I am developing an algorithm in Python, which is supposed to identify the area of a leaf that contains spots to report the severity of a disease.
Therefore to be able to achieve the goal, I need to segment the image on foreground (leaf) and background.
During my research, I found out about LeafSnap (State of the Art) and follow the paper to segment the leaf on the image using OpenCV Expectation Maximization, which is trained using S and V form HSV color space; however, it still returns some false positives due to reflection or shadow.
So, I'm trying to figure out a way to avoid or decrease the incidence of false positives. Any hint on it?