一种新的红外目标跟踪方法

A Novel Infrared Object Tracking Algorithm

提出了一种基于灰度加权和Kalman滤波预测的红外目标跟踪方法。利用带有空间位置信息的目标区域直方图描述红外目标,克服了灰度直方图对图像描述缺少空间位置信息的缺陷,增加了描述的鲁棒性。同时,对跟踪目标的实时位置数据利用卡尔曼滤波校正和预测,增强了跟踪的实时性。实验结果表明,该方法是有效和稳健的。

A novel infrared object tracking method is proposed based on combined average gray using Kalman filter. The infrared object is represented by the object region histogram with space location information, which is better than only by grayscale histogram.Meanwhile, the position data of the tracking object is adjusted and predicted by Kalman filter. Experimental results show that the proposed method is efficient and robust.