远程定位瞄准系统的设计和实现

本文针对远程目标的定位瞄准问题,开发了一套以人脸为基点对远程目标进行定位瞄准的自动化系统。将基于Adaboost方法的肤色检测器和人脸检测器应用于系统中,提高定位、瞄准的效率。首先用肤色检测器对被检测视频进行预处理,将肤色区域从背景中分割出来,然后利用人脸检测器对肤色区域进行检测,定位并瞄准人脸目标。实验表明,对正面人脸目标和具有小角度偏转的人脸目标具有较好的检测效果,在300米左右的距离上可以准确地定位、瞄准目标,同时完全满足实时性要求。实际应用中可以减少人工定位、瞄准目标的复杂性,可以用于战场以及解救人质等紧急情况的辅助瞄准。

Design and Implementation of a Long-Range Target Localization and Collimation System

A long range targets automatic localization and collimation system which takes the human face as the anchor point has been developed. In order to improve the performance,the skin tone detector and the face detector based on AdaBoost are used in our system. First,the skin tone detector checks the available video to separate the skin area from others,then the skin tone area is processed with the AdaBoost face detection method to implement real time localization and collimation. According to the experience,for the front faces and the faces with small angles,the system has a good result. Within the distance of 300m,it localizes and collimates the targets precisely and satisfies real time requirement. The system is able to reduce the complexity and increase the precision of localization and collimation,and is suitable for the front line of a battle and saving hostages and so on.