Designed for programmers, software engineers and machine vision specialists capable of operating at the library routine level, this mvBlueLynx-compatible version of EconoCR enables users to quickly create standalone industrial OCR systems powered by factory floor proven industrial strength software that has been used extensively in a variety of laboratory and production environments since the late 90s. EconoCR’s library routine format also allows it to be easily combined with 1D bar and 2D matrix decoding software to produce more general-purpose video camera-based automatic identification solutions.
The characters to be read can be uniformly or proportionally spaced, dot matrix or continuous stroke, appreciably larger or smaller than those used in training, misaligned, rotated and may even be from different fonts. On-site training of any font of interest is accomplished in minutes.
Reading accuracy is virtually 100% on characters of relatively good quality. Similar performance is achievable on mediocre and in many cases poor quality characters when EconoCR’s “fielding” option is used to specify which characters are allowed to occupy each position in a properly constructed character string or other advanced options are activated that are specifically designed to deal with touching characters, slanted characters, extraneous marks, nonuniform backgrounds, and less than ideal lighting conditions.
Typical applications for EconoCR include the automatic recognition of characters and symbols that have been machine-marked (laserscribed, dotpeened/pinstamped, inkjet printed, laserprinted, electrochemically etched, etc.) on PCBs/substrates, glass lenses, molds & containers, electronic components and carriers, medical/dental devices and instruments, automotive parts, aerospace components, direct mail pieces and packaged goods, consistent with identifying, matching, sorting, tracking, or verifying these products as they move through various laboratory, manufacturing or material handling processes.