Case Studies

A selection of projects we have done.
Computer Vision

Automatic planning of solar systems

Our client is a company from the solar sector, which offers the planning, consultation and installation of solar systems for private households. The aim of this project is to automate the planning process of solar systems. Specifically, this involves preparing an individual offer for customers so that they can estimate the costs and benefits of the investment and get an aesthetic impression of the final result.
Goal

The aim of this project is to automate the planning process of solar systems.

Challenges

Recognition of small obstacles, different input quality, estimating (3-dimensional) roof dimensions, handling of special roof shapes, separation of different roof sides, interfaces.

Solution

The entire process was automated up to the visualization of the solar panels.

Computer Vision

Automatic spotting of artisanal and small scale mines

Artisanal and small scale mining (ASM) is responsible for 10% of global gold production, with an estimated market value of US $14bn, affecting the lives of 10 to 20 million miners (and an additional 80 million miners for other raw materials) and their families. ASM sites are located in more than 80 developing countries and associated with lower environmental and work safety standards. In order to enforce concession rights, labour legislation or to research migration patterns, it is imperative...
Goal

Develop a Computer Vision tool to automatically segment ASM sites in satellite images.

Challenges

Gather a sufficient set of training data.

Solution

We train a Deep Neural Network with a U-Net architecture. ASMSpotter can be developed into a cloud service, which analyzes satellite images upon request.

Computer Vision

Convective clouds detection

Our client here is Deutscher Wetterdienst (DWD). Part of their responsibility is to prepare weather reports for pilots in a system called METAR. We had been tasked with creating a model to assist the detection of convective clouds using satellite imagery to support other forms of detection.
Goal

Reliably detect Cumulonimbus and Towering Cumulus clouds using AI.

Challenges

The input data is in an unusual format that requires extensive preprocessing using domain specific knowledge before Machine Learning can be applied.

Solution

A neural network was trained using the satellite data as the input. We were able to obtain a 98% accuracy.

Natural Language Processing

Legal review of rental contracts

Our client is an online tenant protection club that offers its clients the legal review of their tenancy agreements by experienced partner lawyers. The aim of the project is to develop an intelligent software tool that supports lawyers in analysing contracts. In concrete terms, this involves the automatic evaluation of so-called "cosmetic repair clauses", which are intended to oblige the tenant to carry out regular renovations. They are often incorrectly formulated and therefore ineffective....
Goal

Develop an intelligent tool that supports lawyers in analyzing contracts using AI.

Challenges

Input: photograph of the rental agreement as PDF file
Output: explained and textually substantiated assessment of the effectiveness of the cosmetic repair clause.

Solution

We have developed a software which fully automates all steps from the evaluation of the quality of the document image to the decision on the admissibility of contractual regulations.

Talk to us

And find out where costs can be saved through process automation by Artificial Intelligence