Abstract: Methods and apparatuses are disclosed for refining groupings of edge points that represent contours in an image. The methods and apparatuses decrease data dispersion and data quantization effects. The methods and apparatuses are particularly useful for accurate and robust detection of straight line-segment features contained in noisy, cluttered imagery occurring in industrial machine vision applications. Additionally, a measurement criterion of the quality of the detected line segments is introduced.

Abstract: A method and an apparatus are disclosed for finding an orientation of an object from the image of the object through the generation of, and subsequent evaluation of, at least a portion of a frequency response of the image. The orientation of the object can be used in subsequent image processing.

Abstract: A method and an apparatus are disclosed for generating a model of an object from an image of the object. First, an orientation of the object in the image is determined through the generation of, and subsequent evaluation of, at least a portion of a frequency response of the image. Thereafter, the orientation is used to gauge the object. The gauging provides the necessary dimensional information that becomes a part of the geometrical model of the object. An embodiment is disclosed that generates a geometrical model of a leaded object from the image of the leaded object.

Abstract: An efficient and reliable method and apparatus is disclosed that finds a reference point of an object profile within an image when the object is of an unknown size. The object profile is modeled using a synthetic labeled-projection model, which in conjunction with the image, is projected over a portion of the image of the object profile to derive a histogram. The histogram is normalized and a maximum of a first derivative of the histogram is defined for that position. The position of the labeled-projection model is moved relative to the image, and the process is repeated until a selected portion of the image has been examined. The first derivative of the normalized labeled projection is greatest when a feature of the image and the feature denoted by a specific synthetic label of the labeled-projection model are aligned. The method and apparatus can locate the center of the object with reliability, because use of the labeled-projection model and the histogram minimizes the * effects of image artifacts.

Abstract: The disclosed methods and apparatuses leverage a known value of a characteristic of an object to partially calibrate an imaging system “on-the-fly”, and minimize, if not eliminate, the need for a separate calibration image(s). Specifically, the scale relationship (i.e. the relationship between physical dimensions and image dimensions) is calculated using the known value and a measured value of the characteristic from the image. The same image used to calculate the scale relationship is also processed, such as inspected, for example. The known value can be a measurement of an aspect of many things, including an inherent feature, or a relationship between features, for example. One embodiment uses a model to find the characteristic. A described preferred embodiment inspects an end-face of a fiber-optic cable, wherein the known value is the diameter of an annular cladding of the fiber-optic cable.

Abstract: A method and apparatus are disclosed for analyzing a boundary of an object. An embodiment for determining defects of a boundary to sub-pixel precision and an embodiment for fast correlation scoring are disclosed. The boundary is analyzed by matching a first boundary, such as a model of an ideal object boundary, to a second boundary, such as the boundary of an object being produced at a factory. The boundaries are represented as a set of indexed vertices, which are generated by parsing the boundaries into a set of segments. One embodiment refines the parse through merging segments and reassigning data points near the endpoints of the segments. The model produced is very accurate and is useful in other applications. To analyze the boundaries, the sets of indexed vertices are matched, and optionally the segmentation of the second boundary is refined to increase the extent of matching.

Abstract: A methods and apparatuses are disclosed for determining the state of elements in an image. The elements can be arranged in any regular pattern, such as a matrix of LCD display elements. The state of the elements is determined by extracting a characteristic of the elements, such as intensity, and processing the characteristic to provide comparison criteria, such as a difference value. The comparison criteria are then used to determine the state of the element alone or in combination with other comparisons, such as average intensity comparisons. Several comparisons and variations thereof are disclosed, as well as the preferred ordering and configuration of comparisons for LCD displays. Also disclosed is an embodiment where a model of the elements is generated and used to inspect the elements.

Abstract: Methods and apparatuses are disclosed for refining groupings of edge points that represent contours in an image. The methods and apparatuses decrease data dispersion and data quantization effects. The methods and apparatuses are particularly useful for accurate and robust detection of straight line-segment features contained in noisy, cluttered imagery occurring in industrial machine vision applications. Additionally, a measurement criterion of the quality of the detected line segments is introduced.

Abstract: Methods and apparatuses are disclosed for identifying regions of similar texture in an image. The areas of similar texture include areas conventionally thought of as similar texture regions as well as areas of more varied texture that are treated as regions of similar texture in order to identify them within an image. The method associates frequency characteristics of an image with a spatial position within the image by: applying a frequency analysis on sub-regions of the image, thereby, generating frequency characteristics representative of the sub-regions: and associating the frequency characteristics with the origin of the sub-regions. An embodiment disclosed applies a fast Fourier transform on sub-regions in a given direction to determine a dominant frequency of the sub-region and the power of the dominant frequency, both of which are associated with the respective sub-region by storing the dominant frequency and power in a frequency image and power image, respectively, at the position of the origin.

Abstract: The invention provides methods and apparatuses for generating from an image of an object a model of the object, when the object contains similar features. First, the invention finds the similar features. The similar features are found by searching portions of the image for features that are substantially similar to a feature prototype. More particularly, individual features in the image are located and designated candidate features, and optionally a spatial pattern representing a majority of the candidate features is generated. Next, feature profiles are generated therefrom, and the feature prototype is constructed using at least a subset of feature profiles. Thereafter, the model is generated using the similar features and, optionally, the spatial pattern.

