Short description
The lecture introduces fundamental and modern methods of computer vision with a focus on industrial applications. Students learn how image data can be processed and analyzed using deep learning methods for classification, object detection, segmentation, and anomaly detection. In addition, classical computer vision methods for geometric measurement are introduced. Particular emphasis is placed on combining learning-based and model-based approaches to develop accurate, robust, and interpretable solutions for industrial inspection tasks.
Module content
Part 1: Modern Computer Vision
- Image acquisition, preprocessing, and preparation of image data
- Fundamentals of deep learning for computer vision
- Image classification and object detection
- Semantic and instance segmentation
- Anomaly detection
- Model training, transfer learning, and performance evaluation
Part 2: Geometric Measurement and Hybrid Approaches
- Camera models, calibration, and imaging geometry
- Classical image processing methods for geometric measurement
- Edge, contour, and shape extraction
- Geometric fitting and dimensional measurement
- Measurement accuracy and uncertainty
- Integration of learning-based and classical methods for industrial inspection