Novel Deep Learning Model for Traffic Sign Detection Using Capsule Networks

TL;DR

This paper introduces a capsule network-based deep learning model for traffic sign detection that outperforms CNNs by effectively capturing pose and orientation, achieving 97.6% accuracy on the GTSRB dataset.

Contribution

The paper presents a novel capsule network architecture for traffic sign detection that eliminates manual feature extraction and improves robustness against spatial variances and adversarial attacks.

Findings

Achieved 97.6% accuracy on GTSRB dataset.

Outperformed traditional CNN-based methods.

Enhanced resistance to adversarial attacks.

Abstract

Convolutional neural networks are the most widely used deep learning algorithms for traffic signal classification till date but they fail to capture pose, view, orientation of the images because of the intrinsic inability of max pooling layer.This paper proposes a novel method for Traffic sign detection using deep learning architecture called capsule networks that achieves outstanding performance on the German traffic sign dataset.Capsule network consists of capsules which are a group of neurons representing the instantiating parameters of an object like the pose and orientation by using the dynamic routing and route by agreement algorithms.unlike the previous approaches of manual feature extraction,multiple deep neural networks with many parameters,our method eliminates the manual effort and provides resistance to the spatial variances.CNNs can be fooled easily using various adversary attacks and capsule networks can overcome such attacks from the intruders and can offer more reliability in traffic sign detection for autonomous vehicles.Capsule network have achieved the state-of-the-art accuracy of 97.6% on German Traffic Sign Recognition Benchmark dataset (GTSRB).