Project AutoVis: Automatic Visualization via Webcam

AutoVis is a tool for automating visualization of colored reactions.

Usage

Two computers are needed, one for running the experiment (i.e. taking images using a built-in camera or an external webcam) and one for running the communication manager that detects the accumulation of image data and initiates the data analysis.

Computer 1: For taking images on the webcam, execute:

$ python MainExperiment.py

The user will input the path to the Dropbox directory, reaction ID, duration of webcam, and image-taking frequency.

Computer 2: For initiating the communication manager, execute:

$ python MainAnalysis.py

The user will input the path to the Dropbox directory and reaction ID.

Installation

To install Autovis, the user needs to install the dependencies listed below.

• For opencv-python, we recommend installing via pip: pip install opencv-python. For installation on Windows, we recommend following the steps here.

• For numpy and matplotlib, we recommend installing via pip: python -m pip install --user numpy matplotlib. For more detailed instructions, refer to this set of instructions.

• For seaborn, we recommend installing via pip: pip install seaborn.

Coming soon: we will work on making Autovis pip-installable.

Dependencies and Versions Used

• python 3.5
• opencv-python 4.1.0.25
• numpy 1.16.4
• matplotlib 2.2.2
• seaborn 0.8.1

Key functionalities

Authors

AutoVis was written by Physical Sciences 50 (Spring 2018) at Harvard University.