Deep Learning and Computer Vision in Health (6cr)

Objective

The student is able to
- Identify an image's representation across various color spaces within the frequency domain and execute common image processing operations.
- Extract and analyze fundamental features from an image, design a convolutional neural network (CNN) architecture, and create an automated learning system employing traditional algorithms for image content classification.
- Develop comprehension of the standard architecture of a convolutional neural network (CNN) and its operational principles.
- Solve a moderately intricate image classification problem involving deep learning algorithms for the identification of image objects. Utilize transfer learning or fine-tuning techniques based on pre-trained CNNs.
- Apply deep learning algorithms to identify image objects and autonomously generate multimedia content in healthcare applications. Implement these algorithms using the open CV library and TensorFlow.