Build a Machine-learning/ Image-recognition Prototype

Problem Estimate max. # of solar panels that can be fit on a roof
Input

Normed satellite image(s) of the roof

Output Scalar (-> max. # of solar panels that can be fit on a roof)
Available Data

About 1.000 labeled images (constantly growing), where the label may be the roof pitch, the gross or net size of the roof.

Task Develop prototype (preferably in Python but generally language-agnostic) up to the optimization part (see last sub-task below).

Sub-tasks to solve

(given current understanding of the problem)

  • Find boundaries of the roof

  • Estimate roof pitch (=Dachneigung)

  • Find objects on the roof such as chimney, windows, etc.

  • Estimate net size of the roof  

  • Max. # of equal-sized solar panels that fit onto the roof 

Who are you?

A ML-Aficionado with a thorough understanding of neural nets and some experience in the field of image recognition. Most importantly, you feel up to the task.

Who are we?

 A young, Berlin-based software company specialised in solving business problems through applied mathematics. 

Interested?

Drop us a few lines to info@datenschmiede.org. We are looking forward to get to know you!