Title :
Traffic Sign Recognition With Hinge Loss Trained Convolutional Neural Networks
Author :
Junqi Jin ; Kun Fu ; Changshui Zhang
Author_Institution :
Dept. of Autom., Tsinghua Univ., Beijing, China
Abstract :
Traffic sign recognition (TSR) is an important and challenging task for intelligent transportation systems. We describe the details of our model´s architecture for TSR and suggest a hinge loss stochastic gradient descent (HLSGD) method to train convolutional neural networks (CNNs). Our CNN consists of three stages (70-110-180) with 1162 284 trainable parameters. The HLSGD is evaluated on the German Traffic Sign Recognition Benchmark, which offers a faster and more stable convergence and a state-of-the-art recognition rate of 99.65%. We write a graphics processing unit package to train several CNNs and establish the final classifier in an ensemble way.
Keywords :
gradient methods; graphics processing units; intelligent transportation systems; learning (artificial intelligence); neural nets; object recognition; traffic engineering computing; CNN; German traffic sign recognition benchmark; HLSGD method; TSR; graphics processing unit package; hinge loss stochastic gradient descent method; hinge loss trained convolutional neural networks; intelligent transportation systems; recognition rate; Convolution; Fasteners; Feature extraction; Kernel; Neural networks; Training; Vectors; Convolutional neural networks (CNNs); hinge loss; stochastic gradient descent (SGD); traffic sign recognition (TSR);
Journal_Title :
Intelligent Transportation Systems, IEEE Transactions on
DOI :
10.1109/TITS.2014.2308281