一种利用Gabor小波特征的目标跟踪方法

小波特征跟踪是一种新的目标跟踪方法。这种方法利用Gabor小波提取目标的边界特征来组成特征模板,并由小波滤波器系数组成特征向量。在跟踪过程中不断优化特征模板的旋转、尺度及平移等几何变形参数,使得小波特征向量值在最小平方和的意义上与初始值相匹配,从而实现目标旋转、尺度及平移等几何变形时的自动跟踪。实验表明,该方法能够有效地跟踪飞机、车辆等目标,并且能够抵抗全局亮度的变化和局部遮挡的干扰。

Target tracking method using wavelet feature

Wavelet feature tracking is a new target tracking method. In this method, Gabor wavelets are used to represent target contour features. Feature models are composed of these wavelets, the feature vector is composed of filter response. During tracking, optimizing the rotation, scale and translation parameters of the feature models, the feature vector match its original values is made in a manner of sum of squared differences. Consequently, automatically tracking of target's orientation, scale and translation is realized. Experiments show that this method can track target effectively, such as plane, vehicle, etc.