Abstract: A methods and apparatuses are disclosed for determining the state of elements in an image. The elements can be arranged in any regular pattern, such as a matrix of LCD display elements. The state of the elements is determined by extracting a characteristic of the elements, such as intensity, and processing the characteristic to provide comparison criteria, such as a difference value. The comparison criteria are then used to determine the state of the element alone or in combination with other comparisons, such as average intensity comparisons. Several comparisons and variations thereof are disclosed, as well as the preferred ordering and configuration of comparisons for LCD displays. Also disclosed is an embodiment where a model of the elements is generated and used to inspect the elements.

Abstract: A method and apparatus are disclosed for refining a rough geometric description (GD) and a rough pose of an object having extensions. The invention locates anchor points within an image of the object and uses the anchor points to align in at least one dimension the rough GD. In one embodiment, the anchor points are the tips of the extensions of the object; the rough GD of the object is then aligned along the angular orientation indicated by the tips. Thereafter, other dimensions of the rough GD and the rough pose are measured, measuring the dimensions having less unknowns first. The rough GD and the rough pose are then updated to provide the refined GD and refined pose. For one measurement, an extent of a region is measured using the expected area of region to threshold the region and segment it from the remainder of the image before measuring the extent of the region. An application of refining a GD of a leaded object is disclosed.

Abstract: An efficient and reliable method that identifies possible reference sites in an image is disclosed. The method emphasizes customizing the identification of reference sites for each application by providing for initialization of various parameters, such as the size of the reference site. The method then uses a series of measurable parameters to filter possible reference sites in the image and produces and ordered set of possible reference sites. The highest-ranking site of the ordered set is expected to be chosen as the reference site for production. A preferred implementation is disclosed that uses multi-resolution images to enhance efficiency of the identification and that specifically measures the symmetry, orthogonality and uniqueness of the windows.

Abstract: A method and apparatus is provided for intuitively, graphically parameterizing a region of an input image to be processed by a machine-vision vision system, the region being defined by a deformable window, where the window is deformed by a user using a pointing device. The window pictorially indicates the defined region to the user. The region identified and parameterized is transformed into a destination region. A preferred embodiment is a parameterization of a polar region and the subsequent transformation of the polar region into an OCR rectangular window that is then available for further processing, such as by an OCR software application.

Abstract: Inspection of solder paste on a printed circuit board using a before printing image (pre-application image) to normalize an after printing image (post-application image) of the printed circuit board. Existing lighting and optics used for alignment of the screen-printing stencil to the printed circuit board are used for the solder paste inspection. An embodiment is described wherein pad regions of the printed circuit board are inspected for information about the solder paste applied on the pad regions of the printed circuit board. A stencil in the screen printing process is also inspected using a before printing image (pre-application) to normalize an after printing image (post-application) of the stencil.

Abstract: Methods and apparatuses are disclosed for defining a locus of viable points on an elongated object, such as a locus of viable bond points on a lead of a lead frame. The locus guides the bond point positioning of a wire bonder so that the number of unacceptable wire bonds is reduced. The locus is a region adapted to receive a bond, and is defined such that for any bond centered therein, the probability of forming an unacceptable bond is reduced by offsetting the locus from at least a portion of the lead edge(s) by a minimum-offset distance. The minimum-offset distance is based on various factor(s), such as user tolerance and bond width. Further, the definition of the locus is independent of lead shape, and consequently works with a variety of lead shapes. The locus can also have various shapes for a single lead, and the shapes can vary depending on the application.

Abstract: A system to align oriented (unaligned) objects in an image to permit such operations as dimensional gauging or measurements to be performed. Using the angle of rotation of the object in the unaligned image, the system creates a destination image having rows and columns of pixels aligned with a specified axis. The unaligned image is then aligned by finding a first pixel location in a first column in the destination image; locating a first whole pixel location nearest to the first pixel area in the unaligned image; copying the actual pixel value of the first whole pixel location in the unaligned image to the first pixel location in the destination image; locating a next whole pixel location in the unaligned image; copying the actual pixel value of the next whole pixel location to a next pixel location in the destination image until all pixels in a first column of the window have been copied to the first column of the destination image; locating the next column, and continuing until a last column is located.

Abstract: An apparatus and method for two-dimensional search for a model image within atest image that exploits edge-based matching of edge maps, followed by area-based matching of at least coarse resolution edge maps of respective multi-resolution edge-maps, to efficiently and accurately provide the position of at least one instance of the model image. The invention can find a location of the model image within each test image of a sequence of test images, even when there are non-uniform brightness changes in the test image due to process steps or lighting changes. The postion of each verified instance of the model image within the test image can be provided to an alignment device, a measurement device, or any other device that can use such position information.

Abstract: An apparatus and method are provided for automatic visual inspection of crescent-shaped wire bonds. The apparatus includes image acquisition means for acquiring an image of a wire and an image of a crescent-shaped wire bond. A search model is used that includes a crescent-modeling portion, and can also advantageously include a wire modeling portion. Search functionality is used to find an image of the crescent-shaped wire bond, using the search model, so as to provide location information regarding the location of the crescent base, and shape-match information regarding the reliability of the found location information. The invention is useful for verifying the correct position of the crescent bond to ensure good electrical contact. In addition, the invention greatly simplifies the task of two other important automated visual inspection steps: wire location and crescent shape determination. Presence/absence of the crescent bond can also be determined using this information.

Abstract: Inspection of solder paste on a printed circuit board using a histogram of an image before printing (pre-application image) to normalize an image after printing of the printed circuit board (post-application image) is described. Existing lighting and optics used for alignment of the screen-printing stencil to the printed circuit board are used for the solder paste inspection.