AI4EOSC webinar series (2): Image processing with AI4EOSC

Monday Apr 8, 2024, 2:00 PM → 4:00 PM Europe/Madrid

Judith Sainz-Pardo Diaz (IFCA), Marcin Plociennik (PSNC), Álvaro López García (IFCA)

The second day of our workshop series, focusing on "Image processing with AI4EOSC." Dive into the exciting realm of image analysis and enhancement with the AI4EOSC platform, as we explore cutting-edge techniques for leveraging artificial intelligence (AI) to extract valuable insights from visual data. From image segmentation to superresolution, this workshop offers a comprehensive overview of advanced AI techniques and their applications within the European Open Science Cloud (EOSC) ecosystem. Join us for an immersive learning experience, where you'll gain practical skills and knowledge to unlock the full potential of AI in image processing.

 

All presentations will contain a questions & answers section at the end.

 

Webinar is live streamed in https://www.youtube.com/watch?v=JQOWmsEQANs

 

Questions can be asked via Sliddo: https://app.sli.do/event/986zicefK7NkZUx1LFNS6R/live/questions 

2:00 PM → 2:30 PM The impact of AI in scientific image processing

This presentation highlights the practical applications of AI in enhancing scientific imaging processes. We'll touch on key AI techniques that streamline tasks along the imaging pipeline such as denoising, segmentation, and visualization, demonstrating how they contribute to more efficient data processing and analysis. From processing large datasets with high-quality outcomes, to addressing dataset integrity and training bias considerations, from U-Nets to foundation models, this talk aims to provide a concise overview of AI's benefits and challenges in scientific imaging.

Speaker: Dr Deborah Schmidt (HELMHOLTZ IMAGING - Max Delbrück Center)

2:30 PM → 3:00 PM Image segmentation with AI techniques

Enter into the field of image segmentation with AI techniques, as we uncover strategies for partitioning visual data into meaningful regions. Learn about the state of the art models with a special focus on diffusion models oriented to do data augmentation segmentation. Learn about different segmentation algorithms and their applications, and explore how AI enables researchers to tackle complex segmentation tasks. This session highlights the diverse range of applications and models for image segmentation, showing a particular example concerning the detection of pavement defects for road condition monitoring.

Speaker: Saúl Cano Ortiz (University of Cantabria)

3:00 PM → 3:30 PM Enhancing satellite imagery with AI: superresolution

Learn how to enhance satellite imagery analysis with the power of AI superresolution techniques. Discover how AI4EOSC can enhance the spatial resolution of satellite images, revealing finer details and improving the accuracy of remote sensing applications. From urban planning to agricultural monitoring, learn how superresolution can unlock new insights and opportunities for leveraging satellite imagery within the EOSC platform.

Speaker: Dr Daniel García (ICMAN)

3:30 PM → 4:00 PM Train an AI4EOSC module (using YOLO)

Take your AI4EOSC skills to the next level by learning how to train a custom module using the YOLO (You Only Look Once) object detection algorithm. Explore the fundamentals of deep learning and computer vision, and gain hands-on experience with training and deploying AI models within the AI4EOSC dashboard. Whether you're interested in object detection, tracking, or localization, this session equips you with the tools and techniques needed to develop powerful AI solutions tailored to your specific research goals.

Speaker: Dr Enoc Martinez (UPC)