为了实现对视度和视差的客观检测，提高自动化检测程度，设计了一种基于CCD图像处理的视度和视差自动检测系统。对CCD视度和视差检测原理公式进行了推导，通过分析发现需要测量成像最清晰时CCD光敏面的位置。为此，构建了像面可调的CCD摄像系统和图像采集与处理系统，采用扫描求解图像像素灰度值梯度的方法，实现图像清晰度测量和边缘提取；同时利用视频图像清晰度峰值搜索法，通过控制电机带动CCD光敏面朝图像最大清晰位置运动，实现CCD摄像系统的自动对焦，并完成对成像最清晰时CCD光敏面位置的测量。最后通过对装备视度和视差实际检测，表明视度检测不确定度在0.15 D以内，视差不确定度在0.3分以内，与传统方法相比，检测结果客观准确，自动化程度得到了提高。

In order to achieve objective measurement of diopter and optical parallax and improve the level of automation,an automatic measurement system for diopter and optical parallax was designed based on CCD image processing.The measurement formula was deduced and the analysis showed that the location of CCD sensor needed to be tested when the CCD imaging was the clearest.A CCD camera system whose image plane was adjustable,and an image acquisition and processing system were built up.It achieved the test of image clarity and edge detection with the solution of scanning gray value gradient of image pixel.Under the control of the system,the motor drove the CCD sensor to the biggest definition position,by search of the video image peak to achieve the CCD camera system automatic focusing.Finally,experimental tests on the equipment showed that the uncertainty of diopter was less than 0.15 D,and optical parallax less than 0.3 arcminutes.Compared with the traditional methods,the automation of the measurement introduced in this paper was greatly improved,and the results are more objective and accurate.