Abstract: The object of a method for detecting defects in the inspection of structured surfaces is to ensure a detection of defects which is not dependent on the number of structuring planes and includes structure features in real-time operation for separating defects from good structures. From image point classification in which zones of a recorded image which have similar image point features are assembled, a gray-value intermediate image containing edge structures and corner structures is generated from the image and the behavior of the image point features of every image point in the intermediate image is analyzed with respect to its neighboring image points. The method is used predominantly in statistical process control in the production process of masks, LCD's, printed circuit boards and semiconductor wafers.

Abstract: A three dimensional inspection process for ball array devices. The ball array device is positioned in a fixed optical system. The ball array device is illuminated and a first image of the ball array device is taken with a first camera disposed in a fixed focus position relative to the ball array device to obtain a characteristic circular doughnut shape image from a ball. A second image of the ball array device is taken with a second camera disposed in a fixed focus position relative to the ball array device to obtain a top surface image of the ball. The first image and the second image are processed using a triangulation method to calculate a three dimensional position of the ball with reference to a pre-calculated calibration plane.

Abstract: A method for verification of components installed on a printed circuit board includes the step of capturing an image of an entire printed circuit board and components mounted thereon using a scanning device. After it has been determined that a printed circuit board having no defects, the scanned image of the circuit board has no defects is stored in a memory. The image of the entire printed circuit board under test, and having potential defects is then captured using a scanning device. The image of the printed circuit board under test is then stored in a memory. The stored image of the printed circuit board having no defects is then overlaid in registration with the stored image of the printed circuit board with potential defects on a display in alternating sequence in order to visually provide an indication of a defective component or defective component placement on the printed circuit board under test to an operator.

Abstract: Accuracy of a circuit assembly process (100) is determined by capturing an image (306, 1008) of an assembly formed on a substrate by the circuit assembly process (100), calculating coverage (310, 1012, 1024) of the assembly from the image, and comparing the coverage (312, 1016, 1026) to a threshold.

Abstract: A PCB testing circuit for an automatic inserting apparatus and method thereof can minimize an inferior rate of the PCB.

Abstract: Apparatus for determining at least one edge of an object having a plurality of openings with respective edges. An imaging device generates an image signal of the object to represent light intensity values of a plurality of pixels, the image signal being processed to estimate a linear relationship derived from the light intensity values of a predetermined number of the pixels. A determining device determines whether the linear relationship includes at least one light intensity value corresponding to an edge.

Abstract: A method of inspecting an image to locate a ball grid array surface-mounted device includes the steps of inspecting the image to find its surface features and to determine their locations (referred to herein as their "observed" locations); comparing expected locations of those features with the observed locations to identify missing surface features; reinspecting the image in the vicinity of apparently missing surface features to verify if the feature is really missing or to find those features and to determine their "observed" locations; and determining, from the observed locations of the surface features, the position and/or angle of the ball grid array surface-mounted device. The invention can be used to determine the position and/or angle of ball grid array surface-mounted devices with surface features in any of many array configurations, e.g., a regular lattice, a checker board lattice, a reverse checker board lattice, a regular lattice with a holes, and a custom lattice.

Abstract: The device for recognizing the position of the terminals (A) of components (BE) whose optical unwanted structures proceed in two principal directions (X, Y) has an image transducer (BW) equipped with an objective (O); and at least one segmented ring light illumination (RB; RB1, RB2) whose rays (S; S1, S2) are incident onto the surface of the components (BE) from all sides upon exclusion of angular segments (WS) around the principal directions (X, Y). By suppressing the light directions incident perpendicular and parallel to the unwanted structures, the terminals (A) are presented with clearly higher contrast than the other structures of the components (BE). The device is particularly utilized in components vision systems of automatic equipping units. It enables a reliably recognition of the position of the spherical terminals even given flip chips, bare chips and ball grid arrays and, thus, an exact optoelectronic centering of these components.

