Source-railway-technology.com
Engineering and design company Klas Telecom is to launch a new deep learning computer vision solution for the railway industry.
The solution is due to be launched next week at the InnoTrans 2018 event in Germany.
Alongside providing deep learning computer vision, the solution will offer object detection capabilities to trains and the network edge.
The multiplatform solution is designed to enable railway network operators to deploy a number of applications, including video surveillance and smart cameras.
These applications are capable of replicating human vision and conducting various jobs such as vehicle identification, intruder detection and empty seat recognition to ultimately help improve safety, security and analyse risks.
In addition, the new solution features Klas Telecom TRX Connected Transportation Platform R6 (TRX R6), a router/server and six-modem cellular gateway device, and Intel OpenVINO convolutional neural networks (CNN). All of these features facilitate deep learning.
TRX R6 is designed to serve the railway communications systems market and is equipped with Intel CNN-based machine learning models, in a single 4.5kg, 250mm x 279mm x 76.5mm chassis.
The solution also complies with environmental standards for rail, as well as containing Intel Core i7, i5 and i3 processors and provides 32GB of RAM and up to 8TB of built-in storage.
Klas Telecom Transportation Business Development director Brendan Fleming said: “It is exciting to be able to bring artificial intelligence and machine learning into railway use cases such as traffic monitoring, vehicle identification, pedestrian identification, real-time risk analysis, security and passenger information services.
“It means our customers can save time by using computers to perform intelligent functions and predictive analysis. Ultimately, time means money and we hope to save our customers both.”
Klas Telecom currently has three offices in the US, with another in Dublin, Ireland.