公开(公告)号：US08669523B2

公开(公告)日：2014-03-11

申请号：US13410506

申请日：2012-03-02

申请人： Chien-Huei Chen ,  Peter White ,  Michael J. Van Riet ,  Sankar Venkataraman ,  Hai Jiang ,  Hedong Yang ,  Ajay Gupta

发明人： Chien-Huei Chen ,  Peter White ,  Michael J. Van Riet ,  Sankar Venkataraman ,  Hai Jiang ,  Hedong Yang ,  Ajay Gupta

IPC分类号： H01J37/28 ,  G01N23/04 ,  G06T5/50

CPC分类号： H01J37/28 ,  G01N23/04 ,  G06T5/50 ,  G06T7/001 ,  G06T2207/30148 ,  H01J37/05 ,  H01J37/222 ,  H01J37/244

摘要： One embodiment relates to a method of inspecting a site location on a target substrate. Contours are obtained, the contours having been generated from a reference image using a design clip. A target image of the site location is acquired. The contours are aligned to the target image, and contrast values are computed for pixels on the contours. A threshold is applied to the contrast values to determine contour-based defect blobs. Another embodiment relates to a method of generating contours for use in inspecting a site location for defects. Other embodiments, aspects and features are also disclosed.

摘要翻译： 一个实施例涉及一种检查目标基板上的位置位置的方法。 获得轮廓，使用设计剪辑从参考图像生成轮廓。 获取站点位置的目标图像。 轮廓与目标图像对齐，并且对轮廓线上的像素计算对比度值。 将阈值应用于对比度值以确定基于轮廓的缺陷斑点。 另一个实施例涉及一种产生用于检查位置缺陷的轮廓的方法。 还公开了其它实施例，方面和特征。

公开(公告)号：US08502146B2

公开(公告)日：2013-08-06

申请号：US13252042

申请日：2011-10-03

申请人： Chien-Huei Chen ,  Hedong Yang ,  Cho H. Teh

发明人： Chien-Huei Chen ,  Hedong Yang ,  Cho H. Teh

IPC分类号： G01N23/00

CPC分类号： H01L22/12 ,  G06T7/0006 ,  G06T2207/10012 ,  G06T2207/10028 ,  G06T2207/10061 ,  G06T2207/30148 ,  H01L21/67288

摘要： One embodiment relates to a method of classifying a defect on a substrate surface. The method includes scanning a primary electron beam over a target region of the substrate surface causing secondary electrons to be emitted therefrom, wherein the target region includes the defect. The secondary electrons are detected from the target region using a plurality of at least two off-axis sensors so as to generate a plurality of image frames of the target region, each image frame of the target region including data from a different off-axis sensor. The plurality of image data frames are processed to generate a surface height map of the target region, and surface height attributes are determined for the defect. The surface height attributes for the defect are input into a defect classifier. Other embodiments, aspects and features are also disclosed.

摘要翻译： 一个实施例涉及对衬底表面上的缺陷进行分类的方法。 该方法包括在衬底表面的目标区域上扫描一次电子束，导致二次电子从其发射，其中目标区域包括缺陷。 使用多个至少两个离轴传感器从目标区域检测二次电子，以便产生目标区域的多个图像帧，目标区域的每个图像帧包括来自不同离轴传感器的数据 。 处理多个图像数据帧以生成目标区域的表面高度图，并且为缺陷确定表面高度属性。 缺陷的表面高度属性被输入到缺陷分类器中。 还公开了其它实施例，方面和特征。

公开(公告)号：US07440607B1

公开(公告)日：2008-10-21

申请号：US10980499

申请日：2004-11-03

申请人： Jason Z. Lin ,  Hong Chen ,  Evgeni Shifrin ,  Ashok V. Kulkarni ,  Santosh K. Bhattacharyya ,  Wei Zhao ,  Chien-Huei Chen

发明人： Jason Z. Lin ,  Hong Chen ,  Evgeni Shifrin ,  Ashok V. Kulkarni ,  Santosh K. Bhattacharyya ,  Wei Zhao ,  Chien-Huei Chen

IPC分类号： G06K9/00

CPC分类号： G01N21/8851 ,  G01N21/95607 ,  G01N2021/8825 ,  G01N2021/8887 ,  G06T7/0004

摘要： A method of detecting anomalies in a test image. Test features of pixels within the test image are selected, and reference features of pixels within at least one reference image are also selected. A signal distribution of test features and reference features in a multi-dimensional feature space is created, and stored. Those test features of the test image that do not satisfy a set of criteria for normalcy are selected as candidate points. Those candidate points that are statistical outliers are identified as anomalies. Positions of the anomalies are located using the stored signal distribution within which the defects have been identified as a lookup table.

摘要翻译： 一种检测测试图像异常的方法。 选择测试图像内的像素的测试特征，并且还选择至少一个参考图像内的像素的参考特征。 创建并存储多维特征空间中的测试特征和参考特征的信号分布。 选择不满足正常条件的测试图像的测试特征作为候选点。 统计异常值的候选点被识别为异常。 使用存储的信号分布来定位异常的位置，在该信号分布中已经将缺陷识别为查找表。

公开(公告)号：US06985220B1

公开(公告)日：2006-01-10

申请号：US10644319

申请日：2003-08-20

申请人： Chien-Huei Chen ,  Vivek Bhagat ,  Qiang Song ,  James A. Quigley ,  Ashok V. Kulkarni

发明人： Chien-Huei Chen ,  Vivek Bhagat ,  Qiang Song ,  James A. Quigley ,  Ashok V. Kulkarni

IPC分类号： G01N21/88

CPC分类号： G05B19/41875 ,  G01N21/8851 ,  G01N2021/8874 ,  G05B23/0235 ,  G05B2219/32184 ,  G05B2219/32222 ,  G05B2219/45031 ,  Y02P90/20 ,  Y02P90/22

摘要： A method of tuning an inspection system. An inspection piece is sensed and analyzed to identify anomalies. Level information is analyzed with an initial set of thresholds, and an initial portion of the anomalies are flagging as defects. The inspection system parameters are changed, and the level information is analyzed with a modified set of thresholds. The anomalies are flagged as defects based on the immediately preceding analysis of the level information. The steps of changing the thresholds and reflagging the defects are repeated as desired, and the modified set of thresholds are stored for use in an inspection system recipe.

摘要翻译： 调整检查系统的方法。 检测和分析检查件以识别异常。 使用初始阈值分析级别信息，异常的初始部分标记为缺陷。 检查系统参数发生变化，水平信息用修改后的阈值进行分析。 根据上一级信息分析，将异常标记为缺陷。 根据需要重复改变阈值和重新填充缺陷的步骤，并且修改的阈值集合被存储以用于检查系统配方。