National Repository of Grey Literature 1 records found  Search took 0.00 seconds. 
Identifikace železničních vozů na základě obrazové informace
ŠEPS, Pavel
This master's thesis focuses on the design and implementation of railway wagon indetification system based on the identifying information on the side of each wagon. It describes the search for the optimal image preprocessing, comparison of OCR services, search for the suitable technologies and evaluation of the quality of the finished application. Part of this thesis also describes existing solutions and the application development and functionality. To obtain the images of the wagons, a photo trap is designed and assembled, which is then located near the train track. A custom neural network was built to find the identifying information and a custom annotation application was developed to create the test and training sets. The final identification application then implements this neural network. It also uses Azure OCR services. It is designed to automate the manual process and is ready for use in real operation.

Interested in being notified about new results for this query?
Subscribe to the RSS feed.