ovl.target_filters.contour_filter module

ovl.target_filters.contour_filter.contour_filter(contour_filter_function)

A decorator function used to make a contour filter function. Contours filters are functions that take a list of contours and returns only

Wrapping a function it gives the following capabilities:

• pass parameters before the list of contours
• preserves the properties of the function: name, documentation etc.

The contours filter then changes it behaviour to act as follows:

For a given function:

@contour_filter
def area_filter(contours, min_area, max_area):
    output_list = []
    ratio_list = []

    for current_contour in contour_list:
        if min_area <= cv2.contourArea(current_contour) <= max_area:
            output_list.append(current_contour)
            ratio_list.append(current_contour)
    return output_list, ratio_list

Instead of calling the function like other functions:

some_filter(list_of_contours, 200, 5000)

The function needs to be called as follows:

activator = area_filter(200, 5000)

result = activator(list_of_contours)

Vision objects use functions that are decorated with contour_filter you can just pass the activator to the Vision object like so:

target_filters = [area(min_area=400, max_area=5000), ovl.circle_filter(min_area_ratio=0.75)]

vision = Vision(..., target_filters=target_filters, ...)