← All workCase 06 / 07
PJBUMI Heavy Engineering
Computer vision for oil and gas inspection
Defect detection and asset-health models for an inspection service used with oil and gas clients.
- Role
- Led model development
- When
- 2023 to 2024
- Status
- MLOps apprentice
The problem
Inspecting oil and gas assets is slow, manual work, and defects such as corrosion are cheapest to fix when they are caught early.
What I built
I led AI model development for the company's asset inspection service: defect detection and real-time asset-health assessment, including work on a Petronas engagement.
Alongside it I built InOutVision, a face-recognition attendance system with OpenCV, TensorFlow and Power BI dashboards, and a CNN that tracks stockpile volume, served with Flask.
Where it stands
The models were used in inspection services delivered to oil and gas clients.
Client work, so no inspection imagery is shown here.
- Python
- TensorFlow
- OpenCV
- Flask
- Power BI