Abstract: This invention provides a system and method for reading and decoding ID features located on a surface of a curved, sloped and/or annular object, such as a tire moving on a conveyor. A plurality of 3D sensors are operatively connected to a vision system processor. The sensors are calibrated by calibration parameters to generate a stitched-together 3D image of a field of view in a common coordinate space. A motion conveyance (e.g. a conveyor) causes the object and the 3D sensors to move in relative motion, and the conveyance provides motion information to the vision system processor. An ID finder locates ID features within a version of the 3D image and a decoder (e.g. an OCR reader) generates data from the ID features. The ID finder can locate a trained portion of the ID and the search for variable code elements at a known orientation relative to the trained portion.

Abstract: A computer-implemented method for use in machine vision systems is provided. The method can include receiving a source image and segmenting the source image into one or more segments. The method can further include receiving a selection of a first segment of the one or more segments associated with the source image and generating a first mask image, based upon, at least in part, the first segment. The method can also include determining at least one attribute associated with the first segment and normalizing a masked area of a runtime image using, at least in part, the at least one attribute.