-
公开(公告)号：US5390108A
公开(公告)日：1995-02-14
申请号：US705122
申请日：1991-05-24
申请人： Roman Baldur , Michael R. Barrett
发明人： Roman Baldur , Michael R. Barrett
CPC分类号： G06T7/0044 , F42B35/00 , G06Q10/087 , G06T7/0002 , G06T7/0024
摘要： The apparatus analyzes and compares surfaces of fired bullets. A microscope obtains and amplifies optical signals representative of the characteristics of the surfaces, and the optical signals are converted, first to electrical signals and then to electrical coded digital representations. The coded digital representations are stored in an electronic memory. The coded digital representations of one bullet can then be computer compared to the coded digital representations of another bullet to determine if there is a match between the bullets.
摘要翻译： 该装置分析和比较发射子弹的表面。 显微镜获得并放大表示表面特性的光信号，首先将光信号转换为电信号，然后转换为电编码数字表示。 编码数字表示存储在电子存储器中。 然后可以将一个子弹的编码数字表示与另一个子弹的编码数字表示进行计算机比较，以确定子弹之间是否存在匹配。
-
公开(公告)号：US5386508A
公开(公告)日：1995-01-31
申请号：US745897
申请日：1991-08-16
申请人： Katsuhiko Itonori , Noboru Shimizu
发明人： Katsuhiko Itonori , Noboru Shimizu
CPC分类号： G06F8/34 , G05B19/056 , G06K9/00476 , G05B2219/13113
摘要： A program generator including: a character/figure separating device for separating characters and line graphics from an input flowchart image. A character recognizing device performs character recognition on the separated character portion and a figure recognizing device extracts symbols from the line graphics. A program code generating device generates instructions corresponding to the symbols and generates a program code by combining the generated instructions with the results of the character recognition that corresponds to the positions in which the symbols have been extracted. In the program generator, an image signifying the procedure of program processing which is composed of characters and line graphics, represented in a format easy for users to understand, is separated into an image in the character region and an image in the line graphics region and those images are subjected to individual recognition, with the results of recognition being combined to generate a coded program that is compatible with automatic machine processing. Hence, a desired computer program can be prepared by simple steps in a rapid and accurate way.
摘要翻译： 一种程序生成器，包括：用于将字符和行图形与输入流程图图像分离的字符/图形分离装置。 字符识别装置对分离的字符部分执行字符识别，并且图形识别装置从线图形提取符号。 程序代码生成装置生成与符号相对应的指令，并且通过将生成的指令与对应于已经提取了符号的位置的字符识别结果组合来生成程序代码。 在程序生成器中，表示以用户易于理解的格式表示的字符和行图形的程序处理程序的图像被分离为字符区域中的图像和行图形区域中的图像， 对这些图像进行单独识别，结合识别结果生成与自动机器处理兼容的编码程序。 因此，可以通过简单的步骤以快速和准确的方式准备所需的计算机程序。
-
公开(公告)号：US4809194A
公开(公告)日：1989-02-28
申请号：US901115
申请日：1986-08-28
申请人： Rex J. Crookshanks
发明人： Rex J. Crookshanks
IPC分类号： G06E3/00 , G06F15/66 , G06F15/336 , G06F15/70
CPC分类号： G06E3/005
摘要： An image processing system 10 includes an array (12) of detectors 14, each of which is designed to produce a current proportional to incident radiation. This system provides image processing at a viable sampling rate even for very large arrays and permits very efficient determination of single element detections.The modulation functions supplied from a weighted summer (18). The weighted summer applies an invertible matrix of weights to a series of orthonormal Walsh functions defined over a predetermined sampling interval, the Walsh functions being generated by a function generator (16).The modulated outputs of the array are combined by a summer (20) and distributed among parallel channels by a divider (22). Correlators (24) correlate the signal in each channel with a respective one of the original Walsh functions. The correlated outputs are digitized by analog-to-digital converters for transmission and processing by a digital processor (28). The processor can at least partially reconstruct the detected spatial distribution for output to a display (30).
