Optical character reader

Abstract

The invention provides an optical character reader (OCR) utilizing prioritized horizontal OR ranges for enhanced accuracy and ease of use. A light source illuminates a price tag or the like on which characters have been imprinted generally along a horizontal direction of the sheet. A lens system images the characters onto an image sensor having a field of vision covering a plurality of characters, the image sensor including a plurality of photo-electric conversion elements arranged in a matrix. A binary coding circuit digitizes outputs of the photo-electric conversion elements, which define picture elements (pixels), of an image in the field of vision of the image sensor, into binary signals representing black and white picture elements. A hand holdable scanner incorporates the light source, image sensor, lens system and binary coding circuit. An identifying circuit identifies characters from binary-coded data. The OCR utilizes at least two horizontal OR ranges in the field of vision of the image sensor. When the horizontal OR range of higher priority detects a character line, the result of character line detection is output. When none of the horizontal OR ranges detects a character line, it is determined that there is no character line. The character line detecting operation of each of the horizontal OR ranges is the same as that in the conventional OCR. However, in the OCR according to the present invention, the horizontal OR ranges are spread horizontally. Therefore, even in the case of a price tag which is long enough to have characters arranged along one of its sides, the character line or lines can be positively detected. Since at least two horizontal OR ranges are provided in the field of vision of the image sensor, therefore the pattern of the background of the price tag and the characters on the price tag can be distinguished from each other. This means that the OCR can be used for relatively small in size price tags. The OCR conserves electric power when it is not actively 'reading' data by suspending its character identifying operations when not needed. Lines having no characters are identified and, in response, character identifying operations are stopped.