公开(公告)号：US6038335A

公开(公告)日：2000-03-14

申请号：US901058

申请日：1997-07-25

申请人： Haruhiko Yokoyama ,  Akira Kobayashi

发明人： Haruhiko Yokoyama ,  Akira Kobayashi

IPC分类号： G06T7/00 ,  G06K9/00

CPC分类号： G06T7/0004 ,  G06T7/0083 ,  G06T7/0085 ,  G06T7/602 ,  G06T2207/30164

摘要： A flaw detection apparatus detects a flaw formed on an object's surface based on a digital image data indicative of said object's surface as regions which are groups of plural pixels utilizing the image processing technique. An image processor is provided for detecting an extending direction in which each region extends based on a contour line of each region, so that regions extending in the same direction are clustered as one region. The image processor further produces a flaw signal indicating sizes of thus clustered regions.

摘要翻译： 探伤装置利用图像处理技术，将表示物体表面的数字图像数据作为多个像素组的区域，检测形成在物体表面上的缺陷。 提供了图像处理器，用于检测每个区域基于每个区域的轮廓线而延伸的延伸方向，从而沿相同方向延伸的区域被聚集成一个区域。 图像处理器还产生指示如此聚簇区域的大小的缺陷信号。

公开(公告)号：US5850468A

公开(公告)日：1998-12-15

申请号：US8211

申请日：1998-01-16

申请人： Haruhiko Yokoyama ,  Akira Kobayashi

发明人： Haruhiko Yokoyama ,  Akira Kobayashi

IPC分类号： G06T7/00 ,  G06T7/60 ,  G06K9/00

CPC分类号： G06T7/602 ,  G06T7/0006 ,  G06T7/606 ,  G06T2207/30164

摘要： A flaw detection apparatus detects a flaw formed on an object's surface based on a digital image data indicative of said object's surface as a group of plural pixels utilizing the image processing technique. An image processor is provided for detecting a direction in which each of group extends to combine a plurality of groups extending in substantially the same direction. The image processor further produces a flaw signal indicating sizes of thus combined groups.

摘要翻译： 利用图像处理技术，探伤装置将表示所述物体的表面的数字图像数据作为一组多个像素，检测在物体表面上形成的缺陷。 提供了图像处理器，用于检测每个组延伸的方向以组合沿基本上相同方向延伸的多个组。 图像处理器还产生指示这样组合的组的大小的缺陷信号。

公开(公告)号：US5319720A

公开(公告)日：1994-06-07

申请号：US913190

申请日：1992-07-14

申请人： Haruhiko Yokoyama ,  Masaya Nakao

发明人： Haruhiko Yokoyama ,  Masaya Nakao

IPC分类号： G01B11/24 ,  G06T1/00 ,  G06T7/00 ,  G06K9/00

CPC分类号： G06T7/0006 ,  G06T2207/10024 ,  G06T2207/30148

摘要： An appearance inspecting method includes the steps of dividing a recognized image of an object into regions based on two attributes of image points within the image calculating the minimum distance of each point within a target region to a periphery of the target region, and measuring a size of the target region from the maximum of the calculated minimum distances of the points in the target region. A distance measuring method for measuring a distance of each point within a region to a periphery of the region includes the steps of dividing each region to be measured by lines made of picture elements arranged in one direction, sequentially integrating the distance of each point from the periphery of the region from its outer side to a middle point in each line from one end line to the other end line of the region to thereby obtain the minimum distance of each point, and the distance of each point from the periphery of the region is sequentially integrated from its outer side to the middle point in each line from the other end line to the one end line of the region to thereby obtain the minimum distance of each point.

