Project

Relats

Intelligent wiring inspection using machine vision

Relats

Our process

How it was done, step by step.

It’s continuing to decrease: the block remains on screen, and the project progresses phase by phase, with the actual work carried out by i-mas in each phase.

Phase 1 / 4The project
Project update
Fase 01 · The project

Relats enables the automatic, precise, and real-time detection of defects in cable jackets.

From manual inspection to industrial automation, thanks to a system based on machine vision and Deep Learning, which can identify imperfections in various types of cabling without human intervention. This is a system developed for demanding industrial environments with high production rates.

Deep Learning Connections
Fase 02 · The challenge

Detecting defects in cable covers with maximum reliability.

The client presented the challenge of designing a device capable of inspecting and detecting defects in the protective covers of various types of cabling during the production process.

The aim was to replace traditional manual visual inspection methods with an automated system that would ensure continuous, accurate, and repeatable quality control, regardless of the cable’s diameter or type.

Deep learning system for Relats
Fase 03 · Development

Applying Deep Learning to industrial inspection.

To achieve optimal results, a system was developed based on machine vision technologies, machine learning, and deep learning, which are capable of learning from real data and improving their performance over time.

The system analyzes the cable’s surface in real-time and detects defects in its coating, automatically classifying them without the need for human intervention. The solution has been integrated into a robust industrial unit, designed to operate continuously on the production line.

Inspection using deep learning in Relats
Fase 04 · Final product

Automatic cable wiring inspection at high speed

The final device allows for inspection of cables of various diameters, and operates at detection speeds of up to 120 meters per minute, ensuring consistent quality control without disrupting production.

The solution offers: automatic detection of defects, reduction of human errors, increased reliability in the process, improved traceability, and increased productivity.

The system provides a visual interface that displays in real-time the images captured by the cameras and any detected defects, making it easier for the operator to monitor the process.

Relats represents a significant advance in the automation of quality control within cable manufacturing processes, utilizing artificial intelligence for the industry.

Final product
Sector
Industry

Let's talk

Do you have a specific product in mind?

Tell us what you’re hoping to achieve and where you are in the process. Our product design team will respond to you within 24 working hours.