✔ The Intelligent Bus Stop Recognition System identifies Bus Stops using images acquired from cameras placed
on a bus using a lightweight and simple hybrid nearest neighbor algorithm and implemented using a
RaspberryPi Zero, thereby eliminating the need for sensor networks. This is our final year undergrad thesis.
✔ A Day and a Night Classifier is implemented separately in run-time, so that the Recognition Engine can be
automated 24 x 7, using a light sensor or a timer.
✔ The Bus Stops considered are in the urban city of Chennai with the bus stops being easily scalable, and a
detailed version of the dataset recorded and the implementation is given in the report which can be found
here: Intelligent-Bus-Stop-Recognition-System.
✔ The accuracy and efficiency of the algorithm are also explored, in terms of both memory and space.
✔ We focused on using low-resolution images to achieve the same result while not compromising on our accuracy.
For more details about the data, contact Gautham Krishna or Ateendra Ramesh.
Steps for running the Intelligent Bus Stop Recognition System:
-
Install Python 2.7 depending on your Operating System, via Continuum Anaconda
-
Open "cmd" or "Power Shell" in Windows, or Open a Terminal in Linux.
-
In the terminal, type the required classifier approach code as, "python file_name.py" and press Enter.
-
Repeat the above step for each of the three approaches for day and night.