Skip to content

A web app that recognizes the speed limits on speedsigns on the road and throws an alert on the HUD if your car is going over the speed limit

Notifications You must be signed in to change notification settings

Jhalak151/Speedsign-Detection

 
 

Repository files navigation

Developed a web app that takes a photo of the road as input, recognizes if the photo has any road signs and if the photo has a speed sign it extrapolates the speed limit given on the speed sign. It can also tell you if your car is going over the speed limit by giving a pop-up on the HUD of the car.

Front end - Streamlit + HTML + CSS

Back end - Python, Tensorflow library, Faster R-CNN neural network

Instructions -

  1. In model/faster_rcnn_resnet50 unzip the inference-graph.7z file using the 7-zip extension (or anything else)
  2. After extraction ensure that the structure is of the form model/faster_rcnn_resnet50/inference_graph where - Inference_graph has only 1 file called frozen_inference_graph.pb.
    (thus, path of frozen_inference_graph.pb is - model/faster_rcnn_resnet50/inference_graph/frozen_inference_graph.pb)

About

A web app that recognizes the speed limits on speedsigns on the road and throws an alert on the HUD if your car is going over the speed limit

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 91.9%
  • CSS 4.6%
  • HTML 3.5%