解决检测目标的定位问题是实施自动光学检测的前期工作.针对挠性印制电路基材制成品与标准设计存在的差异导致的匹配误差,本文提出了基于互信息熵与关键区域特征相结合的匹配误差校正方法.首先采用图像灰度互信息熵作为衡量检测目标与设计标准之间差异的指标,建立以互信息熵值为目标函数的搜索方法；根据关键区域内的互信息熵对匹配误差具有更好的敏感性与准确性的特点,选取含灰度信息丰富的关键区域与互信息熵相结合的方法对检测目标进行匹配以提高匹配效率；然后,在焊盘图像中添加椒盐噪声、缺损、冗余等干扰信息,在存在干扰的环境下实现对目标的准确定位与搜索,验证了算法的抗干扰能力.通过最终的实验表明,采用该方法对于75 μm制程该方法定位精度可达22.4μm,优于霍夫变换法与坐标转换法.对于工艺误差、基材胀缩、外形缺损等因素造成的不良影响具有良好的抗扰性,更适于在实际的工业环境中使用.%The method of correcting position error based on mutual information entropy is proposed in this paper. The realistic product is always different from standard template since the FPC manufacturing material is easily affected by thermal expansion and contraction. Correcting matching error is crucial during manufacturing process. The mutual information entropy(MIE) is applied to evaluating the correlation between the images. The MIE increases and reach maximum when the referent and inspecting FPC solders are aligned with the same place. Moreover, The MIE in some image's region which contains more grey information can react to position error strongly. According to the above principle , the MIE and key regional feature is applied to searching best optimal correction parameters to improve positioning accuracy. The method is verified by a simulation and applied to the inspecting system. It demonstrates that the method can correct the position error more effectively compared with Hough transformation and affine transformation method. The matching accuracy can reach 22.4 micron under the condition that manufacturing error is 75 micron.
展开▼