摘要翻译： 图像处理系统10包括检测器14的阵列（12），每个检测器被设计成产生与入射辐射成比例的电流。 该系统即使对于非常大的阵列也以可行的采样率提供图像处理，并且允许非常有效地确定单个元件检测。 从加权夏季（18）提供的调制功能。 加权的加法器将权重的可逆矩阵应用于在预定采样间隔上定义的一系列正交沃尔什函数，沃尔什函数由函数发生器（16）产生。 阵列的调制输出由加法器（20）组合，并通过分频器（22）在并行通道之间分配。 相关器（24）将每个通道中的信号与原始沃尔什函数中的相应的一个相关联。 相关输出由模数转换器数字化，用于数字处理器（28）的传输和处理。 处理器可以至少部分地重建检测到的空间分布以输出到显示器（30）。
-
公开(公告)号：US4783751A
公开(公告)日：1988-11-08
申请号：US016495
申请日：1987-02-17
CPC分类号： G06K9/0014 , G01N15/088 , G01N33/241 , G01N2015/0846
摘要： Analog signals corresponding to color images or pixels of a thin section of a rock reservoir are digitized and then filtered to isolate the pixels representative of pores in the thin section. The pores so isolated are then counted, measured for their total pore perimeter and labelled.Each area of pixels representative of pores of value one is progressively eroded and dilated, pursuant to which one layer of pixels of ones on the perimeter of the area is converted to zeros and, if a seed pixel remains, one layer of pixels of ones is added to the perimeter of the area. Thereafter, the original object undergoes two successive erosions followed by two dilations if a seed pixel remains. Successive iterations of the erosion and dilation cycle continue until the last erosion eliminates the seed pixel of the area. The numbewr of pixels of ones lost with each degree of erosion constitutes a pore spectrum consisting of information relating to the total amount of pore image lost each erosion-dilation cycle, the pore size lost each erosion-dilation cycle and the pore roughness lost each cycle.The spectra devleoped from the erosion-dilation cycle and corresponding to each pore complex are then analyzed into end members and the end member proportions for each field of view are calculated.
摘要翻译： 对应于岩石储层的薄部分的彩色图像或像素的模拟信号被数字化，然后过滤以分离代表薄部分中的孔的像素。 然后计数这样分离的孔，测量其总孔周长并标记。 表示值为1的孔的每个像素区域被逐渐侵蚀和扩张，根据该区域的周边上的一个像素像素被转换为零，并且如果剩余种子像素，则一层像素为 添加到该地区的周边。 此后，如果种子像素保留，则原始物体经历两次连续的侵蚀，随后进行两次扩张。 侵蚀和扩张循环的连续迭代持续到最后的侵蚀消除了该区域的种子像素。 每次侵蚀损失的像素像素数量构成了一个孔隙谱，其中包括与每个侵蚀 - 扩张循环损失的孔隙图像的总量有关的信息，每个侵蚀 - 扩张循环的孔径损失和每个循环的孔隙粗糙度损失 。 然后将从侵蚀扩张循环中去除的并对应于每个孔隙复合物的光谱分析成端部构件，并计算每个视场的端部构件比例。
-
公开(公告)号：US4587617A
公开(公告)日：1986-05-06
申请号：US543038
申请日：1983-10-18
申请人： John C. Barker , Ian A. Cruttwell
发明人： John C. Barker , Ian A. Cruttwell
IPC分类号： G01N21/88 , G01N21/956 , G03F1/08 , H01L21/027 , H01L21/66 , G06F15/70 , G06K9/46
CPC分类号： G01N21/956
摘要： A defect detection system for classifying defects in reticles (such as used in the manufacture of integrated circuit wafers) is described. An inspection unit, scanning the reticle in real time, compares it with the correct form and inputs signals representing each defect and its position in a store. The stored defects are then used to direct the inspection unit to inspect each defect more slowly. The inspection unit classifies each defect as either being an "excess metal" or a "missing metal" defect, and corresponding signals are stored in respective additional stores. The signals stored in these stores are then further processed and compared with the desired form of the reticle so as to classify the "excess metal" defects as either being a "pin spot" defect, an "extension" defect or a "bridge-type" defect. Similarly, the "missing metal" defects are classified as either being a "pinhole" defect an "intrusion" defect or a "break-type" defect. A size measuring unit measures the size of each defect.
