无铅焊点鲁棒定位的灰度积分投影算法

针对自动光学检测系统因焊点定位不准确而导致的误判问题，基于灰度积分投影技术提出了一种焊点定位的鲁棒算法。首先对焊点图像进行预处理，利用焊点的颜色特征对图像进行二值化，将焊点从PCB图像中分割出来，然后分别利用焊点特征的水平和垂直灰度积分投影曲线，以焊盘窗内焊点像素的面积最大化为目标，找到焊点的定位坐标，以实现准确的焊点定位。另外，在算法中，通过引入Blob评价函数区分焊点与噪声，从而有效地减少噪声的干扰，提高了定位算法的鲁棒性。实验结果表明，对变化多样的Chip元件焊点，该算法具有较高的定位精度。

A Robust Positioning Algorithm of the Lead-Free Solder Joint

Considering the automated optical inspection system’s misjudgement due to positioning error of the solder joints, a robust positioning algorithm of the solder joints is presented based on the gray-level integration projection. Firstly, the solder joint is extracted from the PCB (print circuit board) image based on image segmentation and binary. Then, with the help of solder joint feature’s gray-level integration projection curve with respect to horizontal and vertical direction, the accurate solder joint location can be determined by the maximum area of solder joint pixel within the solder-lands windows. In addition, a Blob evaluation function is defined in this algorithm in order to differentiate the information of the solder joint and the noise, which can reduce the disturbance of noise and improve the robust of the algorithm. Experimental results show that the proposed method is valid for various solder joints of chip component, and obtains a relative high positioning precision.