相位一致性图像及其在目标跟踪中的应用(英文)

Phase Congruency Image and its Application in Target Tracking

针对传统实时相关跟踪方法对照度变化敏感的问题,提出了一种基于相位一致性图像的相关跟踪方法.利用相位一致性函数值在[0,1]区间内且无量纲、对图像的亮度和对比度具有不变性等特点,首先对原始图像进行相位一致性检测,得到相位一致性图像,再利用MAD(Minimum Absolute Difference)等相关跟踪算法在相位一致性图像中对目标进行跟踪运算.对可见光和红外图像的实验表明,在图像的亮度和对比度发生剧烈变化的情况下,算法仍能保持对目标的稳定跟踪.该方法可用于解决传统实时相关跟踪方法普遍存在的因照度变化导致跟踪点漂移甚至跟踪失败的问题.

To deal with the problem that traditional real-time correlation tracking methods are sensitive to the light variation, a Phase Congruency（PC） image based on tracking method is proposed which utilizes the properties of PC function falling in [-0, 1] interval, being normalized and invariant to changes of image brightness and constrast. PC detection is performed on the original image, and PC image is obtained. Subsequently, target tracking with correlation tracking methods, such as Minimum Absolute Difference （MAD） is conducted on the PC image. Experiment results show that the proposed method can achieve a satisfactory tracking performance on visible light images and infrared images even though the image brightness and constrast vary abruptly. The phase congruency image based on tracking method can be used to solve the common problems of tracking point drifting, even failed tracking due to image light variation in traditional real-time correlation tracking methods.