摘要翻译： 描述了用于对掩模版中的缺陷进行分类的缺陷检测系统（例如用于制造集成电路晶片的缺陷）。 检查单元实时扫描光罩，将其与正确的形式进行比较，并输入表示每个缺陷的信号及其在商店中的位置。 然后使用存储的缺陷来引导检查单元更慢地检查每个缺陷。 检查单元将每个缺陷分类为“过量金属”或“缺少金属”缺陷，并且相应的信号存储在相应的附加存储中。 然后将存储在这些存储器中的信号进一步处理并与标线的期望形式进行比较，以将“过量金属”缺陷分类为“引脚点”缺陷，“扩展”缺陷或“桥型” “缺陷。 类似地，“缺失金属”缺陷被分类为“针孔”缺陷“入侵”缺陷或“断裂型”缺陷。 尺寸测量单元测量每个缺陷的尺寸。
-
6.发明授权Projection view creation method with vector discrimination onto plane-represented curved surfaces 失效投影视图创建方法，其中矢量识别在平面表示的曲面上
公开(公告)号：US5586232A
公开(公告)日：1996-12-17
申请号：US350143
申请日：1994-11-29
申请人： Toshiharu Yoshida
发明人： Toshiharu Yoshida
CPC分类号： G06T17/00
摘要： A method of creating a view of a curved line projected onto a curved surface in a CAD system by using a vector discrimination technique which reduces computation steps required for projection. The curved line and the curved surface are divided into a group of straight line segments and a group of planar surfaces, respectively. One of the straight line segments is selected and a first vector is formed at an end of the straight line segment. After a discrimination process using the first vector, a combination of one straight line segment and one planar surface pointed to by the first vector is obtained, and the projection operation starts with this combination. Second and third vectors are then formed at both ends of an edge of the planar surface. The straight line segments within the range delineated by the second and third vectors are projected onto the planar surface. If there is a straight line segment falling outside the range, the next planar surface is selected and the excessive part of the line segment is projected onto the newly selected planar surface. The projection operation thus continues until all the straight line segments are projected on the planar surfaces.
摘要翻译： 通过使用减少投影所需的计算步骤的矢量鉴别技术来创建投影到CAD系统中的曲面上的曲线的视图的方法。 曲线和曲面分别分成一组直线段和一组平面。 选择直线段中的一个，并且在直线段的末端形成第一矢量。 在使用第一矢量的辨别处理之后，获得由第一矢量指向的一条直线段和一个平面的组合，并且以该组合开始投影操作。 然后在平面的边缘的两端形成第二和第三矢量。 由第二和第三矢量描绘的范围内的直线段投影到平面上。 如果直线段落在范围之外，则选择下一个平面，并且将线段的过多部分投影到新选择的平面上。 因此，投影操作继续直到所有直线段投影在平面上。
-
7.发明授权
公开(公告)号：US5402346A
公开(公告)日：1995-03-28
申请号：US112842
申请日：1993-08-25
申请人： Didier Lion , Eric de Poulpiquet
发明人： Didier Lion , Eric de Poulpiquet
IPC分类号： G01S13/50 , G01S13/58 , G01S13/64 , G01S13/91 , G01S13/92 , G08G1/01 , G08G1/015 , G06F15/70
摘要： This invention comprises a system for calculating at least one vehicular traffic check parameter. The system according to the invention includes a pulse modulated radar placed at a high point at the side of the lanes to be monitored. The radar emits pulses having a frequency f.sub.o and width .tau. according to a radiation diagram. The radiation diagram is narrow in plan and sufficiently wide in elevation to cover all lanes to be monitored, and inclined in elevation and plan such that at least one range bin of the radar is included in a lane. Echo signals received by the radar are then processed by a signal processing module and a parameter extraction module. The invention is particularly useful in automobile traffic, for calculation of the number of vehicles travelling in lanes, the vehicles speed and length, etc. The advantage of the this inventive system compared with previous systems is that it enables simultaneous checking of several traffic lanes.
