中文摘要：

相机标定是影响视觉测量系统精度的重要环节。为实现对平面靶标成像清晰度合理、稳定的评价,避免CCD相机曝光参数不当造成特征点有效信息的丢失,提出了一种基于曝光补偿的图像绝对中心矩改进算法（ECACM）,研究曝光对靶标图像清晰度评价指数的影响规律。实验结果表明,相比于图像绝对中心矩算法（ACM）,在非过曝光的强度范围内,图像绝对中心矩改进算法给出的评价指数消除了曝光对评价指数的线性影响,对于相同对焦状态不同曝光下所成图像的清晰度可给出一致的评价指数;并且在过曝光的强度范围内,该算法给出的评价曲线下降更快,更明显地体现出图像有效信息的丢失程度。

英文摘要：

Camera calibration was an important part which influences the accuracy of visual measurement system.In order to get a reasonable and stable sharpness evaluation of planar target images,and to avoid loss of effective information caused by improper exposure,an improved ACM（image absolute central moment） algorithm（ECACM） was proposed based on exposure compensation.The effect of exposure on sharpness evaluation values of target images was discussed.Experiments show that,when compared with ACM,the improved algorithm eliminates the linear influence of exposure intensity on sharpness evaluation value.For the same focus state under different exposure,consistent evaluation is given by ECACM.In addition,when CCD camera is overexposed,the evaluation curve presented by ECACM decreases faster.This incline reflects loss of effective information in images more obviously.