基于FPGA的晶圆级芯片封装图像序列配准方法的设计与实现*
电子技术应用
方俊杰,吴泽一,黄煜萧,任青松,王赓
上海交通大学 软件学院,上海 200240
摘要:针对未切割晶圆进行封装后的晶圆级芯片封装(WLCSP),12英寸晶圆以1 μm物理分辨率进行自动光学检测(AOI)面临大幅面、高质量成像和成像速度的技术挑战。晶圆全局图像需由多幅扫描生成的局部图像序列拼接而成,为实现图像序列的高质量、高速配准,在FPGA中采用OpenCL实现相位相关法进行四邻域棋盘配准。首先在构建二维FFT和互功率谱函数内核的基础上,采用双端口缓存和行缓存的设备全局内存对计算过程的频谱数据进行复用并应用内核通道级联提高配准速度,基于最小生成树优化配准结果降低全局图像坐标计算的累积误差,并经实际扫描图像验证配准算法及加速性能。 关键词:晶圆级芯片封装;图像配准;FPGA;OpenCL
中图分类号:TN402 文献标志码:A DOI: 10.16157/j.issn.0258-7998.234273
中文引用格式:方俊杰,吴泽一,黄煜萧,等. 基于FPGA的晶圆级芯片封装图像序列配准方法的设计与实现[J]. 电子技术应用,2023,49(12):90-97.
英文引用格式:Fang Junjie,Wu Zeyi,Huang Yuxiao,et al. Method of image sequence registration for wafer level chip scale packaging based on FPGA[J]. Application of Electronic Technique,2023,49(12):90-97.
中文引用格式:方俊杰,吴泽一,黄煜萧,等. 基于FPGA的晶圆级芯片封装图像序列配准方法的设计与实现[J]. 电子技术应用,2023,49(12):90-97.
英文引用格式:Fang Junjie,Wu Zeyi,Huang Yuxiao,et al. Method of image sequence registration for wafer level chip scale packaging based on FPGA[J]. Application of Electronic Technique,2023,49(12):90-97.
Method of image sequence registration for wafer level chip scale packaging based on FPGA
Fang Junjie,Wu Zeyi,Huang Yuxiao,Ren Qingsong,Wang Geng
School of Software, Shanghai Jiao Tong University, Shanghai 200240, China
Abstract:The Wafer Level Chip Scale Packaging (WLCSP) of chips on 12-inch wafers without dicing poses significant technological challenges for automatic optical inspection (AOI) in terms of high-quality imaging and imaging speed. To achieve high-quality and high-speed registration of the image sequence, a local image sequence generated by multiple scans needs to be stitched together to form a global image of the wafer. To this end, a phase correlation method based on the four-neighborhood checkerboard registration is implemented using OpenCL in an FPGA to address the challenge. Initially, a two-dimensional Fast Fourier Transform (FFT) and cross-power spectrum function kernel are constructed. Then, dual-port and row-buffered device global memory are employed to reuse the computed spectral data and to apply kernel channel cascading to enhance the registration speed. Finally, the registration result is optimized using a minimum spanning tree algorithm to reduce the cumulative error of global image coordinate calculation. The proposed registration algorithm and its accelerated performance are verified using actual scanned images.
Key words :wafer level chip scale package;image registration;FPGA;OpenCL
0 引言
晶圆级芯片封装(Wafer Level Chip Scale Packaging, WLCSP)是一种对尚未切割的整片晶圆进行封装的先进封装技术,该技术生产工序中仅切割合格的芯片颗粒到后续制造工序中[1]。工业制造中使用自动光学检测(Automatic Optical Inspection, AOI)技术对晶圆级芯片封装的整片晶圆表面进行缺陷检测,过程中需获取整片晶圆表面图像,需要高效的图像生成与拼接技术支持。
针对12英寸WLCSP的整片晶圆,以1 m物理分辨率精度获取晶圆表面图像,其图像具有幅面大、精度高的特点,整体像素数量可达1011,存在图像生成空间占用大、图像序列邻接关系复杂、整体算法计算量大的问题与挑战。
扫描生成的局部图像拼接前需进行配准以确定各相邻图像间的位置关系,图像配准主要有基于频域、基于灰度图像及基于特征匹配的方法。对于图像序列,一维或二维排列的场景均存在,上交大潘昕对机器人移动采集的大视差钢卷一维图像序列进行带尺度约束的特征匹配并拼接得到钢卷仓库全局图像[2];德国英飞凌Singla等人对扫描电子显微镜采集的晶圆二维图像序列结合三种配准方法进行局部配准,并利用最大似然估计的方法在拼接过程中最小化全局误差,从而得到纳米尺度晶圆表面结构几何布局[3]。
本文详细内容请下载:https://www.chinaaet.com/resource/share/2000005812
作者信息:
方俊杰,吴泽一,黄煜萧,任青松,王赓
(上海交通大学 软件学院,上海 200240)
此内容为AET网站原创,未经授权禁止转载。