Seth Software Sp. z o.o. has received a grant from the European Union for its “TubSensAI – Artificial Intelligence-supported monitoring system for potato quality in storage based on volatile substance identification and digital imaging” project.
The goal of the project is to conduct R&D and implement an innovative system for monitoring potato quality during the storage process, based on innovative methods for monitoring volatile substances and digital imaging, supported by AI and ML algorithms. By making separate use of digital imaging data and volatile substance identification data, the system will allow more effective monitoring of potato quality in storage and earlier removal of potatoes affected by storage diseases from warehouses.
The project will expand Seth Software sp. z o.o.’s portfolio in the area of IT systems dedicated to the agricultural and food industry to include a product innovation as a result of R&D, i.e. an integrated hardware and system solution for potato quality monitoring based on the identification of volatile substances and digital imaging supported by AI, addressed to potato producers and processors with their own storage facilities.
The project will help the Applicant to improve its position compared to its competitors, on a domestic scale at a minimum, in the area of IT solutions for the agricultural and food industry.
The project is expected to involve the following tasks:
- Formulating a technology concept, model testing, feasibility studies and developing an electronic nose based on volatile compound identification in the potato storage process (R&D module – industrial research)
Testing the volatile compound (VOC) profile of healthy and infested potatoes during storage – model testing;
Determining chemical predictors as indicators of potato spoilage by numerical methods;
VOC profile change tests – evaluating the effectiveness of the designated predictor(s) of potato spoilage (based on measurements under real-life conditions, i.e. in the storage room);
Verifying the concept of sensor-detector system technology for potato spoilage predictors;
Designing and testing sensor-detector system technology for potato spoilage predictors under simulated operational conditions.
- Formulating a technology concept, model testing, feasibility studies and developing a digital imaging technology for potatoes during storage (R&D module – industrial research)
Verifying the concept of a digital potato imaging system;
Model tests under simulated storage conditions of pathogen-inoculated potatoes using a hyperspectral camera. The tests will involve collecting images and data to develop digital imaging technology for potatoes during storage.
Building a demonstrator for a prototype digital potato imaging system;
Developing an algorithm for determining quality parameters from potato imaging based on data acquired during the modelling stage and testing under simulated operational storage conditions.
- Developing a system for integrating and harmonising data sources from storage facilities (R&D Module – industrial research)
Developing a concept for an integrated system of harmonised storage data;
Studying and developing methods for data transmission, collection and storage, conducting interoperability studies;
Developing dashboards and studying the performance and usability of their interactive features.
Studies in user experience optimisation, user behaviour and business benefit analysis;
Studying, developing and validating methods for data filtering, interference elimination strategy analysis;
Integrating an electronic nose and digital imaging into a storage data system;
Developing integrated classification methods for different usage scenarios of the TubSensAI system, fine-tuning the models produced in Task 1 and Task 2;
Studying the technology of harmonised storage data under simulated operational conditions.
- Pilot run of technology for an integrated system for monitoring potato quality during storage (R&D Module – experimental development)
Developing the TubSensAI platform as an MVP version of an integrated system for monitoring potato quality during storage in a real environment;
Pilot run of the integrated system for monitoring potato quality during storage in a real environment.
- Pre-implementation work (Pre-implementation work module)
Patent agency services, including fees, for the registration of exclusive trademark and patent rights;
Market research.
- Implementing R&D results (Implementation work module)
Server for the support of the TubSenseAI client platform (web user interface);
Centralised TubSensAI system monitoring platform.
The project will enable to implement the following:
a product innovation in the form of a new product: the TubSensAI system (a potato quality monitoring system based on AI-supported volatile substance identification and digital imaging);
a process innovation in the form of a fundamental change in the potato tuber quality control process.
Furthermore, the Applicant intends to take the following action as a result of the project:
applying for a patent for the integrated hardware & software solution featuring AI models developed as a result of the proposed project. There are plans to obtain patent protection for the solution through the registration of a EU protection right.
the Applicant intends to file one application with the European Union Intellectual Property Office (EUIPO) for the protection of the TubSensAI trademark.
#EUFunds #EuropeanFunds
Project value: PLN 7,098,648.32
Amount of contribution from the European Funds: PLN 4,933,438.15