This projects uses machine learning algorithms to identify the damage in a composite material. The objective is to input the damage index image and obtain a segmentation of the damage region. For this, we will train different classifiers using image segmentation techniques.
The image below shows some pairs (damage_index, damage_region). We will use this to train in a supervised manner.
The jupyter notebook file knn.ipynb
goes over the implementation of a k-nearest
neigbourhs algorithm. We can see some results in the image below.
- Israfel Salazar -Initial work- israfelsr