This repository is made for testing different computer vision methods by simulating a mission done in Trondheimsfjorden in April 2016
-
Python: 2.7.11 —Anaconda 2.5.0 (64-bit)
-
scipy: 0.17.0
-
numpy: 1.10.4
-
matplotlib: 1.5.1
-
pandas: 0.17.1
-
sklearn: 0.17
- images_L&R_avoided obstacle first try
- repeatExperiment
- images close to transponder
Stereo camera method:
- Dipsarity method
Mono camera methods also called the texture based methods:
-
LBP ROI method
-
Haralick ROI method
-
SLIC Superpixel Locally Binary Pattern method
-
SLIC Superpixel Harlick method
Bellow are some examples:
Image with an obstacle
Predicted image with lbp
Display maskedImage image
Display countour center and boundingbox of the predicted obstacle
Unifrom Local Binary Pattern (watch this to understand better):
- Lars Brusletto Master thesis