RIECNN : Real-time Image Enhanced CNN for Traffic Sign Recognition

This repository is for a research project at Cairo University, computer engineering department. This paper introduces RIECNN: real-time image enhanced CNN for traffic sign recognition is published in Neural Computing and Applications Journal, Springer 2022.

Abstract

Traffic sign recognition plays a crucial role in the development of autonomous cars to reduce the accident rate and promote road safety. It has been a necessity to address traffic signs that are affected significantly by the environment as well as poor real-time performance for deep-learning state-of-the-art algorithms. In this paper, we introduce Real-Time Image Enhanced CNN (RIECNN) for Traffic Sign Recognition. RIECNN is a real-time, novel approach that tackles multiple, diverse traffic sign datasets, and out-performs the state-of-the-art architectures in terms of recognition rate and execution time. Experiments are conducted using the German Traffic Sign Benchmark (GTSRB), the Belgium Traffic Sign Classification (BTSC), and the Croatian Traffic Sign (rMASTIF) benchmark. Experimental results show that our approach has achieved the highest recognition rate for all Benchmarks, achieving a recognition accuracy of 99.75% for GTSRB, 99.25% for BTSC and 99.55% for rMASTIF. In terms of latency and meeting the real-time constraint, the pre-processing time and inference time together do not exceed 1.3 ms per image. Not only have our proposed approach achieved remarkably high accuracy with real-time performance, but it also demonstrated robustness against traffic sign recognition challenges such as brightness and contrast variations in the environment.

Note

This ReadMe must be updated if any installation requirements or prequisities are needed

DataSet Links

• German Traffic Sign Recognition Benchmark

  • Training Data Set images with annotations : https://sid.erda.dk/public/archives/daaeac0d7ce1152aea9b61d9f1e19370/GTSRB_Final_Training_Images.zip
  • Testing Data Set's ground truth annotations : https://sid.erda.dk/public/archives/daaeac0d7ce1152aea9b61d9f1e19370/GTSRB_Final_Test_GT.zip
  • Testing Data Set images : https://sid.erda.dk/public/archives/daaeac0d7ce1152aea9b61d9f1e19370/GTSRB_Final_Test_Images.zip

• Belgium Traffic Sign Recognition https://btsd.ethz.ch/shareddata/

Access Ready DataSet .pkl files

You can access German and Belgium train and test .pkl files for winning models through this link

Installation Requirements

• Clone Project Repo through ssh link using credentials
• Git pull to get most up-to-date master
• Set up Virtual Env locally or through Google Collab
• pip install -r requirements.txt for python package installation
• Move DataSet Directory to be Source_Code/DataSet folder for consistency
  • Training Data_Set to be : Source_Code/DataSet/Training_DataSet
  • Testing Data_Set to be : Source_Code/DataSet/Testing_DataSet
  • Please ignore .txt files in each directory : made only to commit both folders
• For Belgium DataSet Move DataSet to be in Source_Code/DataSet
  • Belgium Training Data_Set to be : Source_Code/DataSet/BelgiumTSC_Training/Training
  • Belgium Testing Data_Set to be : Source_Code/DataSet/BelgiumTSC_Testing/Testing
  • Please remove readMe.txt files inside Training and Testing Folders or else loading will fail
• Move pkl files to Source_Code/Model/Processed_DataSet folder for consistency
  • TrainDataSet.pkl and TestDataSet.pkl files

Presentation Link

Presentation Link : Here