AI Neural Network Transforms Quality Control Processes

This article appeared in the April 2025 issue of MiMfg Magazine. Read the full issue and find past issues online.

Software featuring artificial intelligence frees the time of employees by handling tedious, repetitive jobs, allowing people to focus instead on tasks that require human logic and oversight.

One particular area where AI can optimize processes and minimize human error is in quality control. AI can be trained to recognize defects in products or divergence from normal processes — allowing handlers to address the issue before the faulty product or process proceeds down the line. In addition to accelerating quality assurance processes, having AI solutions in-house rather than through a third party can reduce security concerns, and the auto-mation often can be implemented without extensive programming or technical coding expertise.

“Our mission is to take the power of neural networks out of the clouds and give it to the engineering community so that they can solve problems that were not previously solvable,” says Detect-It Founder Kevin Kerwin, noting that the software, therefore, provides more direct control for greater cyber-security.

Created by Oak Park-based Detect-It, the software can be integrated into any digital camera to detect errors and defects while ensuring processes are followed.

To “train” the audio-visual AI software, customers would take a video of a part in good condition, take a video of the part in bad condition and then draw boxes around errors. They would then use those videos to teach the software to recognize in live video feeds which parts are good and which are bad.

The technology also allows users to be less engaged with these repetitive quality control tasks and not make mistakes, Kerwin says.

“In one customer’s case, they repurposed 12 people to higher-function roles within the business, and their line went faster because bottlenecks no longer occurred during the inspection process,” Kerwin says.

Examples of how the software has been used — or could be used in the future — to ensure security, increase productivity and optimize processes include the following:

• OEMs use the technology for quality control at the end of production lines, such as identifying wrinkles on car seats.
• Detect-It’s technology can help train operators for various workstations within the same manufacturing facility, reducing the need of training new hires before production.
• A borescope integrated with the software has been placed on a sewing machine to perfect stitching.
• The automation assists businesses with inventory control.
• If a person accidentally places a product into an incorrect bin, Detect-It can recognize that mistake.
• In the medical industry, the software could ensure that PPE is worn properly and that equipment is used correctly.
• The agricultural industry could use it to reduce weed growth and recognize unhealthy crops before the problems expand.
• Senior health care professionals could train the technology to recognize when a particular patient is sick, such as by detecting changes in that person’s facial color, body movements and pain levels.

Detect-It offers revolutionary solutions for quality control without the need for advanced coding experience, all while saving manufacturers time and money.

Learn More About Our Gamechanger

Headquarters: Oak Park
Web: detect-it.ai

Have a manufacturing story to tell? E-mail communications@mimfg.org.