, Surveillance camera, Convolutional Neural Network (CNN), Region-based (RCNN) . Volume: 2, Issue: 2, April 2024 Manuscript Code: 717B 105 P-ISSN: 2958-4612 E-ISSN: 2959-5568 Academic Science Journal مراجعة لانظمة الكشف عن حوادث المركبات في الطرق والاخبار عنها بالاعتماد على تقنيات التعلم الالي دعاء هادي نصار و جمال مصطفى التويجري قسم علوم الحاسوب - كلية العلوم - جامعة ديالى الخلاصة مع النمو السريع لحوادث المركبات في طرقنا وتزايد الخسائر البشرية وكلف اصلاح الاضرار الناجمة عن تلك الحوادثلتقليل المخاطر المرافقة للحوادث. الكثير من الباحثين والمختصين اتجهوا لايجاد نظام الكتروني قائم على استخدام تقنياتوخوارزميات التعلم الالي والتعلم العميق او اي من طرق الذكاء الاصطناعي لكشف الحوادث وتوليد اشارة طارئة لتقليلالفجوة الزمنية بين زمن وقوع الحادث ووصول المساعدة الطبية. الغرض الرئيسي لهذه المراجعة هو لجمع كل الدراساتالتي اجريت سابقا للكشف عن الحوادث المرورية بالاعتماد على طرق التعلم الالي و على وجه الخصوص طرق التعلمالعميق. هذه الورقة البحثية قامت باستعراض كل الطرق المستخدمة في تلك الدراسات والنتائج المتحصلة منها والفجواتالبحثية التي تحتويها. لقد اتبعنا خطة بحث ممنهجة تعتمد على تحديد الدراسات ذات الكلمات المفتاحية الاكثر تشابها لضمانان كل الدراسات تتطابق مع موضوع المراجعة. قمنا بتطبيق هذه الخطة بتحميل الدراسات المعنية من المنصات التالية Springer, Elsevier, Electrical and Electronics Engineers (IEEE) xplore ,Google scholar, Arxiv,( Sciencedirect and Researchgate) ومن كل المنصات المذكورة سابقا قمنا بتحديد 35 ورقة بحثية فقط 26 منهاتتوافق مع موضوع المراجعة. الكلمات المفتاحية: حوادث المركبات, تعليم الالي, اكتشاف الاشياء, كاميرة المراقبة المحيطية, خوارزمية الشبكة العصبيةالالتفافية, خوارزمية الشبكة العصبية الالتفافية المعتمدة على تحديد المنطقة. Introduction Every year, according to the numbers, more than a million people die on the world’s roads because of vehicle accident and the costs to overcome the consequences of these Accidents runs into billions [1] . Mobility is the main simplest requirements for people who live in towns or townships. There are numerous methods to mobile from one region to another by airplanes, ships and roads by diverse categ

FAQs

What problem does the review address in vehicle accident detection?

It addresses rising vehicle accidents that lead to higher casualties and increased repair costs, emphasizing the need to reduce the time gap between an accident and medical response.

How did the review select relevant prior studies?

It used a systematic search plan based on keyword similarity and searched multiple academic platforms (including Springer, Elsevier, IEEE Xplore, Google Scholar, arXiv, ScienceDirect, and ResearchGate).

What is the main focus of the surveyed techniques?

The review focuses on machine learning methods for road traffic accident detection, particularly deep learning approaches that can identify events and trigger emergency notification signals.

Aug 4, 2026

Upload by Ezra