Deep-image-processing

Hands on for academic lecture

This lecture focus on the use of codes to run and assess deep model. We are interested on image classification and image segmentation. In view of the range of existing models, we use a classical architecture: CNN and UNet.

Image classification

Image segmentation

Data :

• Nuclei data
• BCCD

Object detection and Foundation model

Data :

• Apple

Liens utiles :

• DINO
• Yolov11
• Hyperparamètres
• Prediction

Deploying model with Napari for public usage

Nous allons travailler sur Napari. Si vous êtes muni de votre PC portable, vous devez installer anaconda puis napari (consignes d'installation dans l'exercice 1).

Conclusion