MaxLing / gmm_color_object_detection

a hybrid Gaussian Mixture Model for color segmentation, and connected component analysis for object detection

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

UPenn ESE650 Learning in Robotics, Project 1

  1. Put training images into folder .\roipoly_annotate\train
    Put testing images into folder .\roipoly_annotate\test
  2. Execute annotate.py to hand-label RedBarrel and TrickyCases. Remember to change colorClass in the code between. TrickyCases are basically light yellow ceiling, windows and dark red floor.
  3. Run train.py to get a GMM Model
  4. Once the model is obtained, run predict.py to detect red barrel. A image with bounding box and centroid will appear, also info like centroid, distance and numbers will be at command window.

About

a hybrid Gaussian Mixture Model for color segmentation, and connected component analysis for object detection


Languages

Language:Python 100.0%