摘要
在基于图像识别的立靶密集度测试系统中,为了识别连发射击后的弹孔位置,提出了一种基于投影原理的多个弹孔识别方法。采用投影原理的矩形分割法将含有弹孔的图像分割成若干个区域,再用基于区域填充的十字模板法识别出弹孔并计算弹孔中心位置。给出了识别算法和计算公式。实弹射击实验结果表明,该方法能够有效地计算出连发射击后的弹孔着靶坐标,测量误差不大于1.5 mm。
In order to identify the positions of the holes on paperboard produced by continuously fired bullet with a vertical target test system based on the image processing,a bullet hole recognition technique based on projection principle was put forward.The primary image was processed and divided into a number of segments by rectangular partitioning.The hole was successfully recognized and the center position of the hole was calculated using the formula described in this paper.The live fire experiment shows that this method can accurately calculate bullet coordinates and its measurement error is less than 1.5mm.
出处
《应用光学》
CAS
CSCD
北大核心
2010年第1期60-64,共5页
Journal of Applied Optics
基金
陕西省教育厅基金项目(03CJ31)
关键词
投影分割
图像减影
着靶坐标
图像处理
projection and segmentation
image subtraction
projectile impact coordinates
image processing
