Abstract: Displacement of a first block of pixels from a first to a second image of a scene is determined for applications of automated distance computation. The selection of corresponding blocks in the two images is performed and verified as follows. A first block is taken from the first image. A first region is selected from a set of first regions in the second image, so that a difference between an image content of the first region and of the first block is minimal among the set of first regions. A second block is taken from the second image, the second block containing at least part of the selected first region. A second region is selected from a set of second regions in the first image, so that a difference between an image content of the second region and of the second block is minimal among the set of second regions. A first displacement between the first block and the selected first region is compared with a second displacement between the second block and the selecting second region.

Abstract: A method of processing an image of an object to obtain a dimension-difference relating amount which relates to a difference of an actual dimension of the object from a reference dimension thereof, the method including the steps of producing a plurality of search lines each of which perpendicularly intersects a corresponding one of a plurality of substantially straight portions of an edge of the image of the object, and obtaining the dimension-difference relating amount based on a sum of a plurality of products each of which is obtained by multiplying a corresponding one of respective edge-point deviations of respective actual edge points where the search lines intersect the straight portions of the edge, respectively, from respective reference edge points on the search lines, in respective directions parallel to the search lines, by a corresponding one of respective parallel-direction distances of the reference edge points from a dimension center of the image of the object in the respective directions paralle

Abstract: A method of training a system to identify inspection sites on a circuit board is described. The method uses a priori information, which includes a region of interest, approximate sizes and approximate spacings of a plurality of nominal pad locations, and a pad count or an aperture count. A region is created which is associated with each one of the nominal pad locations within an image of a printed circuit board. A search tool is run to find pad candidates within each one of the regions. The pad candidates are filtered. The best pad candidates are selected from among each of the found pad candidates. The best pad candidates are averaged to provide an average, and are modified based on the average.

Abstract: A method involves training a system to identify inspection sites on a printed circuit board. The method uses a priori information, which includes a sample pad description. A training region of interest is created within an image of the printed circuit board. A search tool is run to find pad candidates within the training region. The pad candidates are filtered and false pad candidates are eliminated. The filtered pad candidates are averaged to provide an average, and are modified based on the average.

Abstract: The present invention provides a method for automatically cropping a scanned image having irregular boundaries. A pattern sheet is provided on the bottom surface side of the cover of the optical scanner. The pattern sheet is defined by multiple pattern units for serving as referential coordinates. When the scanner is initialized, the pattern sheet is pre-scanned to generate a pre-scanned image. The pre-scanned image of the pattern sheet will be analyzed to get the referential data, including the index, the referential intensity level, and the pixel-positions of each pattern unit. The referential data of each pattern unit will be stored in a database. When auto-cropping an area of interest (AOI) from a pre-scanned image, the pre-scanned image of the original will be logically divided into multiple document blocks according to the positions of the pattern units. Then, compare each document block with its correspondent pattern unit to determine if the document block contains AOI data.