FLIR-human-detection icon indicating copy to clipboard operation
FLIR-human-detection copied to clipboard

A python implementation of a simple yet effective algorithm for detecting humans in low-res thermal images that are usually captured using FLIR cameras such as FLIR Lepton and FLIR One.

Human Detection in Low Resolution Thermal Image Cameras (FLIR Lepton)

This code demonstartes a hand-crafted image processing algorithm for detecting humans in low-res thermal images that are usually captured using FLIR cameras such as FLIR Lepton and FLIR One.

Requirements

  • Python 2.x
  • Numpy 1.13.x
  • OpenCV 2.4.x
  • Tqdm 4.14.0

Content

  • data_utils.py: Utlitiy functions used for serving the main functionality of the code.
  • human_detector.py: The building blocks of human detection algorithm.
  • main.py: Sample code to test the functionality of the human detection algorithm on a given folder of raw radimetry data in .txt files.

Usage

  • In order to test the human detection algorithm on your raw radimoetry data folder, run the main.py script. For more information on the available options/argument to pass to the script, run the following command:
python main.py --help