Skip to content

A-D-I-T-Y-A/Offline-Traffic-Sign-Recognition-On-Mobile-Device

Repository files navigation

Offline-Traffic-Sign-Recognition-On-Mobile-Device

Udacity Machine Learning Engineer Nanodegree, Capstone Project

This project focuses on using deep learning to enable traffic sign recognition without internet connectivity on a mobile device.

Requirements

The following libraries are required to run the notebooks (Python 3.5)

  • Tensorflow
  • Keras
  • OpenCV
  • Numpy
  • sklearn
  • Matplotlib
  • PIL
  • Pandas
  • Pickle

Dataset

The dataset can be obtained from here.

  • Extract the zip file into the root folder, i.e the GTSRB folder should be in the same directory as the notebooks.

Setting up android project

The android project requires the OpenCV-android-sdk to be present in the same directory that contains the project folder. The OpenCV-android-sdk can be downloaded from here

APK

If there is a problem building the project and you just want to use the APK, it can be obtained from here

About

Udacity Machine Learning Engineer Nanodegree, Capstone Project

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published