"Torn Wallpaper Detection"... DL E&C Develops AI Defect Inspection System

Development of Photo-Based Defect Inspection System Using AI Filed for Patent
Starting with Wallpaper Defects, Plans to Expand Application Scope

[Asia Economy Reporter Onyu Lim] DL E&C (formerly Daelim Industrial) announced on the 1st that it has developed an AI-based photo inspection system for defect detection and filed a patent for it. The DL E&C research team independently developed the defect inspection system through data analysis and plans to gradually expand its application scope and improve its accuracy.

The newly developed defect inspection system automatically detects and identifies the location of wallpaper tears by photographing the wallpaper. This allows workers or site managers to replace the difficult task of visually inspecting wallpaper defects with simple photo capture. It is expected to reduce the manpower and time required for defect inspection and minimize defect occurrence before residents move in.

Since early this year, DL E&C has started developing the AI-based defect inspection system using in-house developers and defect data. The company analyzed and trained the AI with 60,000 cases of wallpaper defects collected so far. After three months, they developed a system capable of independently determining the presence of defects and indicating their locations.

Currently, DL E&C is refining the system to accurately detect the presence and extent of wallpaper contamination caused by leaks or condensation, in addition to tears. In the future, they plan to expand the application scope to cover various finish defects that can be visually identified.

A DL E&C representative emphasized, “It is significant that we built the AI defect inspection system using our own capabilities and data,” and explained, “We expect to improve quality and productivity by further refining the system and expanding its application scope.”

This content was produced with the assistance of AI translation services.