公开(公告)号：US20100150426A1

公开(公告)日：2010-06-17

申请号：US12710844

申请日：2010-02-23

申请人： Hiroyuki ONISHI ,  Hiroshi Asai ,  Hiroshi Ogi

发明人： Hiroyuki ONISHI ,  Hiroshi Asai ,  Hiroshi Ogi

IPC分类号： G06K9/00

CPC分类号： G06T7/001 ,  G06T2207/30148

摘要： An operation part in a pattern inspection apparatus includes a defect candidate image generator for generating a binary defect candidate image representing a defect candidate area in an inspection image by comparing the inspection image with a reference image, in an inspection image masking part the inspection image is masked with the defect candidate image to obtain a masked inspection image. In a feature value calculation part, an autocorrelation feature value is obtained from the masked inspection image, and outputted to a classifying part. The classifying part comprises a classifier outputting a classification result on the basis of the autocorrelation feature value and a classifier construction part for constructing the classifier by learning. It is thereby possible to easily perform the high accurate classification of defect candidate using the autocorrelation feature value which is hard to characterize as compared with geometric feature value or feature value representing density.

摘要翻译： 图案检查装置中的操作部分包括缺陷候选图像生成器，用于通过将检查图像与参考图像进行比较来生成表示检查图像中的缺陷候选区域的二值缺陷候选图像，在检查图像掩蔽部中，检查图像 用缺陷候选图像掩蔽以获得掩蔽的检查图像。 在特征值计算部中，从掩蔽检查图像获得自相关特征值，并输出到分类部。 分类部分包括基于自相关特征值输出分类结果的分类器和通过学习构建分类器的分类器构造部分。 与几何特征值或表示密度的特征值相比，可以容易地使用难以表征的自相关特征值来执行缺陷候选的高精度分类。

公开(公告)号：US07689029B2

公开(公告)日：2010-03-30

申请号：US11235288

申请日：2005-09-27

申请人： Hiroyuki Onishi ,  Hiroshi Asai ,  Hiroshi Ogi

发明人： Hiroyuki Onishi ,  Hiroshi Asai ,  Hiroshi Ogi

IPC分类号： G06K9/00

CPC分类号： G06T7/001 ,  G06T2207/30148

摘要： An operation part in a pattern inspection apparatus includes a defect candidate image generator for generating a binary defect candidate image representing a defect candidate area in an inspection image by comparing the inspection image with a reference image, in an inspection image masking part the inspection image is masked with the defect candidate image to obtain a masked inspection image. In a feature value calculation part, an autocorrelation feature value is obtained from the masked inspection image, and outputted to a classifying part. The classifying part comprises a classifier outputting a classification result on the basis of the autocorrelation feature value and a classifier construction part for constructing the classifier by learning. It is thereby possible to easily perform the high accurate classification of defect candidate using the autocorrelation feature value which is hard to characterize as compared with geometric feature value or feature value representing density.

摘要翻译： 图案检查装置中的操作部分包括缺陷候选图像生成器，用于通过将检查图像与参考图像进行比较来生成表示检查图像中的缺陷候选区域的二值缺陷候选图像，在检查图像掩蔽部中，检查图像 用缺陷候选图像掩蔽以获得掩蔽的检查图像。 在特征值计算部中，从掩蔽检查图像获得自相关特征值，并输出到分类部。 分类部分包括基于自相关特征值输出分类结果的分类器和通过学习构建分类器的分类器构造部分。 与几何特征值或表示密度的特征值相比，可以容易地使用难以表征的自相关特征值来执行缺陷候选的高精度分类。

公开(公告)号：US20060067570A1

公开(公告)日：2006-03-30

申请号：US11235288

申请日：2005-09-27

申请人： Hiroyuki Onishi ,  Hiroshi Asai ,  Hiroshi Ogi

发明人： Hiroyuki Onishi ,  Hiroshi Asai ,  Hiroshi Ogi

IPC分类号： G06K9/00

CPC分类号： G06T7/001 ,  G06T2207/30148

摘要： An operation part (50) in a pattern inspection apparatus (1) comprises a defect candidate image generator (511) for generating a binary defect candidate image representing a defect candidate area in an inspection image by comparing the inspection image with a reference image, in an inspection image masking part (521) the inspection image is masked with the defect candidate image to obtain a masked inspection image. In a feature value calculation part (531), an autocorrelation feature value is obtained from the masked inspection image, and outputted to a classifying part (54). The classifying part (54) comprises a classifier (541) outputting a classification result on the basis of the autocorrelation feature value and a classifier construction part (542) for constructing the classifier (541) by learning. It is thereby possible to easily perform the high accurate classification of defect candidate using the autocorrelation feature value which is hard to characterize as compared with geometric feature value or feature value representing density.

