Manufacturer Mirtec announces the first release of its AI-based smart factory automation solution Intelli-Pro, an advanced software and algorithm package specifically designed to improve the performance and ease of use of Mirtec's AOI systems. It consists of proprietary deep learning-based functional components for automated part search and teaching, parameter optimization, character recognition (OCR), foreign object detection (FOD), component placement inspection algorithms and automatic defect classification.
The auto-matching function with AI-based deep learning maximizes the efficiency and accuracy of AOI programming, reducing processing time by up to 90% compared to traditional manual teaching and by 50% compared to automatic teaching without deep learning. Untrained operators can also be used.
Mirtec's Optimum Inspection Tool (OIT) is an off-line software solution that performs automatic debugging and parameter optimization by analyzing multiple inspection results. After the teaching process with the Deep Learning Auto Matching and Teaching Tool, the programmer inspects the board and reviews the results for each defect to determine whether the component in question is defective or good. The software then suggests optimal parameter values.
The Optical Character Recognition (OCR) algorithm reads the text labeling on the components and recognizes whether the matching component has been installed in the correct position and orientation. Mirtec also uses AI-based deep learning here, which can also detect whether the text is damaged. This is used by the automatic character teaching function. The same applies to the Foreign Object Detection (FOD) algorithm, which checks the board for foreign objects, ignoring regular elements on the board.
Defect classification through deep learning and continuous storage of the instances found is also an important step in the inspection, as it prevents false alarms. In future, this could be done automatically, i.e. without the intervention of an operator.
