Computer Vision Quality Control
Objective
Incorporate vision-based language models for quality and visual inspections, providing impartial ratings and identifying upsell opportunities in cleaned facilities.
Challenges
Small technical team, so adding another database was not desirable, so we use the same SQL server already being maintained for source of truth.
Solution:
- Implemented smart phone based quality control computer vision system
- Created a natural language processing framework that allowed rules to be applied to the images
- Integrated with existing work order management system
- Built ranking and scoring system to provide feedback to the cleaners
Outcome:
The implementation of our computer vision quality control system transformed the client's quality control process, automating and well as standardizing the process for quality control and inspection.
Key Benefits:
- Significant improvement in defect detection accuracy
- Substantial reduction in quality control labor costs
- Notable decrease in defective products reaching customers
- Faster feedback for facilities monitoring and reporting
- Seamless integration with existing work order management system