摘要翻译： 外观检查方法包括以下步骤：基于图像内的图像点的两个属性，将目标区域中的每个点的最小距离计算到目标区域的周边，将目标的识别图像划分为区域，并且测量尺寸 从目标区域中的点的计算出的最小距离的最大值来计算目标区域。 用于测量区域内的每个点到该区域的周边的距离的距离测量方法包括以下步骤：将由沿一个方向布置的像素组成的线划分成待测量的每个区域，依次将从每个点的距离 该区域的边缘从该区域的一端线到另一端线的每条线的外侧至中点，从而获得每个点的最小距离，并且每个点与该区域的周边的距离为 从其另一端线到该区域的一端的每条线从其外侧到中点顺序地集成，从而获得每个点的最小距离。

公开(公告)号：US06240202B1

公开(公告)日：2001-05-29

申请号：US09125341

申请日：1998-08-14

申请人： Haruhiko Yokoyama ,  Masatoshi Nakamura

发明人： Haruhiko Yokoyama ,  Masatoshi Nakamura

IPC分类号： G06K948

CPC分类号： G01N21/8851

摘要： An external appearance inspection method for electronic components for detecting the presence of unevenness in the outline thereof. An approximately square-shaped outline (1) of an electronic component under inspection is extracted, this outline (1) is differentiated, and the presence of unevenness in the outline (1) is determined according to the absolute value of the difference between the maximum and minimum values (10, 11) of the differential values and the median (12) thereof.

摘要翻译： 一种用于检测轮廓中存在不均匀性的电子部件的外观检查方法。 提取正在检查的电子部件的大致四边形的轮廓（1），该轮廓（1）不同，轮廓（1）的不均匀性的存在根据最大值 和微分值的最小值（10,11）及其中值（12）。

公开(公告)号：US5570298A

公开(公告)日：1996-10-29

申请号：US249164

申请日：1994-05-25

申请人： Noriyuki Suzuki ,  Haruhiko Yokoyama

发明人： Noriyuki Suzuki ,  Haruhiko Yokoyama

IPC分类号： G06K9/46 ,  G06F17/00

CPC分类号： G06K9/4647

摘要： A dot pattern-examining apparatus examines a dot pattern displayed on a display screen by picking up a two-dimensional image of the display screen and by performing image processing with respect to the dot pattern on the two-dimensional image. An image projection data generating section generates image projection data indicative of a set of density values by accumulating, on a predetermined axis, density values of individual dots of the two-dimensional image of the display screen for each line. Then, a run-length data generating section generates run-length data indicative of a set of combinations of an accumulated density value and a length thereof, based on the image projection data of the two-dimensional image. Thereafter, a run-length data matching section compares the run-length data of the two-dimensional image with a reference run-length data indicating a reference dot pattern and generated in advance by the run-length data generating section, to thereby determine a position having run-length data which most match the reference run-length data as the position where the dot pattern to be examined exists.

摘要翻译： 点阵图案检查装置通过拾取显示屏幕的二维图像并且对二维图像上的点图案执行图像处理，来检查显示在显示屏幕上的点阵图形。 图像投影数据生成部通过在规定的轴上累积各行的显示画面的二维图像的各个点的浓度值来生成表示浓度值的图像投影数据。 然后，游程长度数据生成部基于二维图像的图像投影数据生成指示累积浓度值和其长度的组合的游程长度数据。 此后，游程长度数据匹配部分将二维图像的游程长度数据与指示参考点图案的参考游程长度数据进行比较，并由游程长度数据生成部分预先生成，从而确定 具有与参考游程长度数据最匹配的游程长度数据作为要被检查的点图案的位置的位置。

公开(公告)号：US5392364A

公开(公告)日：1995-02-21

申请号：US885837

申请日：1992-05-20

申请人： Haruhiko Yokoyama ,  Masaya Nakao

发明人： Haruhiko Yokoyama ,  Masaya Nakao

IPC分类号： G06K9/46 ,  G06K9/62 ,  G06K9/64 ,  G06T1/00 ,  G06T7/40 ,  G06K9/68

CPC分类号： G06K9/64

摘要： A method for discerning whether an object to be inspected is acceptable or not is based on feature values with respect to a binary-coded image of the object. The method includes the steps of coding image data of the object into binary digits to obtain the binary-coded image, calculating at least three feature values based on a predetermined sample group of acceptable objects and a predetermined sample group of unacceptable objects, obtaining a Mahalanobis' generalized distance between the sample groups of the acceptable objects and the unacceptable objects with respect to each of the calculated feature values, comparing each of the distances with a first predetermined value and then selecting as a first representative feature value the distance which is not smaller than the first predetermined value, obtaining a Mahalanobis' generalized distance between groups of acceptable objects and unacceptable objects with respect to the feature values except for the feature value selected as the first representative feature value and the first representative feature value, and comparing each of the distances with a second predetermined value and then selecting as a second representative feature value the distance which is not smaller than the second predetermined value, so that it is discerned whether the object is acceptable or not based on the first and/or first and second feature values with respect to the binary-coded image of the object.

