Product

Use Case

Resource

Company

Partner

Contact us

Assembly inspection solution

AI-based deep learning assembly inspection
comprehensively detects everything from missing parts to fastening defects.

Neurocle's Solution Clients

Verify assembly quality with an assembly inspection solution to enhance process stability.

Check the position, orientation, and connection status of the parts to automatically determine any assembly omissions or misassemblies.

How to apply the inspection

-

See real examples

-

Type

Subtitle

Title

Type

Subtitle

Title

Type

Subtitle

Title

How to apply the inspection

-

See real examples

-

Type

Subtitle

Title

Type

Subtitle

Title

Type

Subtitle

Title

Main features

The powerful differentiator provided by the Neurocle assembly inspection solution.

Precisely inspect complex assembly conditions.

The position accuracy of components, whether they are fastened, and the assembly condition can be assessed simultaneously in a single inspection, allowing for a consistent inspection system to be established even in complex assembly processes.

Precisely inspect complex assembly conditions.

Configuration of specialized deep learning models for assembly processes

By combining various deep learning models, it is possible to design an optimized assembly inspection environment according to process, vehicle type, and product structure.

Configuration of specialized deep learning models for assembly processes

By combining various deep learning models, it is possible to design an optimized assembly inspection environment according to process, vehicle type, and product structure.

Field Application Case

Application cases of assembly inspection solutions created with rich on-site experience.

Pipe welding defect inspection

문제점

The inspection site for determining the quality of the welds on the pipes is enclosed, and there were significant issues with detection accuracy due to external influences such as dust, foreign materials, and glare.

솔루션

Detected the welding area and simultaneously determined the quality of the weld as acceptable or defective within the detected area. The robot blocked the movement of any pipes identified as defective, allowing for cutting and re-welding to proceed. This not only reduced unnecessary human and time costs but also ultimately increased the accuracy of defect detection to 99%.

Pipe welding defect inspection

문제점

솔루션

Automobile body assembly inspection

문제점

For 10 to 15 seconds until the vehicle structure moves, a person inspected the number of bolts in all inspection areas. The accuracy varied depending on the type of vehicle and the inspector's capabilities, negatively affecting the consistency of product quality.

솔루션

The assembly inspection solution analyzes the acquired images to determine assembly anomalies, and the method was changed so that only those vehicles judged to be abnormal are confirmed by the operator for final judgment. After using a deep learning model in the first phase, the detection accuracy improved from 97% to 99.8%, an enhancement of approximately 2.8%.