<?xml version="1.0" encoding="UTF-8"?><xml><records><record><source-app name="Biblio" version="7.x">Drupal-Biblio</source-app><ref-type>17</ref-type><contributors><authors><author><style face="normal" font="default" size="100%">Seghir,  Zineb</style></author><author><style face="normal" font="default" size="100%">Lyamine Guezouli</style></author><author><style face="normal" font="default" size="100%">Kamel Barka</style></author><author><style face="normal" font="default" size="100%">Boubiche, Djallel-Eddine</style></author><author><style face="normal" font="default" size="100%">Homero, Toral-Cruz</style></author><author><style face="normal" font="default" size="100%">Martínez-Peláez, Rafael</style></author></authors></contributors><titles><title><style face="normal" font="default" size="100%">A Real-Time Consensus-Free Accident Detection Framework for Internet of Vehicles Using Vision Transformer and EfficientNet</style></title><secondary-title><style face="normal" font="default" size="100%">AI </style></secondary-title></titles><dates><year><style  face="normal" font="default" size="100%">2026</style></year></dates><urls><web-urls><url><style face="normal" font="default" size="100%">https://doi.org/10.3390/ai7010004</style></url></web-urls></urls><volume><style face="normal" font="default" size="100%">7</style></volume><language><style face="normal" font="default" size="100%">eng</style></language><abstract><style face="normal" font="default" size="100%">&lt;p style=&quot;text-align: justify;&quot;&gt;
	&lt;i&gt;first_page&lt;/i&gt;
&lt;/p&gt;

&lt;p style=&quot;text-align: justify;&quot;&gt;
	&lt;i&gt;settings&lt;/i&gt;
&lt;/p&gt;

&lt;p style=&quot;text-align: justify;&quot;&gt;
	&lt;a href=&quot;https://www.mdpi.com/2673-2688/7/1/4/reprints&quot; id=&quot;js-button-reprints&quot;&gt;Order Article Reprints&lt;/a&gt;
&lt;/p&gt;

&lt;article&gt;
	&lt;p style=&quot;text-align: justify;&quot;&gt;
		Open AccessArticle
	&lt;/p&gt;

