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: This invention provides an apparatus and method for locating a curvilinear symbol by generating a grey scale model of a segment of a curvilinear symbol and interpolating curvature at the segment's periphery; reflecting this segment around the axes of symmetry to generate a full size model; generating full resolution subfeature models; assigning coordinate axes weighting to each subfeature model x and y axis; defining tolerance thresholds for deviations from an idealized curvilinear symbol; searching a digitized image for shapes corresponding to the subfeature models; assigning weights to scores returned for positions found corresponding to the subfeature models; producing weighted positions combining both coordinate axis weighting and score weighting; computing a best fit for the image found; comparing the positions found to tolerance thresholds and discarding those positions outside the thresholds; substituting and weighting alternate positions found, if any, to compute the location of the curvilinear s

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: A method and apparatus for locating curvilinear objects of variable scale in digitized images using feathered fiducials. The leathered fiducials are constructed by: defining a range of tolerated object size scale variations; generating an enlarged submodel within the range of tolerated scale variations and assigning background pixel values exterior to the enlarged submodel; generating a reduced size submodel within the range of tolerated scale variations and assigning foreground pixel values to the reduced size submodel; disposing the reduced size submodel within and tangent to the enlarged submodel at the application point; uniformly changing grey scale pixel values in a graduated manner from foreground to background between the reduced size submodel and the enlarged one. The curvilinear objects are then located by conducting searches using the constructed fiducials.

Abstract: A frontal illumination system includes a light source and a finite plurality of reflecting surfaces. The reflecting surfaces can be characterized as the inner surfaces of a set of conical frustums arranged to approximate the illumination coming from the inner surface of a hemisphere. The light source is located within or adjacent to at least one of the reflecting surfaces. It emits light that is reflected off opposing portions of at least one, and preferably all, of the surfaces. The system is capable of illuminating the object from a very large solid angle, even though it is disposed in a substantially short axial space.