实现微小芯片的凸点和基板天线的互连是射频识别( RFID)电子封装设备的关键技术,其中芯片上凸点的定位是精密操作中的难点之一。研究RFID标签封装设备中芯片上凸点的视觉定位,提出一种基于改进的SUSAN ( Smallest Univalue Segment Assimilating Nucleus )角点检测的算法。用改进的SUSAN角点检测算法提取凸点特征,根据凸点特征计算出凸点的中心坐标,以此得到凸点的位置信息,实现芯片上凸点的精确定位。目前提出的算法已成功应用到RFID封装设备HEI-DIII上,实验结果表明,与常规的特征匹配算法相比,所提出的凸点识别的RFID微小芯片定位算法耗时短,定位精度较高,对扰动和芯片的大角度旋转不敏感。
The realization of combination the chip and antenna is the key technologies of the electronic packaging devices ,which the visual location is one of the difficulty in the whole precision operation .Research the visual location algorithm for the bumps recongnition in RFID packaging equipment and proposal a new visual location algorithm based on modified SUSAN corner detec -tion algorithm.The algorithm uses the modified SUSAN corner detection algorithm to extract the feature of bump and calculate center coordinates of the bump ,in order to get the location information of the bump .The algorithm has been tested and applied in the RFID package equipment .Experiment results show that compared with conventional algorithms ,the new algorithm achieves in-tended purpose and promise application values with its short consuming-time and high prediction accuracy ,and not sensitive to disturbance and large angle rotation of the chip .