Python Deep Learning Projects

Post-processing of a mask

Once we are able to create our mask successfully, we need to perform some morphological operations, which are basic image processing operations used for the analysis and processing of geometrical structures. 

First, we will create a kernel that will perform various morphological operations over the input image:

kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (15, 15))

Closing: Dilation followed by erosion is helpful to close small pieces inside the foreground objects or small black points on the object.

Now let's perform the close operation over the mask:

mask_closed = cv2.morphologyEx(mask, cv2.MORPH_CLOSE, kernel)

The opening operation erosion followed by dilation is used to remove noise.

Then we perform ...