摘要翻译： 用于识别被检查对象是否可接受的方法是基于关于对象的二进制编码图像的特征值。 该方法包括以下步骤：将对象的图像数据编码为二进制数字以获得二进制编码图像，基于可接受对象的预定样本组和预定的不可接受对象样本组计算至少三个特征值，获得马氏距离 相对于所计算的每个特征值，可接受对象的样本组和不可接受对象之间的广义距离，将每个距离与第一预定值进行比较，然后将不等于的距离选择为第一代表特征值 相对于除了选择为第一代表特征值和第一代表特征值的特征值之外的特征值，获得可接受物体组和不可接受物体之间的马氏距离广义距离，并且比较第一预定值， 距离与第二预定值然后 选择不小于第二预定值的距离作为第二代表特征值，使得基于相对于二进制编码的第一和/或第一和第二特征值来辨别对象是否可接受 对象的形象。

公开(公告)号：US5568567A

公开(公告)日：1996-10-22

申请号：US824173

申请日：1992-01-22

申请人： Haruhiko Yokoyama

发明人： Haruhiko Yokoyama

IPC分类号： G06T3/00 ,  G06T5/00 ,  G06T7/60 ,  G06K9/46

CPC分类号： G06T7/0083 ,  G06T7/604 ,  G06T2207/30164

摘要： Two-dimensional image data of an object is processed to inspect a circular arc configuration of the object. Initially, the two-dimensional image data is subjected to a data extraction process in which image edge points of the circular arc configuration of the object are extracted from the two-dimensional image data. The image edge points are then applied to a process to approximate the circular arc configuration of the object by defining a circle which minimizes a mean value of a fourth power error of the image edge points relative to the circle, the thus defined circle being used to approximate the circular arc configuration of the object.

摘要翻译： 处理对象的二维图像数据以检查对象的圆弧配置。 首先，对二维图像数据进行数据提取处理，其中从二维图像数据中提取对象的圆弧结构的图像边缘点。 然后将图像边缘点应用于通过限定使图像边缘点相对于圆的第四功率误差的平均值最小化的圆来近似对象的圆弧结构，由此定义的圆被用于 近似对象的圆弧形状。

公开(公告)号：US5138671A

公开(公告)日：1992-08-11

申请号：US613983

申请日：1990-11-14

申请人： Haruhiko Yokoyama

发明人： Haruhiko Yokoyama

IPC分类号： G06T1/00 ,  G06K9/38 ,  G06T5/40

CPC分类号： G06K9/38 ,  G06T7/0081 ,  G06T2207/20148

摘要： An image processing method distinguishes an object from within an image area by determining a stable threshold value of light intensity values from within the image area. Received light is converted into image data denoting light intensity values of the received light at a plurality of points from within the image area. A bright portion average value denoting an average of the light intensity values which exceed a predetermined threshold and a dark portion average value denoting an average of the light intensity values which are less than the predetermined threshold are calculated. A new threshold is obtained by applying the bright portion average value and the dark portion average value to a predetermined dividing ratio. The object within the image area is distinguished using the new threshold. The method is also applicable to the use of multivalue thresholds.

摘要翻译： 图像处理方法通过从图像区域内确定光强度值的稳定阈值来区分图像区域内的对象。 接收的光被转换成表示来自图像区域内的多个点处的接收光的光强度值的图像数据。 计算表示超过预定阈值的光强度值的平均值的亮部分平均值和表示小于预定阈值的光强度值的平均值的暗部平均值。 通过将亮度部分平均值和暗部平均值应用到预定的分频比来获得新的阈值。 使用新阈值区分图像区域内的对象。 该方法也适用于使用多值阈值。