Abstract: The present invention stores digital data of a scanner and includes a component information database for storing information as data of components to be inserted in the printed circuit board and a non-insertion inspection database for storing data of lead wire clinching directions inserted in each hole from predetermined numeric control data. A control section collects hole information as data from the digital data of the scanner, reads the clinching directions of the components inserted in each hole from the non-insertion inspection database, and judges insertion states of the components according to whether or not brightness exists by inspecting a brightness change in the clinching direction around a hole.

Abstract: An automated object inspection system involving locating an object in an input image obtains a digital representation of an image including the object. Expected geometric parameters representing expected geometric features of the object are obtained. A set of pixels required for input to an object locating process is selected. A subsampling factor is defined based upon at least the expected geometric parameters. The subsampling factor defines a reduction in resolution of the set of pixels. The set of pixels is subsampled in accordance with the defined subsampling factor. The object locating process is run on the subsampled set of pixels. In defining the subsampling factor based upon at least the expected geometric parameters, the system takes into account constraints such as time an uncertainty constraints.

Abstract: An inspecting apparatus is arranged on an electronic component mounting line for inspecting whether a mounting state of the component mounted on a printed circuit board by a mounting apparatus in a precedent process or a printing state of a cream solder by a cream solder printing apparatus in a precedent process is non-defective or defective.

Abstract: A method for determining spatial characteristics of a selected object in an image that has pixel values representing heights of corresponding portions thereof is characterized by the steps of finding one or more heights of portions of the object, determining a volume of the object at and above (or below) a selected height by summing a count representing a cross-sectional area of the object at that height with a count representing a volume of the object at a greater (or lesser) height, and outputting the volume so determined. A related method determines a cross-sectional area of an object in such a "height" image by finding one or more heights of portions of the object and summing a count of pixels in the object at the selected height with a count representing a cross-sectional area of the object above that height.

Abstract: In a method for recognizing corners of an angular component, an image of a subject including the angular component is taken and outline pixels of the angular component are detected. Vector sums of first and second position vectors are calculated based on the outline pixels. A reference value having a predetermined magnitude is sampled from the vector sums. A threshold value for detecting corners of the angular component is calculated based on the reference value. The vector sums are sequentially compared with the threshold value in order to judge whether or not the corners thereof are detected according to the comparison result. It will be decided whether an operation of recognizing the corners thereof has succeeded through detecting corners of the angular component. The corners of angular component can be recognized at a high speed through the exact performance of the recognition processes, thereby mounting the components to an accurate position of the printed circuit board.

Abstract: A hybrid surface mount component inspection system which includes both vision and infrared inspection techniques to determine the presence of surface mount components on a printed wiring board, and the quality of solder joints of surface mount components on printed wiring boards by using data level sensor fusion to combine data from two infrared sensors to obtain emissivity independent thermal signatures of solder joints, and using feature level sensor fusion with active perception to assemble and process inspection information from any number of sensors to determine characteristic feature sets of different defect classes to classify solder defects.

Abstract: A system for transporting and inspecting, seriatim, semiconductor devices with plural prong type or solder ball type leads includes a head for transporting the semiconductor devices from one support structure, such as a tray or tube, to a second support structure, such as a tray or tape, and wherein two dimensional and three dimensional measurements of the positional accuracy of the leads is carried out during the transport process. The inspection apparatus is interposed in the transport path and includes a first optical sensor such as a CCD camera oriented to capture a two dimensional image of the semiconductor device package and compare the image with a predetermined two dimensional image store in a central processing unit (CPU).

Abstract: Disclosed is a method for recognizing a printed circuit board fiducial mark in order to decide an origin point in a chip mounter so that electronic parts can be mounted on a desired position of the printed circuit board. Since the method can be used for a distorted image, it can exactly recognize a damaged object as well as a general object. The method can also reduce a calculation amount relative to a polarity number of a candidate object when the candidate object has the polarity number. The method can recognize a candidate object of the same pattern fiducial mark regardless of its size.

