Contour extraction of a moving vehicle in visible images based on image fusion

Infrared (IR) and visible images are commonly used in the detection and tracking of a moving target. A novel contour extraction scheme for a moving vehicle is proposed in this paper. Firstly, a motion segmentation technology is applied to get an initial contour. Secondly, a dynamic contour is applied to represent the initial contour and then converges to the target´s contour. Lastly, a novel image fusion is proposed, which minimizes the B-spline norm´s square of control point vector´s difference in two modal images. Image registration is not required. The experiment indicates that for visible images, average contour extraction error decreases by 58.14% after the fusion. Moreover, a fast iteration algorithm of dynamic contour based on the Newmark algorithm is devised. This image fusion is implemented only with control point vectors and is suitable for real-time processing.