Automated Visual Inspection

What is Automated Visual Inspection?

Automated = 6-axis articulated arm to perform a manual task automatically.

Visual Inspection = the inspection of an object's surface to detect random surface defects such as dents, nicks, pits, scratches, tooling marks, etc.

Blade Dove Tail Inspection

AV&R Vision & Robotics' team started acquiring knowledge in Visual Inspection more than 15 years ago, but the official birth of the Centre of Excellence occurred in 2005 with a system inspecting finished automotive roller bearings. The technique was rapidly transferred in the fields of aerospace and energy to inspect compressor and gas turbine blades.

Today, AV&R's systems use machine vision, algorithms completely adapted by its team of engineers and robots mimicking human tasks to perform completely automated inspections on gas turbine parts such as blades, variable vanes and blisks / IBRs.

Benefits

Platform Blade Inspection

Here are some benefits of an Automated Visual Inspection system:

Elimination of human subjectivity in visual inspections;
Complete 100% inspection with guaranteed results;
Feedback on the process quality;
Higher production rates;
Elimination of the possibility of a non-conform part being shipped;
Ergonomic design to enhance operator safety and the work environment;
Increased customer satisfaction.

Automated Challenges Overcome

To automate high level human tasks, it requires considering many variations in the product's shape and in the inspection conditions for which humans are usually not even aware. For an automated machine, a small lighting variation or process variation can be seen as anomalies leading to a mismatch between the customer's actual quality evaluation and the system output, consequently rejecting the part.

Special algorithms are needed to have intuitive results from a human-like point of view and minimize false defect detection. Here are some fluctuations taken into account:

Normal variation in the shape of a part;
Surface finish;
Color variation of some processes (such as heat treatment);
Positioning of some known features (not to be considered as anomalies);
Multiple part models to inspect through the same system.

Furthermore, to ensure that every portion of a part has been inspected, a special handling and image processing synchronization is needed.

2D Visual Inspection

Automated Visual Inspection of complex surfaces can be applied everywhere a human is presently manipulating a part to look all around it, searching for anomalies. AV&R's two dimensional (2D) Visual Inspection systems detect every defect a conscientious human eye can see such as handling defects (dents, nicks, scratches, chips, etc.) or process defects (polishing marks, grinder marks, inadequate shot peening, coating chip, etc.).

The inspection systems can also recognize 2D characters and ensure a feature is positioned properly or is within the required dimension. This inspection also includes surface profile analysis and the software tools use caliper and morphologic analysis.

AV&R has designed systems which can validate assembly with its given capability to archive images of the assembly or specific components and to perform optical character recognition to validate that a specific component is present and conform.

Here are examples of 2D Visual Inspection projects:

Inspection of turbine and compressor blades, variable vanes and blisks / IBRs (looking for defects);
Validation of cooling hole presence;
Optical Character Recognition (OCR);
2D matrix dot marking;
Assembly validation of complete aircraft engines.

3D Defect Characterization

AV&R's three dimensions (3D) inspection system characterizes surface defects found by the 2D system such as dents, nicks, scratches, tooling marks, etc.

Finding defects is not the end of the inspection process; following, comes the evaluation whether the part can be reworked or not. To make this decision, information on the shape and topography of a defect are necessary and require a 3D data acquisition system*.

AV&R's 3D defect characterization systems:

Can distinguish a localized defect from a depth of ± 20 microns / 0.8 thousandth of an inch (i.e. depth of a dent);
Finds the profile of a defect;
Helps with defect classification (pit, nick, dent, etc.);
Helps decide if the part inspected can be reworked or not, based on dimension and depth of the defect;
Improves repeatability of measurements compared to human decision.

AV&R's 3D Automated Visual Inspection systems are currently used to inspect gas turbine blades, variable vanes and blisks / IBRs used in aerospace and energy.

* The 3D data acquisition technology was developed by the National Research Council of Canada (NRC).