The RailVision solution enables improved safety of rail operations, by providing the driver with advanced, real-time, day/night imaging together with obstacle detection and rail scene analysis using deep learning technologies.
The solution generates rapid and reliable alerts to the train driver (or command and control centers) of impending obstacles and related risk instances, hence improving the driver’s situation awareness and response time.
RailVision’s imaging, environmental and dynamic sensors architecture, all provide visualization and detection ranges of more than four times the range compared to the naked eye of qualified drivers and are operable in all weather and illumination conditions.
These capabilities dramatically reduce the risk of a broad range of accidents (including driver-induced errors) and provide an additional safety layer for the operator’s decision making process when alerting, slowing down or potentially stopping moving trains.
RailVision’s solution offers wide field-of-view scene coverage for improved visualization of both rail vicinities and broader surrounds.
Video imagery as well as data from dynamics and environmental sensors are all synchronously compiled and analyzed, thereby generating a variety of security alerts and suspected instances, associated with available geo-located information.
We at RailVision leverage technology, deep learning and data storage to address Home Land Security (HLS) concerns of the railway industry.
The RailVision solution supports modular imaging and processing configurations that support preventive maintenance of railway infrastructure, both along the railway itself, as well as in its surrounding.
For example, the detection of line of sight obstructions along the railway such as vegetation generates a maintenance alert for the removal of such obstructions.
The detection of rail integrity deficiencies is another important condition monitoring function that can provide preventive maintenance and avoiding major damage to the rail infrastructure.
This prominent feature supports the maintenance of various rail-related components in the vicinity of the rail and track ecosystem. Catenary wire inspection and analysis is another important condition monitoring function for electric rail applications, as is measuring wire changes over time and predicting and preventing major defects along the tracks.
The system transfers the collected data to a command and control center where RailVision provides a data storage and a processing platform for designating and analyzing defects.
RailVision’s main line solution provides a range of tools for data archiving, distributed data processing, database management and review of results.
Operators, analysts and maintenance personnel can search through raw video material in the database to retrieve information, draw out statistics data and build behavioral trends and environmental analyses.
Data is collected using the system’s roof-mounted electrooptical, dynamic and environmental sensors and is analyzed using deep learning.