Identify 2D polygons from a point cloud using a genetic algorithm
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
Test_data
Test_images
doc
ga
.gitignore
License
Readme.md
corners.csv
generate_test_data.py
identify.py
polygon.py
results.txt
test.py

Readme.md

Introduction

This is a tool to identify polygons from noisy point clouds, using a genetic algorithm-based optimizer. It can take as its input data a CSV file of data points (one point per line) or a PNG image of the sonar data, where pixels with non-zero red values are considered to be data points.

Requirements:

  • Python (2.7 tested)
  • Numpy
  • Matplotlib
  • PyTables

Usage:

python identify.py Test_data/data.csv

OR python identify.py Test_images/sample_pent.png