Development of an industrial method for automatic evaluation of technological parameters and grain classification using image analysis
Due to the ever-increasing demands placed on producers regarding the quality of the cereal grain produced, there is a need for an objective, reproducible, fast and low-cost quality assessment system. These conditions can be met by a system based on multispectral vision systems. The aim of the project is to develop an automated comprehensive technology for assessing the quality of grain mixtures.
The system will be built using advanced vision systems. It is envisaged to develop proprietary algorithms analysing images of objects, on the basis of which the actuators will divide the mixture into individual quality groups. The performance of the developed algorithms and executive module will be verified under industrial conditions. The results obtained in the project, will be used to build an industrial system, which will have applications in the grain industry, including malting plants, grain elevators, mills and manufacturers of sorting and cleaning equipment.
Project leader: dr hab. P. Zapotoczny, Uniwersytet Warmińsko-Mazurski w Olsztynie
Leader in Lodz University of Technology: dr hab. Piotr Szczypiński
Years: 2015 – 2018
Project type: NCBiR, PBS Grant; applied research
Key publications resulting from the project:
Computer vision quality assessment of barley kernels, Izabella Korczyńska, Aleksandra Jarosik, Piotr M. Szczypiński, Piotr Zapotoczny, EFITA/WCCA/CIGR Conference, Poznań (referat) 3D reconstruction of barley kernels, Mateusz Owczarek, Piotr M. Szczypiński, EFITA/WCCA/CIGR Conference, Poznań 2015 (referat)
Master's theses:
Aleksandra Jarosik, Opracowanie obrazowej metody klasyfikacji ziarniaków jęczmienia o różnym charakterze defektów, 2015, magisterska
Dominik Dubiak, Opracowanie ontologii ziarniaków jęczmienia dla celów budowy systemu ekspertowego, 2015, magisterska