Recently, the smart world, smart cars, and so on plays a major role. To ensure Traffic Safety, the development of smart cars requires the detection and recognition of traffic signs. The algorithm is the extended work on the classical LeNet-5 CNN model. The proposed technique makes use of Gabor based kernel followed by a normal convolutional kernel after the pooling layer. The optimizer technique used here is the Adams method. Hue, Saturation Value color space features have a speed of detection is faster and low suffering from illumination. The proposed technique for traffic sign recognition is evaluated using the German Traffic Sign Recognition Benchmark. The proposed technique gives an accuracy of nearly 99%.
Traffic Sign Recognition using Deeplearning for Autonomous Driverless Vehicles
08.04.2021
2514374 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Dis/orientations on driverless driving and autonomous vehicles?
Taylor & Francis Verlag | 2019
|DSCC2017-5187 Traffic Sign Recognition in Autonomous Vehicles Using Edge Detection
British Library Conference Proceedings | 2017
|AUTONOMOUS VEHICLES - Grand Challenge sparks interest in driverless US military vehicles
Online Contents | 2005
|