https://github.com/aggrathon/trafficsignrecognizer

A neural network for recognizing traffic signs in images

https://github.com/aggrathon/trafficsignrecognizer

Science Score: 13.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
  • DOI references
  • Academic publication links
  • Committers with academic emails
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (5.3%) to scientific vocabulary

Keywords

ai android android-app machine-learning neural-network python python3 tensorflow
Last synced: 5 months ago · JSON representation

Repository

A neural network for recognizing traffic signs in images

Basic Info
  • Host: GitHub
  • Owner: Aggrathon
  • License: apache-2.0
  • Language: Python
  • Default Branch: master
  • Size: 9.11 MB
Statistics
  • Stars: 0
  • Watchers: 2
  • Forks: 0
  • Open Issues: 0
  • Releases: 2
Topics
ai android android-app machine-learning neural-network python python3 tensorflow
Created over 8 years ago · Last pushed over 4 years ago
Metadata Files
Readme License

README.md

TrafficSignRecognizer

A neural network for recognizing traffic signs in images. The network is used in an Android app for recording images through the windscreen of a car and remembering the last sign. Even though that it is function, do not use it while driving since it is mostly a distraction from the road.

App

The Android app can be downloaded here.
When evaluated on the training material it reached a precision of 99.6%. Since it was trained on roughly 30 000 images this accuracy doesn't seem to be due to overfitting. In practise it has difficulties with tunnels and anything not recorded from a road but is otherwise pretty accurate. Since it is only trained on Finnish signs, your experience may vary.

Neural Network

The data folder contains some scipt for easily sort through source material and prepare it for learning.
Use train.py to train the network and export.py to prepare the trained model for use in the app. In the model.py is the layout of the network defined and it looks like this:

| Convolution 1 | Convolution 2 | Convolution 3 | Fully Connected 1 | Fully Connected 2 | Prediction | | ------------- | ------------- | ------------- | ----------------- | ----------------- | ---------- | | Size: 32 | Size: 48 | Size: 64 | Size: 256 | Size: 64 | Size: 1 | | Conv2d ReLU | Conv2d ReLU | Conv2d ReLU | ReLU | ReLU | Sigmoid | | Max Pooling | Max Pooling | Max Pooling | Dropout | Dropout | | | Normalization | Normalization | Normalization | | | |

Dependencies

  • Python 3
  • Tensorflow
  • Pygame (for source material sorting)

Owner

  • Name: Anton Björklund
  • Login: Aggrathon
  • Kind: user
  • Company: @edahelsinki

GitHub Events

Total
Last Year

Committers

Last synced: 7 months ago

All Time
  • Total Commits: 36
  • Total Committers: 1
  • Avg Commits per committer: 36.0
  • Development Distribution Score (DDS): 0.0
Past Year
  • Commits: 0
  • Committers: 0
  • Avg Commits per committer: 0.0
  • Development Distribution Score (DDS): 0.0
Top Committers
Name Email Commits
Aggrathon a****o@g****m 36

Issues and Pull Requests

Last synced: 7 months ago

All Time
  • Total issues: 0
  • Total pull requests: 0
  • Average time to close issues: N/A
  • Average time to close pull requests: N/A
  • Total issue authors: 0
  • Total pull request authors: 0
  • Average comments per issue: 0
  • Average comments per pull request: 0
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
Past Year
  • Issues: 0
  • Pull requests: 0
  • Average time to close issues: N/A
  • Average time to close pull requests: N/A
  • Issue authors: 0
  • Pull request authors: 0
  • Average comments per issue: 0
  • Average comments per pull request: 0
  • Merged pull requests: 0
  • Bot issues: 0
  • Bot pull requests: 0
Top Authors
Issue Authors
Pull Request Authors
Top Labels
Issue Labels
Pull Request Labels