Abstract: In a fine pattern inspection device for inspecting a fine pattern comprising a plurality of pattern elements which have the same form and which are formed on an inspection sample, the device detects a defect in the plurality of pattern elements by the use of an image derived from the fine pattern. An image obtaining unit obtains the image from the fine pattern and produces an image signal representing the image. The image signal is divided into first and second divided image signals. A processing unit extracts a defect image and an inverted defect image that is inverted in lightness of the defect image, by carrying out a first process that gives a predetermined delay to the first divided image signal and then inverts the lightness thereof to obtain a processed image signal and carrying out a second process that adds the processed image signal and the second divided image signal to obtain a difference image signal representing a difference image which includes the defect image and the inverted defect image.

Abstract: A surface image of a semiconductor wafer having a defect is picked up as a inspection image while a surface image of a semiconductor wafer having no defect is stored in an image memory as a reference image. A density difference image between the inspection image and the reference image. By extracting the defect in wiring and non-wiring regions from the density difference image, extract images are obtained. Two luminance information for wiring and non-wiring regions are obtained from extract images. Based on the luminance information, the type of the defect is determined and a production process where the defect has occurred is detect.

Abstract: A wiring pattern inspecting method comprises irradiating a test object provided with wiring patterns formed in a plurality of layers with x-rays, obtaining a variable-density image signal corresponding to the thickness of the wiring patterns superposed in a plurality of layers including superposed sections of the wiring patterns, extracting a plurality of image signals, the number of which corresponding to that of the superposed wiring patterns, from the variable-density image signal, and comparing end point information or isolated point information about the wiring patterns obtained from image signal representing a larger number of superposed wiring patterns, and branch information about the end points or the isolated points on the wiring points, obtained from the extracted image signal representing a smaller number of superposed wiring patterns to inspect the wiring patterns for defects in the wiring patterns. The wiring pattern inspecting method is carried out by a wiring pattern inspecting system.

Abstract: The apparatus includes a light source section, a recognizing section, a moving device, and an operating section. The light source section is disposed transversely of a board so as to illuminate one side of the board. The recognizing section includes lenses, which are disposed transversely of the board so as to face the board from an opposite side relative to the light source section. The recognizing section also includes line image sensors disposed transversely of the board for receiving light rays from the lenses. The moving device is operable to move both the light source section and the recognizing section without changing the relative positions of these sections but moves them in such a way as to change their respective positions relative to the board. The operating section is operable to calculate image data for the entire board from image information obtained by the recognizing section and from position information obtained by the moving device.

Abstract: Defect scanner sensitivity and accuracy are improved for light scattering defect scanners and pattern matching defect scanners by calibrating the defect scanners to each die on a wafer using preset marks on the corresponding die. The marks have a predetermined size based on the sensitivity of the defect scanners and a predetermined position relative to the circuit pattern on the corresponding die. Alignment of the defect scanners to a specific die provides improvement in coordinate accuracy over alignment with respect to an entire wafer. A layout mapping defect filtering system collects defect scan data and determines the interaction between the detected defects and a circuit layout. The layout mapping defect filtering system provides automatic identification in real time of killer defects that cause failure of the completed integrated circuit, and classifies and analyzes defects to identify potential killer defects within specified defect classes to identify defective die.

Abstract: Automated component placement apparatus includes a table for holding a board having a placement site, a placement head for holding a component to be placed on the placement site, an alignment assembly, a first actuator and a second actuator. The alignment assembly includes alignment optics for obtaining a first image of the component on the placement head and a second image of the placement site. The alignment assembly further includes an image processor for processing the first and second images to determine a relative position error between the component and the placement site and for providing a control signal representative of the position error. The first actuator causes relative movement between the placement head and the placement site, typically by movement of the placement head, in response to the control signal so as to correct the position error.

