参数轮廓模型的目标跟踪  

Object tracking based on parameterized contour model

作　　者：孙晓云[1] 王书朋[1] 郑棋[2] 

机构地区：[1]西安科技大学通信与信息工程学院,陕西西安710054 [2]西安科技大学电气与控制工程学院,陕西西安710054

出　　处：《西安科技大学学报》2010年第1期123-126,共4页Journal of Xi’an University of Science and Technology

摘　　要：为了减少背景噪声和不相干边缘的影响,提出了一种基于方向能量的边缘选择方法,根据边缘检测的结果和由前一帧跟踪结果得到的预测轮廓构造能量函数,通过最小化能量函数进行轮廓跟踪,由于能量函数包含了预测轮廓和目标边缘信息,使得在跟踪目标轮廓变化的同时,又能保持目标形状的稳定性。实验表明:提出的跟踪方法可以在复杂背景下准确跟踪目标轮廓。A new object contour tracking algorithm for the complex background image sequences is presented. A edge selection method is proposed to reduce the effect of the background noise and irrelative edges. Contour tracking is achieved by minimizing the energy function that is constructed based on detected edges in current and the predicted contour. Since the edge of the object and the predicted contour are contained in the energy function, the robustness of contour tracking is improved. The experimental results show that the proposed algorithm can achieve accurate tracking results in complex background.

关 键 词：目标跟踪 活动轮廓模型 边缘检测 

分 类 号：TU991.2[建筑科学—市政工程]