ChinaManufacturing2023 – ongoingTCL Technology Group Corporation
CSOT automatic panel-defect classification and repair
TCL reported panel-defect classification accuracy rose from 85% to 95%
Problem
CSOT needed to identify panel defects across different sizes and manufacturing processes.
Approach
CSOT applied AI to classify defects automatically. In 2023, it extended detection into defect assessment and repair, piloting the system on its t10 line before expanding to other large-panel lines.
Results
95%defect-classification accuracy, up from 85%, reported by TCL
Takeaway
The reported accuracy measured defect classification, rather than the success rate of repairs.
Categories
- Size
- Large enterprise
- Industry
- Manufacturing
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Computer vision
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
[1]TCL11 Dec 2025
TCL reported classification accuracy increased from 85% to 95%. It described the repair system's rollout from t10 to other large-panel lines.
Full details are in the original sources.
Added 1 Nov 2026 · Updated 1 Nov 2026