Adopting the method combines improved Pal-King fuzzy edge detection with dynamic B-spline curve

can solve the problem of outline and area which is slowly deformed and needs to predict the next stage. Firstly

it collects at least 2 clear object images during its deforming preprocess and pre-processes them with improved PalKing algorithm and performs edge detection. Secondly

a group of dynamic B-spline curves is adopted to converge to the detected edge

and obtains the area change rate through calculating the area inside the edge. Finally

predicted contour is obtained through extending

according to the normal direction of the controlled point of the final B-spline curve combined with the area change rate. The experiment proves that the predicted contour has a high similarity with the actual contour

and the prediction algorithm has the advantages of fast calculation speed and simple implementation.