摘要翻译： 图案检查装置（1）中的操作部（50）包括缺陷候补图像生成部（511），用于通过将检查图像与参照图像进行比较来生成表示检查图像中的缺陷候补区域的二值缺陷候补图像， 检查图像掩蔽部分（521），用缺陷候选图像掩蔽检查图像，以获得掩蔽的检查图像。 在特征值计算部（531）中，从掩蔽检查图像获得自相关特征值，并输出到分类部（54）。 分类部分（54）包括基于自相关特征值输出分类结果的分类器（541）和用于通过学习构建分类器（541）的分类器构造部分（542）。 与几何特征值或表示密度的特征值相比，可以容易地使用难以表征的自相关特征值来执行缺陷候选的高精度分类。

公开(公告)号：US20060133660A1

公开(公告)日：2006-06-22

申请号：US11289708

申请日：2005-11-30

申请人： Hiroshi Ogi ,  Hiroyuki Onishi ,  Yasushi Sasa ,  Yuichiro Hikida

发明人： Hiroshi Ogi ,  Hiroyuki Onishi ,  Yasushi Sasa ,  Yuichiro Hikida

IPC分类号： G06K9/00

CPC分类号： G06T7/001 ,  G06T2207/30148

摘要： In a defect detection apparatus 1, in a reference image inspection circuit 42 compared are a reference image representing a pattern in a die which is determined as a reference on a substrate 9 and a plurality of supervisory images which represent patterns in selected block areas, respectively, to detect defects included in the reference image. Subsequently, in the target image inspection circuit 44, a target image representing a pattern in another die and the reference image are compared to detect a plurality of defect candidates included in the target image. Then, a defect detector 45 excludes defect candidates overlapping with the defects included in the reference image from the plurality of defect candidates on the basis of at least positional information of the defects included in the reference image. This makes it possible to detect defects existing in the pattern in another die accurately while eliminating effects of the defects existing in the pattern in the die which is determined as the reference.

摘要翻译： 在缺陷检测装置1中，在比较的参照图像检查电路42中，分别表示作为基板9的基准的芯片中的图案的参考图像和分别表示所选块区域中的图案的多个监视图像 以检测参考图像中包含的缺陷。 随后，在目标图像检查电路44中，比较表示另一管芯中的图案的目标图像和参考图像，以检测包括在目标图像中的多个缺陷候选。 然后，缺陷检测器45至少基于包含在参考图像中的缺陷的位置信息，排除与来自多个缺陷候补的参考图像中包含的缺陷重叠的缺陷候选。 这使得可以精确地检测存在于另一个模具中的图案中的缺陷，同时消除存在于作为基准的芯片中的图案中存在的缺陷的影响。

公开(公告)号：US09097910B2

公开(公告)日：2015-08-04

申请号：US13557417

申请日：2012-07-25

申请人： Hiroshi Ogi ,  Satoshi Takahashi ,  Yoshiki Umemoto

发明人： Hiroshi Ogi ,  Satoshi Takahashi ,  Yoshiki Umemoto

IPC分类号： H04N7/18 ,  G02B21/36 ,  G01N21/64 ,  G02B21/16

CPC分类号： G02B21/365 ,  G01N21/6428 ,  G02B21/16 ,  H04N7/18

摘要： As light is allowed to impinge from above a microplate M and an imaging unit 13 (a line sensor 131 and an imaging optical system 132), which moves in a scanning motion along the bottom surface of the microplate M, receives transmitted light, images of wells W formed in the microplate M are captured. The scope of imaging by the line sensor 131 is greater than the diameter of at least one well W, or preferably, encompasses a plurality of wells. When the depth of field of the optical system 132 is 0.6 mm or less, the influence of reflection of side walls of the wells upon the images is reduced.

摘要翻译： 当允许光从微孔板M和沿微孔板M的底表面沿扫描运动移动的成像单元13（线传感器131和成像光学系统132）从上方入射时，接收透射光， 捕获在微板M中形成的阱W。 线传感器131的成像范围大于至少一个阱W的直径，或者优选包含多个孔。 当光学系统132的景深为0.6mm或更小时，井的侧壁反射对图像的影响减小。

公开(公告)号：US08923596B2

公开(公告)日：2014-12-30

申请号：US13495627

申请日：2012-06-13

申请人： Hiroshi Ogi

发明人： Hiroshi Ogi

IPC分类号： G06K9/00 ,  G06K9/34

CPC分类号： G06K9/0014 ,  G06K9/342

摘要： Light is allowed to be incident from above wells provided on a microplate M and the light transmitted to the lower surface is received to obtain an original image of the wells (Step S101). Detection target areas in the original image are specified by an appropriate image processing (Step S102), and peripheral areas as backgrounds surrounding the respective detection target areas are specified (Step S103). By calculating a density value of the detection target area Ri using luminance information of the detection target area Ri and that of the peripheral area Si surrounding this detection target area Ri for each detection target area Ri (Steps S105, S106), the influence of a well wall surface reflected on the background is eliminated.

