I need to develop an algorithm for a drone line tracking school project.
The goal of this part of the project is to detect the nearest and biggest line of a photo.

OpenCV is used to detect line shapes of a photo.
Some shape filtering are processed with Numpy.
Matplotlib is used to analyse the results.

Filtering criterias have been defined to make line detection more reliable.
Instead of switching lines frequently during the same flight, the algorithm analyses the green lines accurately and chooses the red one as the right line to track based on the chosen criteria.

The LENGTH filter is simply based on the distance between the lines' ends.
The distance calculus is focused on the Y axis.

The result of the FORWARD filter is likely to be a short and straight line.
This filter is based on the comparison of the line ends' positions on the X axis to be sure the line is not too oriented.

{
    "names" : {
        "LINES" : 0,
        "FILTERED_LINES" : 1,
        "LONG_LINES" : 2
    },
    "values" : [
        "lines",
        "filtered_lines",
        "long_lines"
    ],
    "criterias" : {
        "MIN_LENGTH" : 250,
        "MAX_VERTICAL_POSITION" : 200,
        "MAX_CENTER_DISTANCE" : 150,
        "MIN_ANGLE" : 1
    },
    "filters" : {
        "LENGTH" : 0,
        "FORWARD" : 1,
        "BOTTOM" : 2
    }
}

MIN_LENGTH is the minimum length of the line to be detected.
MAX_VERTICAL_POSITION corresponds to the maximum position the bottom end of the line can appear.
MAX_CENTER_POSITION defines distance that separates the line from the center.

  pip install -r requirements.txt
  bash init.sh
  python3 detection.py