Automated Grading Of Bundles Steel Wires Using AI-Based Machine Vision Technology

By September 2, 2021September 16th, 2021Steel Industries Case Studies, Use Cases
Steel Fiber Grading Inspection- Automated Grading Of Bundles Steel Wires Using AI-Based Machine Vision Technology

CLIENT/INDUSTRY BACKGROUND

The client is a leading steel manufacturer and has a presence across more than 10 countries. They produce a substitute product of rebars used for reinforcement in construction.

WHERE STEEL WIRES/FIBRES ARE USED?

Steel wires (Steel Fibers) are used as a substitute for rebars for the reinforcement of concrete. These steel wires are closely packed for uniform distribution of load whereas gaps can reduce the load-bearing capacity at that particular point.

CLIENT REQUIREMENTS

To classify the grades of steel wire bundles, based on the gaps between the wires.

Steel Fiber Grading Inspection

CLIENT’S PROBLEMS

  • There is no automation in place to stop the line exactly at the time the incorrect bundles are being produced.
  • Incorrect grading mechanism

PROBLEM IMPLICATIONS

Customer complaints and returns are the two major implications. Larger gaps in the steel fibers cannot distribute a load of concrete equally.

CURRENT PROCESS

The inspection is being done by the operators deployed at multiple lines. The operators randomly pick up a bunch of products and visually inspect them with their naked eyes. Through visual inspection, they decide the grade of the steel fibers and segregate them accordingly.

SOLUTION USING MACHINE VISION

A solution was developed to acquire the images of all grades of steel wire bundles. The acquired images are trained in the AI platform (Qualitas EagleEye® Cloud) with the help of annotations. A pre-trained deep learning module for ‘Classification’ is used for developing the models.

SETUP

A machine vision setup will be installed at the inspection station (before packaging). The setup includes a monochrome camera of 2.3 megapixels having 15 FPS to capture the images of the steel fibers. The distance b/w the parts (to be inspected) and the lens i.e. working distance would be 350 mm and therefore a 16 mm lens will be configured. 150×300 mm sq. will be the Field Of View (FOV). Red dome light will be used to illuminate the area of interest (where the parts will be placed). Red lights are used to get even reflection on the metallic surface for best image acquisition and processing.

HOW IMAGES ARE TRAINED IN QEP (QUALITAS EAGLE-EYE® PLATFORM)

Steel Fiber Grading Inspection 1

CONCLUSION

  • The potential benefits of deploying a machine vision system will be –
    Inspection of each product is possible with an inspection accuracy of approx 98 percent.
  • With the help of PLC (Programmable Logic Controller), the steel fibers can be rerouted to their corresponding grading conveyor.
  • The return rate could be completely reduced (almost by 99 percent).
Get In Touch With Us


Leave a Reply

Schedule A Demo
close slider


SCHEDULE A CALL