Automatic optical inspection of regular grid patterns with an inspection camera used below the Shannon-Nyquist criterion for optical resolution
| Autor: | Flávio P. Ferreira, Boris P. J. Bret, Paulo M.F. Forte, Paulo Eduardo Reis Felgueiras, Michael Belsley, Eduardo J. Nunes-Pereira |
|---|---|
| Přispěvatelé: | Universidade do Minho |
| Jazyk: | angličtina |
| Rok vydání: | 2017 |
| Předmět: |
010302 applied physics
Automated optical inspection Quality Control Shannon-Nyquist Science & Technology business.industry Computer science ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Filter (signal processing) 01 natural sciences fractional defects Regular grid 010309 optics Automated X-ray inspection Nyquist stability criterion 0103 physical sciences AOI Optical Metrology Computer vision Artificial intelligence Image sensor Automatic Optical Inspection business Image resolution |
| Popis: | An Automatic Optical Inspection (AOI) system for optical inspection of imaging devices used in automotive industry using an inspecting optics of lower spatial resolution than the device under inspection is described. This system is robust and with no moving parts. The cycle time is small. Its main advantage is that it is capable of detecting and quantifying defects in regular patterns, working below the Shannon-Nyquist criterion for optical resolution, using a single low resolution image sensor. It is easily scalable, which is an important advantage in industrial applications, since the same inspecting sensor can be reused for increasingly higher spatial resolutions of the devices to be inspected.The optical inspection is implemented with a notch multi-band Fourier filter, making the procedure especially fitted for regular patterns, like the ones that can be produced in image displays and Head Up Displays (HUDs). The regular patterns are used in production line only, for inspection purposes. For image displays, functional defects are detected at the level of a sub-image display grid element unit. Functional defects are the ones impairing the function of the display, and are preferred in AOI to the direct geometric imaging, since those are the ones directly related with the end-user experience. The shift in emphasis from geometric imaging to functional imaging is critical, since it is this that allows quantitative inspection, below Shannon-Nyquist. For HUDs, the functional detect detection addresses defects resulting from the combined effect of the image display and the image forming optics. This research is sponsored by the Portugal Incentive System for Research and Technological Development. Project in co-promotion n 002797/2015 (INNOVCAR 2015-2018). info:eu-repo/semantics/publishedVersion |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|