Abstract: A cryogenic apparatus for microscopy of physical properties of an object including a thin, stiff, transparent substrate or window within the outer wall of the vacuum space of a dewar and a cryogenic sensor within the vacuum space and spaced very close distances to the window. This construction allows for positioning a sample for measurement outside of the vacuum space, at room temperature or higher and for microscopy of physical properties of the sample by monitoring the output from the cryogenic sensor as it is scanned along the surface of the sample.

Abstract: Analog image data a SEM are converted into digital data, and are processed by a spatial filtering processing, histogram processing, threshold value setting, three-valued image data processing, noise reduction and the like. Area of a pattern in the three-valued image data is calculated by a labelling and calculation processing, and a pattern is sequentially detected by comparing the area of the pattern with a reference area value. The comparison and detection of the same or similar patterns repeated in the SEM image are performed by using the area of the pattern, and are not performed by a shape of the pattern, thereby resulting a precise detection at high speed by using a microprocessor. Since it is possible to perform a pattern recognition from the area value even though the pattern does not have a characteristic, it is possible to precisely detect and recognize a pattern image in high speed.

Abstract: A machine vision system for inspection of a mark (such as a multi-character mark) on an IC device as shown. The system provides search and defect analysis for the mark, individual characters of the mark, and the foreground and the background of the mark. The system provides for search and defect analysis reports thereof. It includes components for operator training search and defect analysis models for the whole mark, and for automatic training of such models for individual characters of the mark.

Abstract: If a first pattern defect is detected at the (n-2)th step, a pattern defect is retrieved at the (n-1)th step. It is decided that all pattern defects detected at the (n-1)th step in a predetermined region where the first pattern defect is detected are generated at the (n-2)th step. If second and third pattern defects are detected in other regions than the predetermined region, it is decided that the second and third pattern defects are generated at the (n-1)th step. Then, a pattern defect at the n-th step is retrieved. It is decided that all pattern defects detected at the n-th step in a predetermined region where the second and third pattern defects are detected at the (n-1)th step are generated at the (n-1)th step. If fourth, fifth and sixth pattern defects are detected in other regions than the predetermined region where the second and third pattern defects are detected, it is decided that the fourth, fifth and sixth pattern defects are generated at the n-th step.

Abstract: An operator actuates keyboard (22) to register board name of a board for teaching (ST11), sets a reference board (1S) on a Y-axis table unit (4) to actuate a start key (ST12), inputs reference points of the reference board (1S) to be set to an original position (ST13), takes a picture of a predetermined area on the reference board to teach a mounting position (ST14), and reads data about component image and its decision criteria of the corresponding component from a component kind table (19) to be stored as teaching data (ST15). This operation is repeatedly executed until all components are taught to a mounted component inspection device which inspects mounting qualities of a plurality of components mounted on an inspected board.

Abstract: An integrated machine vision inspection and rework system is provided for inspecting and reworking printed circuit boards. The system includes an inspection unit that acquires image data of printed circuit boards mounted on a machine framework and a vision computer that processes the image data contained within a chamber in the framework. Also mounted on the framework is a three stage asynchronous conveyor with sensors, and drive and stepping motors that communicate with I/O boards connected to the vision computer. A rework station also mounted on the framework provides a dual reporting scheme, which includes both a laser indicator and a fault display monitor.

Abstract: A method and apparatus for inspecting patterns of reticle data includes carrying out a sizing process in which source data for designing a given device are combined with each other; carrying out a slit filling process in which any slit, that may occur in patterns obtained by the sizing process and is unconformable to a predetermined rule, is deleted by enlarging and reducing patterns; separately storing patterns obtained by the sizing process and patterns obtained by the slit filing process; carrying out a logical operation for patterns obtained by the sizing process and patterns obtained by the slit filling process; storing patterns obtained by logical operation as graphic patterns; detecting patterns each having dimensions equal to or smaller than the predetermined value on the basis of coordinates of the graphic patterns; and deeming the thus detected patterns to be false defect patterns and distinguishing false defect patterns from true defect patterns.