摘要翻译： 允许光从微板M上提供的上述孔入射，并且接收透射到下表面的光以获得孔的原始图像（步骤S101）。 通过适当的图像处理来指定原始图像中的检测对象区域（步骤S102），并且指定作为各检测对象区域的背景的周边区域（步骤S103）。 通过使用检测对象区域Ri的亮度信息和围绕该检测对象区域Ri的周边区域Si的每个检测对象区域Ri计算检测对象区域Ri的浓度值（步骤S105，S106），影响a 消除了背景反射的井壁表面。

公开(公告)号：US20140320513A1

公开(公告)日：2014-10-30

申请号：US14366544

申请日：2012-05-29

申请人： Hiroshi Ogi

发明人： Hiroshi Ogi

IPC分类号： G06T7/00 ,  G06K9/00 ,  G09G5/30

CPC分类号： G06T7/0012 ,  G02B21/34 ,  G06K9/0014 ,  G06T2200/24 ,  G06T2207/10056 ,  G06T2207/30024 ,  G06T2207/30072 ,  G09G5/30

摘要： In a technique for displaying images of recess portions formed on a sample holding plate, an image display apparatus or an image display method is provided which has a displaying function which makes it possible for a user to compare and observe the images from various perspectives. From a total image obtained by shooting a micro plate as a whole, areas corresponding to wells are cut out. Image processing, such as color coding and contour enhancement, is performed on images of the wells based on classification results obtained using a classification criterion designated by the user, thereby forming material images. The material images are arranged in such an arrangement in accordance with the sequence on the micro plate, thereby forming an image for displaying. Classification results obtained using different classification criteria are displayed side by side, which makes it easy to compare and observe from different perspectives.

摘要翻译： 在用于显示形成在样本保持板上的凹部的图像的技术中，提供了具有显示功能的图像显示装置或图像显示方法，其使得用户可以从各个角度来比较和观察图像。 从通过拍摄微板作为整体获得的总图像中，切出与孔相对应的区域。 基于使用由用户指定的分类标准获得的分类结果，对孔的图像执行诸如颜色编码和轮廓增强的图像处理，从而形成材料图像。 材料图像按照微板上的顺序排列成这样的布置，从而形成用于显示的图像。 使用不同分类标准获得的分类结果并排显示，从而可以从不同的角度进行比较和观察。

公开(公告)号：US07409081B2

公开(公告)日：2008-08-05

申请号：US10391802

申请日：2003-03-20

申请人： Hiroshi Ogi

发明人： Hiroshi Ogi

IPC分类号： G06K9/00 ,  G06F3/048

CPC分类号： H01L21/67288 ,  G06T7/0004 ,  G06T2207/30141 ,  G06T2207/30148

摘要： An inspection system 1 includes an image pickup apparatus 2 for picking up an image of a defect, an inspection and classification apparatus 4 for performing inspection and automatic classification of defects, and a host computer 5. The host computer 5 performs learning for automatic classification at the inspection and classification apparatus 4. For creation of training data to be used for learning, defect images are arranged on a display of the host computer 5 on the basis of sizes of defects or imaging positions for picking up images of defects. A visual sign is provided to the defect image indicating a category assigned thereto. Further, in response to an operation by an operator, a statistical value of feature values of defect images included in a category, data obtained in inspection, images after being subjected to image processing, similar images or dissimilar images similar to or dissimilar to a defect image targeted for classification, an area directed to calculation of feature values in a defect image targeted for classification, and the like, are suitably displayed.

摘要翻译： 检查系统1包括用于拾取缺陷图像的图像拾取装置2，用于执行缺陷的检查和自动分类的检查和分类装置4以及主计算机5。 主机5在检查分类装置4中进行自动分类的学习。 为了创建要用于学习的训练数据，基于用于拾取缺陷图像的缺陷或成像位置的大小，将缺陷图像布置在主计算机5的显示器上。 向指示分配给其的类别的缺陷图像提供视觉符号。 此外，响应于操作者的操作，包括在类别中的缺陷图像的特征值的统计值，在检查中获得的数据，经过图像处理后的图像，类似于或不同于缺陷的类似图像或不同的图像 适当地显示针对分类的图像，针对用于分类的缺陷图像中的特征值的计算的区域等。