PhD Thesis - Industrial Vision System for Advanced Surface Inspection

TL;DR

PhD Thesis (Industrial Vision): Fundamental research on 3D acquisition technologies, illumination spectra, and novel imaging techniques for advanced surface inspection in battery production with an accent on multimodal AI algorithms for defect detection and 3D reconstruction. Focus on designing low-latency deep learning architectures, multi-sensor data fusion, and real-time integration into R&D production lines.

Grenoble, Auvergne-Rhône-Alpes

Company

hirify.global is a battery technology company building gigafactories and advanced manufacturing systems for electrode production.

What you will do

• Investigate 3D acquisition principles, illumination spectra, and novel optical configurations to optimize defect contrast on battery surfaces.
• Develop multimodal deep learning architectures for joint processing of 3D, 2D, multispectral, and thermal data, optimized for in-line inference on edge devices.
• Address AI challenges like class imbalance, scarce data, and generalization using self-supervised methods and synthetic data generation.
• Design real-time data fusion pipeline integrating vision system with line-scan cameras, IR thermography, and other sensors.
• Prototype and integrate the full acquisition-to-decision pipeline on R&D production lines, collaborating with process and equipment teams.
• Conduct market study on hardware, software, and provide buy-vs-build recommendations.

Requirements

• MSc or Engineering degree in computer vision, optics, image processing, or related field.
• Fluent English required
• Strong programming skills (Python, C/C++) and experience with deep learning frameworks.

Nice to have

• Knowledge of optics, 3D metrology, or acquisition systems.
• French appreciated.