Feature analysis for shadow detection

In minimally invasive surgery (MIS), depth estimation is also one of the most essential challenges. With the urgent demands of improving surgical safety, a method of improving the surgeon's perception of depth by introducing an "invisible shadow" in the operative field has been proposed. A novel approach is presented for adaptive shadow detection by incorporating four different attributes and using a classification algorithm to make a distinction between shadow and non-shadow regions. Experimental results show that shadow region can be detected quite well and thus the method could potentially be used as an instrument navigation aid in minimally invasive surgery.