Defects Of Fashion Clothing

The solution allows to carry out an automated analysis of the defects present on the garments.

Reco 3.26 is developing an industrial research and experimental development project of artificial intelligence techniques for detecting defects on clothing. The following types of defects have been identified for which it is possible to apply Computer Vision solutions for their automatic detection::

  • Color defects: During the production of the garment, the color applied to the fabric may appear different from that of the sample garment made available.

The Reco solution includes:

  1. The development of a lighting system and acquisition set up
  2. The analysis of the appropriate color space and chromatic distance
  3. The implementation of an artificial vision system for the matching / detection of chromatic anomalies (handcrafted computer vision technology).
  • Defective seams: During the production of the garment, the sewing process can lead to results that differ from those specified in the original paper

Machine learning-based computer vision systems will be used to detect defects. The Reco solution includes:

  1. The development of the lighting system and acquisition set up
  2. The creation of the training dataset
  3. The implementation of the defect detection subsystem
  4. The implementation of the defect classification subsystem (computer vision technology based on machine learning).

All Reco 3.26 systems can be easily integrated into the customer's pre-existing infrastructure

Implementation

Solution applied for the inspection of clothing for companies in the textile and fashion sector

Background

OVS