	&lt;h1 itemprop=&quot;name&quot; style=&quot;text-align: justify;&quot;&gt;
		A Real-Time Consensus-Free Accident Detection Framework for Internet of Vehicles Using Vision Transformer and EfficientNet
	&lt;/h1&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		by&amp;nbsp;
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		Zineb Seghir
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&lt;sup&gt;&amp;nbsp;1&lt;/sup&gt;&lt;span contenteditable=&quot;false&quot; tabindex=&quot;-1&quot;&gt;&lt;a data-widget=&quot;image&quot; href=&quot;https://orcid.org/0009-0001-6327-7782&quot; rel=&quot;noopener noreferrer&quot; target=&quot;_blank&quot;&gt;&lt;img src=&quot;https://pub.mdpi-res.com/img/design/orcid.png?0465bc3812adeb52?1779174145&quot; title=&quot;ORCID&quot;&gt;&lt;/a&gt;&lt;/span&gt;,
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		Lyamine Guezouli
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&lt;sup&gt;&amp;nbsp;2,*&lt;/sup&gt;&lt;span contenteditable=&quot;false&quot; tabindex=&quot;-1&quot;&gt;&lt;a data-widget=&quot;image&quot; href=&quot;https://orcid.org/0000-0002-7406-7633&quot; rel=&quot;noopener noreferrer&quot; target=&quot;_blank&quot;&gt;&lt;img src=&quot;https://pub.mdpi-res.com/img/design/orcid.png?0465bc3812adeb52?1779174145&quot; title=&quot;ORCID&quot;&gt;&lt;/a&gt;&lt;/span&gt;,
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		Kamel Barka
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&lt;sup&gt;&amp;nbsp;1&lt;/sup&gt;&lt;span contenteditable=&quot;false&quot; tabindex=&quot;-1&quot;&gt;&lt;a data-widget=&quot;image&quot; href=&quot;https://orcid.org/0000-0002-6824-1594&quot; rel=&quot;noopener noreferrer&quot; target=&quot;_blank&quot;&gt;&lt;img src=&quot;https://pub.mdpi-res.com/img/design/orcid.png?0465bc3812adeb52?1779174145&quot; title=&quot;ORCID&quot;&gt;&lt;/a&gt;&lt;/span&gt;,
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		Djallel Eddine Boubiche
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&lt;sup&gt;&amp;nbsp;2,*&lt;/sup&gt;&lt;span contenteditable=&quot;false&quot; tabindex=&quot;-1&quot;&gt;&lt;a data-widget=&quot;image&quot; href=&quot;https://orcid.org/0000-0002-1566-3005&quot; rel=&quot;noopener noreferrer&quot; target=&quot;_blank&quot;&gt;&lt;img src=&quot;https://pub.mdpi-res.com/img/design/orcid.png?0465bc3812adeb52?1779174145&quot; title=&quot;ORCID&quot;&gt;&lt;/a&gt;&lt;/span&gt;,
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		Homero Toral-Cruz
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&lt;sup&gt;&amp;nbsp;3,*&lt;/sup&gt;&lt;span contenteditable=&quot;false&quot; tabindex=&quot;-1&quot;&gt;&lt;a data-widget=&quot;image&quot; href=&quot;https://orcid.org/0000-0002-4421-3775&quot; rel=&quot;noopener noreferrer&quot; target=&quot;_blank&quot;&gt;&lt;img src=&quot;https://pub.mdpi-res.com/img/design/orcid.png?0465bc3812adeb52?1779174145&quot; title=&quot;ORCID&quot;&gt;&lt;/a&gt;&lt;/span&gt;&amp;nbsp;and
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		Rafael Martínez-Peláez
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&lt;sup&gt;&amp;nbsp;4,5&lt;/sup&gt;&lt;span contenteditable=&quot;false&quot; tabindex=&quot;-1&quot;&gt;&lt;a data-widget=&quot;image&quot; href=&quot;https://orcid.org/0000-0003-2188-9892&quot; rel=&quot;noopener noreferrer&quot; target=&quot;_blank&quot;&gt;&lt;img src=&quot;https://pub.mdpi-res.com/img/design/orcid.png?0465bc3812adeb52?1779174145&quot; title=&quot;ORCID&quot;&gt;&lt;/a&gt;&lt;span style=&quot;background:rgba(220,220,220,0.5);background-image:url(https://pg.univ-batna2.dz/profiles/openscholar/libraries/ckeditor/plugins/widget/images/handle.png)&quot;&gt;&lt;img draggable=&quot;true&quot; height=&quot;15&quot; src=&quot;data:image/gif;base64,R0lGODlhAQABAPABAP///wAAACH5BAEKAAAALAAAAAABAAEAAAICRAEAOw==&quot; title=&quot;Cliquer et glisser pour déplacer&quot; width=&quot;15&quot;&gt;&lt;/span&gt;&lt;/span&gt;
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&lt;sup&gt;1&lt;/sup&gt;
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		LaSTIC Laboratory, Computer Science Department, University of Batna 2, Batna 05000, Algeria
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&lt;sup&gt;2&lt;/sup&gt;
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		Laboratory of Renewable Energy, Energy Efficiency and Smart Systems (LEREESI), Higher National School of Renewable Energies, Environment and Sustainable Development (HNS-RE2SD), Batna 05000, Algeria
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&lt;sup&gt;3&lt;/sup&gt;
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		Departamento de Ingeniería y Tecnología, Universidad Autónoma del Estado de Quintana Roo, Chetumal 77019, Mexico
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&lt;sup&gt;4&lt;/sup&gt;
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		Unidad Académica de Computación, Universidad Politécnica de Sinaloa, Mazatlán 82199, Mexico
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&lt;sup&gt;5&lt;/sup&gt;
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		Departamento de Ingeniería de Sistemas y Computación, Universidad Católica del Norte, Antofagasta 1270709, Chile
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&lt;sup&gt;*&lt;/sup&gt;
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		Authors to whom correspondence should be addressed.
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&lt;em&gt;AI&lt;/em&gt;&amp;nbsp;&lt;b&gt;2026&lt;/b&gt;,&amp;nbsp;&lt;em&gt;7&lt;/em&gt;(1), 4;&amp;nbsp;&lt;a href=&quot;https://doi.org/10.3390/ai7010004&quot;&gt;https://doi.org/10.3390/ai7010004&lt;/a&gt;
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		Submission received: 12 November 2025&amp;nbsp;/&amp;nbsp;Revised: 14 December 2025&amp;nbsp;/&amp;nbsp;Accepted: 16 December 2025&amp;nbsp;/&amp;nbsp;Published: 22 December 2025
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&lt;a aria-controls=&quot;drop-supplementary-1824006&quot; aria-expanded=&quot;false&quot; data-dropdown=&quot;drop-download-1824006&quot;&gt;Download&lt;i&gt;keyboard_arrow_down&lt;/i&gt;&lt;/a&gt;&amp;nbsp;
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&lt;a data-counterslink=&quot;https://www.mdpi.com/2673-2688/7/1/4/browse&quot; data-target=&quot;article-popup&quot; href=&quot;https://www.mdpi.com/2673-2688/7/1/4#&quot;&gt;Browse Figures&lt;/a&gt;
	&lt;/p&gt;

