An image segmentation algorithm for segmentation-based video sequence
coding is presented. The method uses a new criterion based on similari
ties in both motion and brightness. Brightness and motion information
are incorporated in a single segmentation procedure. The actual segmen
tation is accomplished using a region-growing technique based on the w
atershed algorithm. It comprises the processes of extracting the initi
al regions called seeds and their growing by merging similar neighbori
ng pixels. In addition, a tracking technique is used in subsequent fra
mes to achieve a coherent segmentation through time. Simulation result
s show that the proposed method is effective in determining object bou
ndaries not easily found using the static criterion alone. (C) 1997 So
ciety ni Photo-Optical Instrumentation Engineers.