自动选取辅助目标的建筑物目标间接定位方法

针对序列图像中目标因不显著、被遮挡等情况而出现的虚警或丢失问题，提出一种基于自动选取辅助目标的建筑物目标间接定位方法，提高目标跟踪稳定性。该方法首先自动选取红外图像中形状稳定、灰度显著的区域作为辅助目标。定位建筑物目标后，提取其与目标相对位置关系, 之后利用该相对位置间接定位，最后融合直接识别及间接定位结果定位目标，并能实时更新辅助目标，保证其时刻处于视场中。实验结果表明:通过间接定位能解决目标虚警或丢失问题，提升目标定位过程中算法的准确性及鲁棒性。

Indirect location method for building target based on automatic selection of auxiliary target

Aiming at the false alarm or missing problems due to the insignificant and obscured targets in sequence images, an indirect localization method for building targets based on automatic selection of auxiliary targets was proposed to improve the stability of target tracking. In this method, firstly, the region with stable shape and significant gray level in the infrared image was automatically selected as the auxiliary target. After locating the building target, its relative positional relationship with the target was extracted, and then, the relative position was used to indirectly locate the target. Finally, the target was located by fusing the result of direct target recognition and the result of indirect target location, which could update the auxiliary target timely to ensure that it is always in the field of view. The experimental result shows that the false alarm or missing problems can be solved by indirect location, and it can improve the accuracy and robustness of the algorithm during the process of target location.