	&lt;p style=&quot;text-align: justify;&quot;&gt;
		&amp;nbsp;&lt;a href=&quot;https://www.mdpi.com/2673-2688/7/1/4/notes&quot;&gt;Versions&amp;nbsp;Notes&lt;/a&gt;
	&lt;/p&gt;

	&lt;section&gt;
		&lt;p style=&quot;text-align: justify;&quot;&gt;
			&amp;nbsp;
		&lt;/p&gt;

		&lt;section id=&quot;html-abstract&quot;&gt;
			&lt;h2 id=&quot;html-abstract-title&quot; style=&quot;text-align: justify;&quot;&gt;
				Abstract
			&lt;/h2&gt;

			&lt;p style=&quot;text-align: justify;&quot;&gt;
				&lt;b&gt;Objectives&lt;/b&gt;: Traffic accidents cause severe social and economic impacts, demanding fast and reliable detection to minimize secondary collisions and improve emergency response. However, existing cloud-dependent detection systems often suffer from high latency and limited scalability, motivating the need for an edge-centric and consensus-free accident detection framework in IoV environments.&amp;nbsp;&lt;b&gt;Methods&lt;/b&gt;: This study presents a real-time accident detection framework tailored for Internet of Vehicles (IoV) environments. The proposed system forms an integrated IoV architecture combining on-vehicle inference, RSU-based validation, and asynchronous cloud reporting. The system integrates a lightweight ensemble of Vision Transformer (ViT) and EfficientNet models deployed on vehicle nodes to classify video frames. Accident alerts are generated only when both models agree (vehicle-level ensemble consensus), ensuring high precision. These alerts are transmitted to nearby Road Side Units (RSUs), which validate the events and broadcast safety messages without requiring inter-vehicle or inter-RSU consensus. Structured reports are also forwarded asynchronously to the cloud for long-term model retraining and risk analysis.&amp;nbsp;&lt;b&gt;Results&lt;/b&gt;: Evaluated on the CarCrash and CADP datasets, the framework achieves an F1-score of 0.96 with average decision latency below 60 ms, corresponding to an overall accuracy of 98.65% and demonstrating measurable improvement over single-model baselines.&amp;nbsp;&lt;b&gt;Conclusions&lt;/b&gt;: By combining on-vehicle inference, edge-based validation, and optional cloud integration, the proposed architecture offers both immediate responsiveness and adaptability, contrasting with traditional cloud-dependent approaches.
			&lt;/p&gt;
		&lt;/section&gt;
	&lt;/section&gt;
&lt;/article&gt;
</style></abstract><issue><style face="normal" font="default" size="100%">1</style></issue></record></records></xml>