When Perfection Matters: Using Photometry-based Imaging Systems for Cosmetic Surface Inspection
| By: Radiant Vision Systems
Advanced Vision—a new class of automated visual inspection technology—combines the automation benefits of traditional machine vision with the sensitivity of true human vision to detect even the most subtle cosmetic defects and ensure a flawless product.
Today, customers expect flawless electronic devices right out of the box. Scratched, dented, and otherwise imperfect products can result in damage to your brand reputation and future business. How can you ensure cosmetically perfect devices while keeping up with demand?
Human quality inspectors apply their acute visual perception and immediate qualitative judgment to spot imperfections, regardless of where or how unexpectedly they may occur. However, faster production rates make it difficult for humans to spot subtle defects consistently while keeping up with throughput demands. Automated (machine-based) inspection systems provide greater speed to keep up with production, traditional machine vision systems just can’t match the adaptability and visual sensitivity of humans.
Photometry-based Vision Systems
Photometry-based imagers are engineered to detect subtle variations in light. Traditionally applied for measuring brightness and uniformity over large spatial areas—like an illuminated display—these systems are correlated to human visual sensitivity. Radiant’s Advanced Vision approach uses photometry-based imaging to combine the speed and consistency of machine automation with the precision and adaptability of human vision. Our capabilities exceed those of traditional machine vision cameras when it comes to inspection of cosmetic defects on surfaces or particles and imperfections in glass. Like the human eye, Advanced Vision solutions detect anomalous features on surfaces and grade these defects based on their location, scope, and severity.
Scientific Imaging. Radiant’s visual inspection systems pair specialized optics and imaging technology with advanced algorithms to locate and characterize cosmetic defects. Our
imaging systems feature scientific-grade image sensors with extremely high resolution, dynamic range, and signal-tonoise ratio for precise and repeatable measurements. Image noise is carefully reduced by testing each sensor with optics, electronics, calibrations, and other optimizations applied, enabling our imaging systems to detect more surface variations than other high-resolution measurement systems.
Luminance variations are detected across a device surface using Radiant’s ProMetric Y29 (29-megapixel) Imaging Photometer (top), indicating surface defects like scratches or dents. Only a limited set of these luminance variations are detected by a 71-megapixel CMOS imager (bottom), missing one peak variation at around 0.08 meters that indicates a cosmetic scratch, which was detected by the ProMetric camera.
Radiant solutions include sophisticated software algorithms to locate and characterize randomly-occurring cosmetic defects on device surfaces, including unknown defects in unknown or unpredictable locations. By taking a complete spatial image of a device, our systems perform uniformity checks to detect hairline scratches, subtle dents, or even fingerprints across device surfaces. These systems can classify the severity of each defect against acceptable tolerances and determine pass/fail, just as humans use their judgment to determine a reasonable margin of error.
Just Noticeable Differences. Because Advanced Vision solutions rely on photometric software tools, they leverage the advantages of light measurement technology to detect and measure surface-level defects. One such advantage is the application of “just noticeable difference” (JND) to perform uniformity evaluation. JND is the amount of change over a surface area that will be noticeable to an average person at least half of the time. This value is used to qualify automated defects and correlate severity with human visual discernment.
Radiant photometric imagers use JND qualification to evaluate non-uniform areas in illuminated displays that would be deemed unacceptable by a human observer. As with illuminated displays, JND measurement can be applied to non-illuminated surfaces for detecting and grading “uniformity” issues on surfaces caused by scratches, dents, smudges, debris, and other defects.
Scratches on the back of a laptop are identified and qualified based on their JND values (severity as compared with the rest of the surface). The values shown in this analysis image provide contrast ratios relative to a background level.
Data Capture & Analysis. Unlike human inspectors, Radiant software applies numeric values to each defect to determine severity—quantifying defects by size, shape, location, frequency, and rate of occurrence and building trend analysis data that allow you to fine-tune processes, improve product quality, and increase production efficiency.
Flexibility. Built on Radiant’s automated visual inspectikon software, our Advanced Vision systems can simultaneously perform inspections for part orientation, assembly verification, fit & finish analysis, and label legibility, eliminating the need for a separate machine vision system to perform these inspections on the line.
The length and width of a dent can be measured against defined thresholds to determine if the product passes or fails
Eliminating Particle Defects on Surfaces
With extremely high imaging resolution and dynamic range, Radiant systems can also detect unwanted particles on surfaces, glass, and transparent layers. Dynamic range increases the number of grayscale values in each image, ensuring that subtle contrast variations are identifiable to inspect fine details. The specifications of Radiant cameras ensure that pixel-sized particles are clearly discernable from image noise, so they can be identified and removed before shipping.
TrueTest™ Measurement Image
Radiant automated visual inspection software detects particles on this metallic surface, which can cause unwanted variations in surface finish.
The Power of Advanced Vision
Used for cosmetic surface inspection, Advanced Vision systems provide all the benefits of automated inspection, including objective analysis and quantifiable results, while capturing the data needed for error tracking to improve production operations. Radiant software provides an extensive set of machine vision inspection tools for production-level monitoring and pass/fail test sequencing to enable manufacturers to automate any visual quality inspection process.
Radiant’s photometry-based systems can detect defects that may go unnoticed by human inspectors, and can also be used to inspect components with imperceptibly precise tolerances, such as keyboards (measuring gaps between keys, characters, etc.), glass & transparent layers (particle detection, air bubbles), and more.
Advanced Vision inspection addresses assembly verification and all aspects of appearance, fit, and finish.