Recognition Science has pioneered advanced techniques for recognizing manufactured parts, configuring camera and lighting systems, developing high-precision calibration software, and creating algorithms for defect detection in automated inspection. Achieving high-quality recognition algorithms is crucial, but to maintain the very low false alarm rates necessary for an efficient machine vision inspection system, precise illumination and exposure control are also essential.
Optimized Inspection Systems
In automated inspection systems, various factors such as camera resolution, lens distortion, camera perspectives, pointing accuracy, lighting colors, evenness, color temperature, polarization, and dynamic range must be meticulously considered. This holistic approach ensures accuracy and speed within realistic cost constraints. We have created routines to assist in automated board inspection for the presence or absence parts, as well as the precision of part placement, ensuring comprehensive quality control.
System analysis for initial design to full prototyping have been created for automation of medical kit assembly, lumber evaluation, extrusion dies databases update are among our many diverse projects.