IN4393 Computer Vision - Final Project

Road Sign Detection on Google Streetview

Set up environment

In order to run our code, please check out our repository with git. After downloading your copy of our code, please unzip the training_data.zip file that is located in the folder data.

Before the code can be run, the python environment has to be prepared. To do this, please install Anaconda

Now that Anaconda is installed, please start Anaconda Navigator.

1. click on Environments
2. click Import.
3. Browse your harddisk to the path where you placed our code and select the environment.yml file in the env directory.
4. Press Ok and press Import. The python environment will now be prepared.

In the mean time, please add our project directories to your PYTHONPATH.
On windows this can be done by pressing windows_key + r
type: SystemPropertiesAdvanced and press Enter, then go to the tab Advanced .

1. press the button Environment Variables.
2. Press New... at user variables.
3. Variable name: PYTHONPATH
4. Variable value: path_to_our_code\src;path_to_our_code\src\data_preparation;path_to_our_code\src\detection;path_to_our_code\src\features;path_to_our_code\src\test;path_to_our_code\src\webapp
   Where path_to_our_code should be replaced with the path to our code.

Once the python environment is installed by Anaconda, an environment named road_sign_detection should be visible.

Run our webapplication

Please open a Command Prompt and navigate to path_to_our_code\src\webapp and type activate road_sign_detection.
Now run python server.py. The following message should appear:

 * Restarting with stat
 * Debugger is active!
 * Debugger PIN: 
 * Running on http://127.0.0.1:5000/ (Press CTRL+C to quit

The webserver is runnning and the webapp can be used by visiting http://127.0.0.1:5000 in your browser.
The result should look like this: