Speakers
Description
Syllabus outline:
Use pretrained Convolutional Neural Networks (CNNs) to classify inserts. Hands-on: 120 minutes.
Visualization of the structure of a CNN architecture.
Training and test data visualization.
Estimation of the class of independent examples using a pretrained model.
Objective competences:
1. To observe in practice the application of a CNN in image processing.
2. To identify the building blocks of a CNN architecture.
3. To learn to use pretrained CNNs to get descriptors to classify the level of wear of milling inserts.
3.1. To get started with non-handcrafted descriptors.
3.2. To apply this knowledge to an Industry 4.0 problem.
Intended learning outcomes:
To classify inserts as having high or low wear using features extracted using pre-trained CNNs.
To identify the parts of an image processing system.
To know how to evaluate the performance of a machine learning model.