摘要翻译： 本发明包括一种用于计算至少一个车辆交通检查参数的系统。 根据本发明的系统包括放置在要监视的车道侧的高点处的脉冲调制雷达。 雷达根据辐射图发射具有频率fo和宽度τ的脉冲。 辐射图在平面上是狭窄的，并且在高度上足够宽以覆盖要监视的所有通道，并且在高程和倾斜方向上倾斜，使得雷达的至少一个范围箱包括在车道中。 然后由信号处理模块和参数提取模块处理由雷达接收到的回波信号。 本发明在汽车交通中特别有用，用于计算车辆行驶的车辆数量，车辆速度和长度等。与先前系统相比，本发明的系统的优点是能够同时检查几个行车道。
-
8.发明授权
公开(公告)号：US5289374A
公开(公告)日：1994-02-22
申请号：US843715
申请日：1992-02-28
申请人： Kunio Doi , Tsuneo Matsumoto , Maryellen L. Giger , Akiko Kano
发明人： Kunio Doi , Tsuneo Matsumoto , Maryellen L. Giger , Akiko Kano
CPC分类号： G06T7/0012 , G06F19/345 , G06K9/346 , G06F19/321 , G06K2209/05 , G06T2207/30064
摘要： A computerized method and system for reducing the number of false-positive detections of nodule candidates in the detection of abnormalities in digital chest radiography. The image is initially subjected to an image difference technique where the detection sensitivity is increased so as to avoid missing small nodules which might otherwise go undetected. Such a technique tends to increase the number of false-positives, however, leading to possible incorrect diagnoses of the radiographs. To reduce the number of false-positives, feature extraction techniques are applied to grown regions around the nodule candidates, in order to provide computer generated information concerning the candidates. A data base of parameters common to false-positives is compared to calculated parameters of a candidate of interest. The candidates with grown region parameters within the data base range common to false-positives are eliminated as being probable false-positive detections due to normal background anatomical features.
摘要翻译： 一种计算机化方法和系统，用于在数字胸部X线检查异常检测中减少结节候选者的假阳性检测次数。 图像最初经受图像差异技术，其中检测灵敏度增加，以避免丢失可能未被检测到的小结节。 这种技术倾向于增加假阳性的数量，然而，导致可能的错误诊断的X光片。 为了减少假阳性的数量，将特征提取技术应用于结核候选者周围的生长区域，以便提供关于候选者的计算机产生的信息。 将假阳性共同参数的数据库与感兴趣的候选者的计算参数进行比较。 具有假阳性共同的数据库范围内生长区域参数的候选者被消除为由于正常背景解剖特征引起的可能的假阳性检测。
-
公开(公告)号：US5287293A
公开(公告)日：1994-02-15
申请号：US76085
申请日：1993-06-14
申请人： Hsien-Yei Chen , Chung-Mu Hwang , Jung-Lung Chang , Chieh-Yu Lin , Ruey-Yang Wang , Jyh-Jye Jeng
发明人： Hsien-Yei Chen , Chung-Mu Hwang , Jung-Lung Chang , Chieh-Yu Lin , Ruey-Yang Wang , Jyh-Jye Jeng
CPC分类号： G06T7/0006 , G06T7/0083 , G06T2207/10056 , G06T2207/20148 , G06T2207/30164
摘要： A method and apparatus for the automatically inspecting the contours of a gear for manufacturing errors. According to the method, a gear to be inspected is imaged on an imaging surface to form a gear image. A digitized image indicative of the gear image is generated. The digitized image comprises a plurality of pixels. The pixels each have a gray value and coordinates which represent the location of the pixel within the digitized image. Each pixel is classified into a first or a second class by comparing the gray value of the pixel to a predetermined threshold value. The coordinates of pixels corresponding to the contours of the gear are determined from the classes of pixels. The manufacturing errors of the gear are calculated by comparing the coordinates of the pixels corresponding to the contours of the gear to a standard gear pattern. The apparatus, in its preferred embodiment, includes a base and collimated light source to illuminate the gear, a microscope to magnify an image of the gear and a CCD camera to generate a digitized image indicative of the magnified gear image. A digital computer is also included to assign a gray value to each pixel in the digitized image, to classify each pixel into a first or second class by comparing the gray value of the pixel with a predetermined threshold value, to determine the coordinates of pixels corresponding to the contours of the gear and to calculate the manufacturing errors of the gear by comparing the coordinates of pixels corresponding to the contours of the gear with a standard gear pattern. The manufacturing errors include runout error, pitch error and profile error.
