The railway maintenance is a particular application context required in order to prevent any dangerous situation. With the growing of the high-speed railway traffic, automatic inspection systems able to detect rail defects, sleepers' anomalies, as well as missing fastening elements, become strategic since they could increase the ability in the detection of defects and reduce the inspection time in order to guarantee more frequent maintenance of the railway network. This paper presents a patented fully automatic and configurable real-time vision system able to detect the presence/absence of the fastening bolts that fix the rails to the sleepers. It gets an accuracy of 99.9%, and, thanks to its parallel processing allowed by a Graphic Processing Unit, reaches an average throughput of 187 km/h, speeding up of about 287% the performance of a quadcore CPU implementation.
A GPU-based vision system for real time detection of fastening elements in railway inspection / De Ruvo, P.; Distante, A.; Stella, E.; Marino, Francescomaria. - (2009), pp. 2333-2336. (Intervento presentato al convegno 2009 IEEE International Conference on Image Processing, ICIP tenutosi a Cairo, Egypt nel November 7-12, 2009) [10.1109/ICIP.2009.5414438].
A GPU-based vision system for real time detection of fastening elements in railway inspection
MARINO, Francescomaria
2009-01-01
Abstract
The railway maintenance is a particular application context required in order to prevent any dangerous situation. With the growing of the high-speed railway traffic, automatic inspection systems able to detect rail defects, sleepers' anomalies, as well as missing fastening elements, become strategic since they could increase the ability in the detection of defects and reduce the inspection time in order to guarantee more frequent maintenance of the railway network. This paper presents a patented fully automatic and configurable real-time vision system able to detect the presence/absence of the fastening bolts that fix the rails to the sleepers. It gets an accuracy of 99.9%, and, thanks to its parallel processing allowed by a Graphic Processing Unit, reaches an average throughput of 187 km/h, speeding up of about 287% the performance of a quadcore CPU implementation.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
