Creative coding using openFrameworks - Workshop: Simple Object Detection using OpenCV
Switch branches/tags
Nothing to show
Clone or download
Nico Reski Nico Reski
Nico Reski and Nico Reski Updated README.
Latest commit 8b76225 Nov 10, 2016
Permalink
Failed to load latest commit information.
src Checked compatibility with OF 0.9.6. Oct 25, 2016
.gitignore Initial commit Oct 18, 2015
LICENSE.md Updated readme and licence. Oct 25, 2016
README.md Updated README. Nov 10, 2016

README.md

Creative coding using openFrameworks - Workshop: Simple Object Detection using OpenCV

complementary slides: GitHub, 9.59 MB

openFrameworks version: 0.9.6

The purpose of this project is to illustrate and teach the access of connected cameras, basic image manipulation as well as some simple object detection capabilities of the OpenCV addon using the the creative coding toolkit openFrameworks. This activity is meant to be held in a workshop-like format and steb by step to the participants using live coding, supported by summarizing slides.

Following, a brief overview about the project's content is given in order as they are illustrated during the workshop:

  1. Setup a camera input source and draw the camera input to the application window. Furthermore, a helper function is created in order to log information about available camera devices.
  2. Create a HSV representation of the original camera input image as well as extracting hue, saturation and value color channel information into separate image instances.
  3. Simple Image manipulation in order to assist the later object detection by taking the difference between a registered background and the saturation color channel image. Additionally, the contrast is increased via defined threshold value.
  4. Using the OpenCV Contour Finder instance, objects ("blobs") are detected and visualized accordingly.
  5. Using the center point of each detected object ("blob"), the color is extracted and visualized as a circle overlay.

License

MIT License, see LICENSE.md