Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Glass Edge Quality Detection using Image Recognition
Image recognition is successfully being applied in many areas, such as self-driving cars, medical diagnosis, and face recognition. In this project, you will work on an image recognition model to accurately predict the glass edge quality after grinding.
The images show the glass surface from above. A horizontal line defines the glass edge, and based on the smoothness of that line, the glass quality should automatically be detected. Because of dirt on the glass and varying image qualities, it is necessary to preprocess the images appropriately. Then, the processed images should be used to train an ML/AI model to identify the grinding quality. Some images with their corresponding labels already exist and will be used as a first step. Later, it might be necessary to conduct additional experiments to gather more data.
Tasks:
- Literature review
- Exploratory analysis of the available data
- Image processing
- Train and validate ML models
- Presentation
- Thesis
Requirements:
- Reasonable understanding of machine learning (in particular, deep neural networks) or very strong interest in learning ML.
- Programming experience in Python and optimally experience with a deep learning framework such as PyTorch.
- Interest in image recognition and image processing.
The images show the glass surface from above. A horizontal line defines the glass edge, and based on the smoothness of that line, the glass quality should automatically be detected. Because of dirt on the glass and varying image qualities, it is necessary to preprocess the images appropriately. Then, the processed images should be used to train an ML/AI model to identify the grinding quality. Some images with their corresponding labels already exist and will be used as a first step. Later, it might be necessary to conduct additional experiments to gather more data.
Tasks: - Literature review - Exploratory analysis of the available data - Image processing - Train and validate ML models - Presentation - Thesis
Requirements: - Reasonable understanding of machine learning (in particular, deep neural networks) or very strong interest in learning ML. - Programming experience in Python and optimally experience with a deep learning framework such as PyTorch. - Interest in image recognition and image processing.
The goal is to implement, train, and validate an image recognition model that can detect the glass edge quality based on a single image. If successful, this model can be used in production to automatically detect the quality during manufacturing.
The goal is to implement, train, and validate an image recognition model that can detect the glass edge quality based on a single image. If successful, this model can be used in production to automatically detect the quality during manufacturing.