Abstract: Graphic data for forming a light-shielding pattern and graphic data for forming a semitransparent pattern are stored in first and second memories, respectively. A synthesis circuit converts the graphic data stored in the first and second memories into bit patterns and writes them in a third memory. A line width determination circuit determines, as a halftone region, a region having the number of continuous bits at the same level, which number is a predetermined number or less, in the bit patterns stored in the third memory. A pattern data generation circuit generates pattern data constituted by the signal levels of a light-transmitting region and a light-shielding region on the basis of an output signal from the synthesis circuit.

Abstract: The position of a hole is measured by image-sensing a holed portion formed in a workpiece and then measuring the central position of the holed portion based on an image of the holed portion on a screen of an image sensing apparatus. A plurality of hole edge points which coincide with a hole edge of the image of the holed portion are picked up. A regression ellipse which represents the image of the holed portion from coordinates of the hole edge points is calculated. The central position of the holed portion is obtained from coordinates of the center of the regression ellipse.

Abstract: In a device used to inspect substrates visually, an imaging region and a magnification are specified ahead of time for every component and stored as library data in a memory. The memory also contains CAD data, which include the class of each component on the substrate and its location. During initialization, a control unit links the location data for a component extracted from the CAD data to corresponding library data and specifies a field of view and a magnification for every component on the substrate. Using this data, the control unit creates on the substrate a number of viewing zones to be inspected at various magnifications and stores the data specifying these zones in the memory. When a substrate is to be inspected, the control unit uses the data specifying these zones to control the position and magnification of an imaging unit to collect image data from each zone successively and display them on a monitor.

Abstract: A solder paste brick inspection and physical quality scoring system 10 employs a neural network 70 trained with a fuzzified output vector. An image of solder paste bricks 64 on a printed circuit board 12 is acquired by a CCD camera 30. Values of a predetermined set of brick metrics are extracted from the image by a computer 28 and used as a crisp input vector to trained neural network 70. A defuzzifier 76 converts a fuzzy output vector from neural network 70 into a crisp quality score output which can be used for monitoring and process control.

Abstract: A method is provided for inspecting an integrated circuit chip by use of a charged-particle microscope, such as an electron or ion-beam microscope, without incurring charge effect and thermal damage to the chip. Fundamentally, the method features the forming of a coating of good electrical and heat conductive material on the back of the target portion such that the charged particles from the microscope, after passing through the target portion, will encounter the coating of good electrical and heat conductive material and thus be drawn away by the same. As a result of this, the adverse consequences of charge effect and thermal damage can be prevented. This also allows for an increase in the resolution of the resultant image of the inspected chip.

Abstract: A mounting state of a component mounted on a printed circuit board with adhesive is inspected by an apparatus which includes an image pick-up unit for picking up in colors an image of the board after the component is mounted, a color extracting part for discriminating an adhesive portion of the image based on a predetermined color distribution of the adhesive and then extracting the adhesive portion as extracted data, a cutting part for setting a window frame for a to-be-inspected portion in the image of the board relative to the extracted data so as to detect bulging of the adhesive, and cutting out data within the window frame, and a deciding part for deciding, based on the cut data, a presence/absence of the bulging of the adhesive from a size or an area of a portion in the window frame which shows a color within the color distribution of the adhesive.

Abstract: A method and resulting system for directing generally collimated illumination from a source at an oblique angle to a surface under inspection (SUS) to produce scattered nonspecular energy substantially normal to the SUS and for observing the scattered nonspecular energy in one of a plurality of fractional windows of a viewed image of the SUS via a plurality of focussing elements. Such arrangement results in simultaneous and significant throughput and sensitivity enhancement over existing art. The method can be enhanced further wherein the observing step is performed through suitable relative motion of the scattering image over the imaging plane to permit over sampling. Furthermore, the DMT is significantly simplified because imaging is utilized to electronically scan the wafer instead of using opto-mechanical scanning means. Imaging sensors with an aggregate 2000.times.2000 pixel resolution could image a 200 mm wafer to 100 microns pitch.