摘要翻译： 用于自动检查齿轮轮廓以制造误差的方法和装置。 根据该方法，将要检查的齿轮在成像表面上成像以形成齿轮图像。 产生指示齿轮图像的数字化图像。 数字化图像包括多个像素。 像素各自具有灰度值和表示像素在数字化图像内的位置的坐标。 通过将像素的灰度值与预定阈值进行比较，将每个像素分类为第一类或第二类。 根据像素的类别确定与齿轮的轮廓相对应的像素的坐标。 通过将与齿轮轮廓对应的像素的坐标与标准齿轮图案进行比较来计算齿轮的制造误差。 在其优选实施例中，该装置包括用于照亮齿轮的基座和准直光源，用于放大齿轮图像的显微镜和CCD相机以产生指示放大齿轮图像的数字化图像。 还包括数字计算机以对数字化图像中的每个像素分配灰度值，通过将像素的灰度值与预定阈值进行比较来将每个像素分类为第一类或第二类，以确定相应像素的坐标 相对于齿轮的轮廓，并且通过将对应于齿轮的轮廓的像素的坐标与标准齿轮图案进行比较来计算齿轮的制造误差。 制造误差包括跳动误差，螺距误差和轮廓误差。
-
公开(公告)号：US4623973A
公开(公告)日：1986-11-18
申请号：US680910
申请日：1984-12-12
申请人： Ingo Hoffrichter , Eggert Jung
发明人： Ingo Hoffrichter , Eggert Jung
CPC分类号： G01J3/46 , G01J3/462 , H04N1/54 , G01J2003/466
摘要： In electronic reproduction technology, a recognition circuit for hues and colors in colored surfaces is provided. Before the actual hue recognition, the hue to be recognized is selected and its prescribed or measured color values of the RGB color space are transformed by a first color transformation into chrominance values of the chrominance/luminance color space, whereby the transformation coefficients are defined such that the transformed color locus of the hue to be recognized lies on a rotated X'Y' color coordinate system in the chrominance plane. During the actual hue recognition, the colored surface is opto-electronically scanned and the measured color values acquired are transformed into chrominance values by a second color transformation with the transformation coefficients defined in the first color transformation. By selection of chrominance values and by quotient formation of the selected chrominance values, a hue signal which is a measure for hue deviation of the scanned hues from the hue to be recognized is generated. After combination of the hue signal with an auxiliary control signal, a hue recognition signal is formed, this being unequal to zero when the scanned hues lie within a sector-shaped hue recognition region. For color recognition, a color saturation signal and a luminance signal are additionally generated. The color saturation signal and luminance signal are adjustably limited in terms of amplitude and are combined with the hue recognition signal to form a color recognition signal.
摘要翻译： 在电子再现技术中，提供了用于彩色表面中的色调和颜色的识别电路。 在实际色调识别之前，选择要识别的色调，并且将其颜色空间的规定或测量的颜色值通过第一颜色变换变换为色度/亮度色彩空间的色度值，由此将变换系数定义为 要识别的色调的变换颜色轨迹位于色度平面中旋转的X'Y'色坐标系。 在实际色调识别期间，彩色表面被光电扫描，并且通过第一颜色变换中定义的变换系数通过第二颜色变换将获得的测量的颜色值转换成色度值。 通过选择色度值和所选色度值的商形成，产生色调信号，该色调信号是被扫描的色调与要识别的色调的色调偏差的度量。 在将色调信号与辅助控制信号组合之后，形成色调识别信号，当扫描的色调位于扇形色调识别区域内时，该色调识别信号不等于零。 为了颜色识别，另外产生色彩饱和度信号和亮度信号。 彩色饱和度信号和亮度信号在幅度方面可调节地限制，并与色相识别信号组合以形成颜色识别信号。
-
-
-
-
-
-
-
-
-
搜索结果
38 条记录 (耗时 0.016 秒)
