中文摘要：

主要研究动态背景下的运动目标检测和跟踪问题；背景补偿差分法是一种常用的动态背景下运动目标检测算法，但检测到的目标轮廓要比其真实轮廓大，检测结果不准确且算法复杂度较高；主动轮廓模型在图像分割和目标提取过程中具有拓扑结构变化灵活性，对数值计算方案的设计更加方便、有效，据此提出一种基于改进c—V模型和卡尔曼滤波的算法，用来检测和跟踪动态背景下的运动目标；提出的算法利用c—V模型曲线演化检测和跟踪目标，使c—V模型在目标的边缘处收敛；结合卡尔曼滤波预测运动目标下一帧位置，从而实现对运动目标轮廓的跟踪；实验结果表明，该方法可以对动态背景下运动目标进行精确的检测与跟踪。

英文摘要：

The moving object detection and tracking technology under dynamic background is mainly researched. Background compensa tion difference is commonly used for moving object detection and tracking technology under dynamic background. Contour of moving object detected by background compensation difference is larger than the real contour, that the detection result is not accurate and the computational complexity is higher. Active contour model is flexible in topology, also convenient and effective in the design of numerical calculation for im age segmentation and object extraction process. An new algorithm which is based on the improved C-V and the Kalman filter model is pro- posed to detect and track the moving objects under dynamic background. The algorithm uses the improved C V model to perform the curve evolution for moving object detection and tracking, making the evolving curve approach to the edge of the object. Kalman filter is used to pre- dict the next frame of the object position, so as to achieve the tracking of the moving object. The experimental results show that the method is suitable for the detection and tracking of objects under dynamic background.