Smart Video Surveillance System For Level Crossing: A Systematic Literature Review

A level crossing is an intersection where a railway line crosses a road or path. The government has made regulation that set level crossings throughout Indonesia, Provisions on traffic and road transport in Law No. 22 of 2009. However, data from PT Kereta Api Indonesia (KAI) shows that during 2020 there was 198 accidents and around 5000 trespasses occurred at level crossings. The surveillance system currently used in level crossings is still conventional, which only relies on supervisory officers and some level crossings monitored by CCTV. Research in the field of video surveillance systems with a computer vision approach has been widely carried out in the past decade. Most researches on video surveillance systems try to replace human functions or traditional systems with video processing algorithms. This paper will review the literature on the application of artificial intelligence for the surveillance system level crossings published from 2015 to 2021. The literature review process consists of analyzing the document to determine the scope of the investigation using intelligent video surveillance systems, frameworks, methods and datasets to detect level crossing violators. In the conclusion of this paper, it will be explained about the challenges and opportunities of research on the object detected violations using intelligent video surveillance system in the future.