RT info:eu-repo/semantics/article T1 A self-adaptive automatic incident detection system for road surveillance based on deep learning A1 Bartolomé Hornillos, César A1 San José Revuelta, Luis Miguel A1 Aguiar Pérez, Javier Manuel A1 García Serrada, Carlos A1 Vara Pazos, Eduardo A1 Casaseca de la Higuera, Juan Pablo K1 Security K1 Systems and Data Security K1 Artificial intelligence K1 Traffic safety K1 Seguridad en carretera K1 Automobiles - Safety appliances K1 Seguridad vial K1 Tráfico, Ingeniería del - Informática K1 Highway communications K1 Machine learning K1 Aprendizaje automático K1 Self-adaptive software K1 Video surveillance K1 Video K1 Artificial intelligence K1 1203.04 Inteligencia Artificial K1 1203.25 Diseño de Sistemas Sensores AB We present an automatic road incident detector characterised by a low computational complexity for easy implementation in affordable devices, automatic adaptability to changes in scenery and road conditions, and automatic detection of the most common incidents (vehicles with abnormal speed, pedestrians or objects falling on the road, vehicles stopped on the shoulder, and detection of kamikaze vehicles). To achieve these goals, different tasks have been addressed: lane segmentation, identification of traffic directions, and elimination of unnecessary objects in the foreground. The proposed system has been tested on a collection of videos recorded in real scenarios with real traffic, including areas with different lighting. Self-adaptability (plug and play) to different scenarios has been tested using videos with significant scene changes. The achieved system can process a minimum of 80 video frames within the camera’s field of view, covering a distance of 400 m, all within a span of 12 s. This capability ensures that vehicles travelling at speeds of 120 km/h are seamlessly detected with more than enough margin. Additionally, our analysis has revealed a substantial improvement in incident detection with respect to previous approaches. Specifically, an increase in accuracy of 2–5% in automatic mode and 2–7% in semi-automatic mode. The proposed classifier module only needs 2.3 MBytes of GPU to carry out the inference, thus allowing implementation in low-cost devices. PB MDPI SN 1424-8220 YR 2024 FD 2024 LK https://uvadoc.uva.es/handle/10324/68101 UL https://uvadoc.uva.es/handle/10324/68101 LA eng NO Sensors, 2024, Vol. 24, Nº. 6, 1822 NO Producción Científica DS UVaDOC RD 27-jun-2024