A Self-Adaptive Automatic Incident Detection System for Road Surveillance Based on Deep Learning.

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.

Indicadores de Atención Pre Hospitalaria para incidentes viales en Colombia / Prehospital care indicators for road incidents in Colombia / Indicadores de Assistência Pré-Hospitalar para incidentes viários na Colômbia

Resumen La Atención Pre Hospitalaria (APH) reduce la letalidad, minimiza la morbilidad y sus secuelas en situaciones de urgencia y emergencia. Evaluar su grado de desarrollo es relevante. Metodología: Revisión sistemática de literatura, consulta a expertos y prueba piloto de un sistema de indicadores de APH para incidentes viales en Colombia implementado en 6 ciudades, un departamento y una concesión vial. Resultados: Según lineamientos políticos, normativos, conceptuales y metodológicos, internacionales y nacionales, se propuso un sistema de indicadores para ser aplicado según el desarrollo territorial de la APH; hubo diversas dificultades para implementarlo. Conclusión: Es necesario fortalecer los servicios de APH y estandarizar un sistema de información para su monitoreo en el país.

Abstract Prehospital care (PHC) reduces lethality and minimizes morbidity and its effects in urgency and emergency situations. Assessing its degree of development is thus relevant. Methodology: the authors used a systematic review of the literature, inquiries to experts and a pilot test with a system of PHC indicators for road incidents in Colombia which was implemented in 6 cities, one department and one highway concession. Results: according to political, regulatory, conceptual, methodological, international and national guidelines, a system of indicators was proposed to be applied in accordance with the territorial development of PHC. However, there were a number of difficulties to implement it. Conclusion: it is necessary to strengthen PHC and standardize an information system for monitoring it in Colombia.

Resumo A Assistência Pré-Hospitalar (APH) reduz a letalidade, minimiza a morbidade e as suas sequelas em situações de urgência e emergência. Avaliar seu grau de desenvolvimento é relevante. Metodologia: Revisão sistemática de literatura, consulta a especialistas e teste piloto de um sistema de indicadores de APH para incidentes viários na Colômbia implantado em 6 cidades, um estado do país e uma concessão viária. Resultados: De acordo com diretrizes políticas, normativas, conceituais e metodológicas, internacionais e nacionais, propôs-se um sistema de indicadores para ser aplicado segundo o desenvolvimento territorial da APH; houve diferentes dificuldades para a sua implantação. Conclusão: É preciso fortalecer os serviços de APH e padronizar um sistema de informação para seu monitoramento no país.

Farm tractors on Swedish public roads--age-related perspectives on police reported incidents and injuries.

BACKGROUND: Agriculture consistently ranks as one of the highest risk industries in North America and Europe. In addition to vehicle injuries and other injuries that occur on farms, farm vehicle drivers are also involved in collisions involving tractors and other slow-moving vehicles (SMVs) on public roads. These collisions often lead to injuries among farmers, family members, farm workers, and other road users. OBJECTIVE: To conduct a demographic analysis of road traffic incidents involving tractors in Sweden during the time period 1992-2009, with special consideration of how incidents vary with driver age. METHODS: Statistics from 2,305 police reports describing tractor-related road traffic incidents in Sweden in 1992-2009 were analysed with respect to driver age, type of incident, severity of injury, type of road user and other circumstances at the accident site. RESULTS: Tractors of all kinds were involved in 128 road traffic incidents annually, with 7 people killed, 44 seriously injured and 143 suffering minor injuries each year. The annual number of fatalities represented 1.2% of all Swedish road fatalities. Most road traffic incidents with farm tractors involved people aged 25-55 years (mean 45 years). However, most farm tractor drivers killed or injured belonged to younger or older age groups. Drivers aged 12-16 years were over-represented in tractor incidents with no other vehicle involved. Older tractor drivers (> 55 years) were more often involved in incidents with passenger vehicles on entering traffic flows. The youngest tractor drivers aged 12-16 years were more often involved in road traffic incidents during school holidays, and both youngest and oldest drivers (>65 years) during harvest time. CONCLUSIONS: Sweden has an ageing fleet of tractors, so increased attention to vehicle maintenance is needed to improve road safety. The over-representation of young children in tractor incidents suggests that it is questionable whether they should be allowed to operate farm vehicles. Farm vehicle drivers suffering the inevitable effects of ageing need increase their awareness of added risks. As road traffic incidents with tractors often involve private vehicles, creating awareness among the public of slow-moving farm vehicles is essential for improving overall road safety.

Critical power two wheeler driving patterns at the emergence of an incident.

The paper proposes a methodology based on Bayesian Networks for identifying the power two wheeler (PTW) driving patterns that arise at the emergence of a critical incident based on high resolution driving data (100Hz) from a naturalistic PTW driving experiment. The proposed methodology aims at identifying the prevailing PTW drivers' actions at the beginning and during critical incidents and associating the critical incidents to specific PTW driving patterns. Results using data from one PTW driver reveal three prevailing driving actions for describing the onset of an incident and an equal number of actions that a PTW driver executes during the course of an incident to avoid a crash. Furthermore, the proposed methodology efficiently relates the observed sets of actions with different types of incidents occurring during overtaking or due to the interactions of the rider with moving or stationary obstacles and the opposing traffic. The observed interrelations define several driving patterns that are characterized by different initial actions, as well as by different likelihood of sequential actions during the incident. The proposed modeling may have significant implications to the efficient and less time consuming analysis of the naturalist data, as well as to the development of custom made PTW driver assistance systems.