News
DataTeam, a student research club based at the Faculty of Geology, Geophysics, and Environmental Protection, is developing a mobile app that detects mould on food. This innovative solution will help people with visual impairments in their daily lives.
The project is aimed at blind and visually impaired people, as they find it difficult to recognise mould on food, particularly if it cannot be detected by touch or smell. In such cases, they have to rely on help from third parties. The mobile app that students are working on will use pattern analysis.
The group of students studying Data Science came up with an idea of an application based on machine learning that could help identify mould.
Provided with a single picture of a selected product taken with a mobile phone, the model predicts whether there is mould on it and provides a voice response indicating whether the food is fit for consumption.
“We obtain the data using a machine learning model, and it is sourced from a platform called Kaggle. We have now developed a mobile app. We are now focusing on ensuring that the app works both with and without an internet connection. In the version without internet access, we’re aiming for lighter models, whilst at the same time trying to improve the prediction accuracy as much as possible. The health and safety of the app’s users is what matters most, which is why we place great emphasis on ensuring that these forecasts are as accurate as possible,” says Cyprian Szot.
The students have established a cooperation with the University of Agriculture. Thanks to Professor Aleksandra Duda, Dean of the Faculty of Food Technology, they were able to obtain a sample of the necessary photographs of food products covered in mould.
The next important step was to consult with the creators of interactive exhibitions at Womai in Krakow, where the young researchers discussed the features their app should include to cater to people with visual impairments.
At present, the students are focusing on integrating a test model that allows them to take photos within the app, so that they can capture them and use them for predictions.
In addition, the basic interface of the app is complete, as is the voice guidance system. The developers are testing how the TalkBack and VoiceOver systems work on Android and iOS. The app should be ready in September. It should also be noted that the range of products available in the app will be limited.
The team is made up by the following students: Kacper Potaczała, Filip Zagała, Jakub Skarbek, Bartek Tasak, Bartłomiej Żurek, Jakub Hus, Kamil Pyzik, Natalia Malcherek and Cyprian Szot. The club’s supervisor is Dr Mateusz Zaręba.
Photograph: Ewa Biśta
New fields of study at AGH University 
AGH University signs a Polish–Italian letter of intent in the field of nuclear energy 
DataTeam is developing an app that detects mould 
AGH University cheerleaders are world championship runners-up 
Our researcher on the podium at FameLab Poland 2026 
AGH University stays high in the Best Global Universities ranking 
AGH University’s rise in QS World University Rankings 2027 
AGH Racing premiers new electric car 
