AI for Damage Inspection Final Presentations: Structural damage detection by drones

@Rijkswaterstaat partnered up with the FruitPunch AI community, @ProRail and the @Nederlandse AI Coalitie to improve remote inspection of dutch infrastructure using drone image data and increase the efficiency of inspection with the help of AI. This challenge worked with the data from preservation inspections a the open CODEBRIM dataset; and aimed to upgrade the existing model used to detect cracks in drone-images of bridges, locks, dykes and other public work and explore further avenues to improve remote inspection.

What will you see? 2 Teams experimenting with their own way to achieve the best performance of the model in detecting and classifying cracks, concrete rot and alkali-silica reaction.

1. Pierre Le Roux presenting for Team 1
2. For Team 2
** Luiza Pozzobon diving into binary multi-label and Learning Curve Analysis
** Kembo Varun taking over for Object Detection
** and Roshan Kotian showcasing the results for multi-class single-label Classification

More info on the Challenge: https://www.fruitpunch.ai/challenges/ai-for-damage-inspection