Abstract: A method and apparatus which incorporate self learning techniques for the detection of solder defects and for statistical process control of solding operations on printed circuit board assemblies (PCBA) are disclosed. The invention includes learning techniques which are used during the inspection of cross-sectional X-ray images of solder joints. These learning techniques improve measurement accuracy by accounting for localized shading effects, which can occur when inspecting double-sided printed circuit board assemblies. Two specific examples are discussed. The first is a method for detection of solder short defects. The second method utilizes learning to improve the accuracy of statistical process control (SPC) measurements.

Abstract: Pieces of feature information in a bi-level image such as a width, a branching point and an ending point of a wiring pattern are detected in a design rule checking section according to a micro inspection to find out pieces of feature information departing from a design rule. Also, feature codes in the bi-level image such as a corner of the wiring pattern are detected in a specific shape detecting section according to a macro inspection. The feature information are compared with pieces of referential feature information pertaining to a non-defective wiring pattern in a first comparing and judging section to judge whether the wiring pattern indicated by the feature information is defective or non-defective. The feature codes are compared with referential feature codes pertaining to a non-defective wiring pattern in a second comparing and judging section to judge whether the wiring pattern indicated by the feature codes is defective or non-defective.

Abstract: An electron beam tester scans a sample with an electron beam to provide a secondary electron image, matches wiring patterns of the secondary electron image with wiring patterns prepared from CAD data, measures voltages of the wiring patterns, and corrects deformation of the secondary electron image. The electron beam tester comprises a pattern matching unit, a wiring pattern tester, a secondary electron image corrector, a wiring pattern inspection unit, and a pattern matching processor. The pattern matching unit simply and quickly matches the wiring patterns of the secondary electron image with the wiring patterns prepared from the CAD data. The wiring pattern tester detects slippage between layers of a multilayered semiconductor chip and correctly positions the electron beam on the chip during a pattern matching operation. The secondary electron image corrector accurately and automatically corrects deformation of the secondary electron image.

Abstract: A method and apparatus for inspecting a bonded joint between components. A cross-sectional image of the joint is analyzed by determining the location of a first characteristic of the joint, the centroid of the joint in the cross-sectional image; and then measuring a second characteristic of the joint in reference to the location of the centroid. This measurement may be used by comparing it with a predetermined specification expected for the measurement for the purpose of determining the quality of the joint. The invention is particularly advantageous for studying cross-sectional X-ray images of solder joints between electronic components and substrata upon which they are mounted and can be implemented for analyzing the images produced in cross-sectional X-ray inspection machines such as scanned beam X-ray laminography systems or digital tomosynthesis systems.

Abstract: A printed circuit board inspection system acquires images for purposes of inspection from viewing fields defined along the surface of the printed circuit board to be inspected which can be varied in size to accommodate the density and tolerance of components associated with the printed circuit board under inspection. To this end, the cameras of the inspection head associated with the printed circuit board inspection system are provided with zoom lenses, which are capable of controlled operation responsive to the existing microprocessor systems of the printed circuit board inspection system. This allows the magnification of images acquired by the cameras to be varied, as desired, in turn varying the size of the viewing fields defined for inspection purposes. The inspection head is then operated at a rate commensurate with the size of the viewing fields which have been selected.

Abstract: Disclosed is a new self-reference signal processing technique for detecting the location of any nonregularities and defects in a periodic two-dimensional signal or image. Using high-resolution spectral estimation algorithms, the proposed technique first extracts the period and structure of repeated patterns from the image to sub-pixel resolution in both directions, and then produces a defect-free reference image for making comparison with the actual image. Since the technique acquires all its needed information from a single image, on the contrary to the existing methods, there is no need for a database image, a scaling procedure, or any apriori knowledge about the repetition period of the patterns.

