基于均值漂移算法的冷空炽热体红外目标跟踪

Tracking of Fiery Infrared Target in Cool Space Based on Mean Shift Algorithm

Mean Shift（均值漂移）算法对彩色序列图像目标跟踪是有效的,而对红外目标跟踪是不稳健的。本文对Mean Shift算法做适当改进,将其用于冷空炽热体红外目标跟踪。先对红外图像做伽马变换,把目标从背景中分离出来;用直方图非均匀量化改进传统Mean Shift算法量化方式,提高目标定位准确性。仿真结果表明,该方法对冷空炽热体红外目标跟踪是有效的和稳健的。

Mean Shift algorithm is often effectively used to color target tracking,but it lacks robustness when used for infrared targets.This paper presents an improved Mean Shift algorithm used in single fiery infrared target tracking in cool space.The interested target is separated from its background by gamma transform.Uniformity quantization is performed on candidate regions to improve the accuracy of target detection.The tracking results show that this approach is effective and robust for fiery infrared targets in cool space.