Abstract:
An image processor based system and method for recognizing predefined-types of coating density imperfections in a web, specifically continuous type or streak imperfections. Continuous type imperfections are recognized in a continuous web moved at a certain rate through an imaging region illuminated by a stripe of substantially constant illumination. A time-delay integrating CCD camera is focused on the illuminated imaging region. The TDI CCD camera comprises an array of N rows of M light sensitive CCD elements each imaged on a fixed discrete pixel-related image area of the illuminated imaging region. The charge levels accumulated in the CCD elements of each row are shifted to the succeeding row or CCD elements and summed with the charge levels therein at a line shift clock frequency that ensures that an asynchronous relationship exists with respect to the incremental movement of the web. During the clock cycle of the N rows, the corresponding pixel areas of the illuminated web shift asynchronously or creep through the discrete pixel-related image areas. The accumulated pixel charge values derived from the pixel-related image areas of the illuminated region of said moving web emphasize imaging of longitudinal streak imperfections in the web due to the asynchronous movement the web.
Abstract:
The inspection method, in order to detect and locate defects (5) included within the thickness of the transparent material (2) to be inspected, consists in uniformly illuminating a bright background (7) placed relative to the camera (4) behind the material (2) in order to be viewed by transparency through the material (2), covering the field of view of the camera (4) and serving as contrast reference, in laterally illuminating the surface of the material (2) in order to distinguish defects (5) included within the material (2) from parasitic elements (11) deposited on its surface, in viewing by transparency, by the camera (4), placed in the vertical to the surface of the material (2), a sequence of contrasted images reproducing the thickness of the material (2), and in processing information acquired by the successive images which are representative of the material (2) seen in its thickness in order to detect and locate the defects (5) included within the thickness of the material (2).
Abstract:
A system and process for detecting and monitoring defects in large surfaces such as the field joints of the container segments of a space shuttle booster motor. Beams of semi-collimated light from three non-parallel fiber optic light panels are directed at a region of the surface at non-normal angles of expected incidence. A video camera gathers some portion of the light that is reflected at an angle other than the angle of expected reflectance, and generates signals which are analyzed to discern defects in the surface. The analysis may be performed by visual inspection of an image on a video monitor, or by inspection of filtered or otherwise processed images. In one alternative embodiment, successive predetermined regions of the surface are aligned with the light source before illumination, thereby permitting efficient detection of defects in a large surface. Such alignment is performed by using a line scan gauge to sense the light which passes through an aperture in the surface. In another embodiment a digital map of the surface is created, thereby permitting the maintenance of records detailing changes in the location or size of defects as the container segment is refurbished and re-used. The defect detection apparatus may also be advantageously mounted on a fixture which engages the edge of a container segment.
Abstract:
A method for inspecting garments for holes includes front-lighting the garment against a contrasting background and forming a pixel image thereof, isolating garment pixels from background pixels by automatically selecting an optimum binarization threshold grey level, binarizing the image at that level, determining the garment boundary, testing for consistency of grey level within the garment boundary on the binarized image along rows and columns of pixels within the garment boundary and noting inconsistent pixels, and identifying as holes only pixels which have been noted as inconsistent in both horizontal and vertical tests.
Abstract:
In a cobwebbing detection device of the type in which light illuminates the end surface of a package and a cobwebbed yarn is detected by photographing the end surface by a CCD camera, a cobwebbing detection device has a binary encoder which converts each pixel of the image data into a binary coded data, a storage device, for storing the binary-coded pixel data, and a detection circuit which detects each pixel [1] or [0] in each scanning.
Abstract:
A part scanning and part calibration apparatus and mechanism for the inspection of printed circuit boards and integrated circuits include a camera and two rotating mirrors to scan an image of a pattern mask retical upon which a precise pattern has been deposited. Small parts are placed upon the retical to be inspected. The third overhead mirror is provided to view the part under inspection from another perspective. The scene of the part is triangulated and the dimensions of the system can thus be calibrated. A precise retical mask is provided with dot patterns which provide an additional set of information needed for calibration. By scanning more then one dot pattern the missing state values can be resolved using an iterative trigonomic solution.
Abstract:
A video inspection system includes first and second video cameras mounted along parallel axes. A lighting array is pulsed, and resultant light is reflected from a specimen to both cameras. Orientation of the specimen is determined in accordance with an image generated from a first camera. This data is used to isolate a selected portion of the specimen for analysis by an image generated from the second camera.
Abstract:
As an object moves along a conveyor (A), a position sensor (B) generates a trigger signal. The trigger signal causes a strobe (14) to emit a high intensity flash of light into an examination region. A lens (20) of a CCD camera (C) focuses light from the examination region both before and during the flash on an image section (24) of a CCD array (22). Subsequent to the flash, a trigger signal delay circuit (16) causes a control circuit (D) to shift the lines of pixel values from the image section into a storage section (26), from the storage section to shift registers (32), and serially from the shift register as a video signal. A video signal channel (52) refines the video signal which is transferred to a computer system (E) which implements a preselected quality control algorithm selected in accordance with a product to be examined. The control circuit includes a pulse generator (74), which provides the appropriate clocking pulses to transfer the data among the image section, the storage section, and the shift registers. A line counter (80 ) counts the lines of pixel values transferred from the image section. When the count corresponds to one field of a video image, the counter causes gates (96, 98) to stop passing clock pulses to the image and storage sections for transferring pixel values therebetween.
Abstract:
An optical inspection system for tubular products utilizes a feeler for engaging the internal surface of the tube to be inspected and a miniature charge coupled device video camera for viewing the interface between the feeler and the tube. An image processing computer responsive to the output of the video camera produces information for rapidly assessing damage or distortion of the tube. The camera is pulled or pushed through the tube while continuously monitoring the optical contrast between the feeler and the tube wall. The camera image of the tube ID is fed to the image processing computer to determine and record changes in ID contour is a function of the axial position of the camera and feeler. These data in turn can be used in tubing life prediction considerations. The data may be analyzed manually if desired.
Abstract:
Automated inspection of solder bumps (18--18) on a major surface (14) of a chip carrier (10) is accomplished by placing the chip carrier on a platform (22) beneath a ring light (28) which is in registration with a television camera (30). Light from the ring light, which is directed at an angle towards all sides (12--12) of the chip carrier, is only reflected upwardly into the television camera by the solder bumps. The output signal of the television camera, which varies with the intensity of the light reflected from the solder bumps, is processed by a vision system (32) to obtain a one-dimensional plot of the light intensity. The one-dimensional intensity plot is analyzed automatically by the vision system to detect for missing, bridged or excessive solder bumps on the chip carrier.