DENKblatt

New AI level at DENKweit: analyse orthophotos!

Our image analysis AI DENKnet now also supports a setup that is crucial for many geodata-based use cases: the evaluation of orthophotos.

Orthophotos = high-resolution, georeferenced aerial photographs that are true to scale thanks to rectified perspective. Some of them cover several kilometres. These are taken by drones or satellites, for example, and are essential for applications in construction monitoring, inspection or land management. 

✅ AI training is performed on the individual, original aerial photographs – i.e. the raw images before they are assembled into an orthophoto.

✅ Prediction/evaluation then takes place on the finished, large orthophoto – where it counts.

A case study:

Our partner FlyNex supports companies in the GDPR-compliant management of aerial image data, particularly drone use in an industrial context. The image data provided by FlyNex is used to train the AI in our DENK VISION AI Hub, while the analysis of the large orthophotos is carried out locally at FlyNex, directly via a provided Docker container including API.

As far as we know, there is currently no other system on the market that evaluates orthophotos.

🚀 A step that makes our Vision AI even more flexible, scalable and powerful. 

👉 If you have your own orthophotos or would like to process them, we look forward to hearing from you!  connect@denkweit.de  The application possibilities are HUGE.

The DENK VISION AI Hub has been updated for clarity and ease of use:

New design.
Faster loading times.
Improved (full-text) search in projects and networks.

The menu item ‘Models’ (the DENKnets) is now called ‘Model Manager’. Here you will find the DENKnets belonging to your projects. Here, you still have the option to add a new model to this project, rename an existing one, delete it, etc.

If you encounter any difficulties in the DENK VISION AI Hub, we will be happy to help you: support@denkweit.de.