Abstract: A method and apparatus for compensating substrate distortion in automatic optical inspection (AOI) systems eliminates false error signals during printed circuit boards (PCB) defect detection. The present invention is used in feature edge transition AOI systems that employ pixel to pixel comparison between CAD and scanned PCB feature databases. The present invention divides the substrate into a plurality of regions or "tiles". A distortion vector for each tile is calculated and is used to generate a shift of pixels within each tile which removes any variation between the preferred pixel location and the actual pixel location, thereby enabling the AOI system to avoid generating error signals when features are only misplaced on the substrate.

Abstract: An optical inspection system for inspecting faults on circuit boards like printed circuit boards and wire scribed circuit boards is disclosed. A video camera for receiving desired images of hole sites on a wire circuit board is provided. The camera converts each image to electrical video signals. A frame grabber converts electrical video signals to a plurality of pixels and stores information representing color element values contained in each of the pixels. A panel feature detector receives the color element values of the pixel that has to be analyzed and provides a wire indication when the color element value of the pixel corresponds to the wire feature. A fault detector circuit receives the wire indication from the panel feature detector corresponding to predetermined pixels associated with the hole site for determining faults on the circuit board.

Abstract: A wiring pattern inspection apparatus for inspecting an abnormality of a wiring pattern formed on a printed circuit board, which is equipped with an optical image pickup device for optically illuminating a surface of the printed circuit board including the wiring pattern to photoelectrically convert optical information of the printed circuit board surface due to the optical illumination into a grey level image. This grey level image is converted into a bi-level image which separates the grey level image into the wiring pattern side and a background side of the wiring pattern. Thereafter, the bi-level image is once contracted by a first size and then expanded by a second size so as to eliminate a micro conductive portion left on the printed circuit board or a micro pinhole which can be disregarded in the abnormality inspection, thereby preventing the excessive detection.

Abstract: An automated electronic module inspection system having illumination sources that produce digitized gray scale images that are then analyzed by a computer. The inspection system illuminates the module including solder joints, components and printed circuit board using a sequence of illumination that produce visual images which are used to identify defects. The inspection system inspects for all defects as specified in MIL-STD-2000A (Section 4.4) by using both surface reflectance properties as well as topological properties of objects. The system comprises two video cameras. A positioning table aligns a module-under-test with the optical axis of each camera thereby enabling the same field-of-view to be observed under different lighting conditions. A dark field illumination associated with a first camera preferentially illuminates non-flat surfaces, but not vertical surfaces. A bright field illumination associated with a second camera preferentially illuminates relatively flat surfaces.

Abstract: An apparatus for visual inspection detects defects on an object to be inspected, stores the defects, calculates the distance between two detected defects from the positions of the two detected defects, and judges the two detected defects to be different portions of a single actual defect when the distance between the two defects is less than a predetermined threshold value. The apparatus also stores the actual defects, as well as class-of-actual-defect inference rules necessary for inferring the class of the actual defects from the class of each detected defect. The class of the actual defect is judged from the class of each detected defect on the basis of the class-of-actual-defect inference rules, and the class of the actual defect is then stored. A method of visual inspection and a method of removing the cause of an abnormality are also disclosed.

Abstract: A calibration system for circuit inspection tools in which analog threshold levels are automatically set by digital signals. The inspection too includes an optical detector system that provides video signals to channels each containing an analog processor. In the analog processor an analog reference signal, an analog video inspection signal which may contain a defect indication and a digital threshold signal are employed. The digital threshold signal is converted into an analog threshold signal which is summed with the analog reference signal that has been inverted. The inverted, thresholded reference signal is then summed with the analog video signal. The summed analog video signal and thresholded reference signal is compared with a zero level. If the sum is greater than zero a defect is indicated.