Optoelectronic sorting system for the highest precision in grape processing — developed in a ZIM research project with the Fraunhofer Institute IOSB and Hochschule Geisenheim.
The ZIM project
In a ZIM research project funded by the German Federal Ministry for Economic Affairs and Energy, ARMBRUSTER Kelterei-Technologie developed an optical sorting system together with Hochschule Geisenheim and the Fraunhofer Institute IOSB in Karlsruhe — with the goal of adapting the optical components and image processing so that, alongside sorting out foreign objects, berries can also be sorted by various ripeness criteria.
Objectives
The aromatic diversity of the individual quality grades in wine is to be enhanced through targeted berry selection by quality-relevant criteria such as Oechsle content, acidity, nitrogen content, rot, and whole vs. damaged berries. Mash and must quality should generally improve by removing unwanted particles (MOG — material other than grapes).
Further goals: optimised gentleness during grape handling, complete material stabilisation on a smooth conveyor belt, simplified cleaning compared with existing systems, and intuitive software so that even non-experts can achieve excellent sorting results.
Research
In initial trials, image capture with visible light proved capable of detecting differences in berry ripeness. A further approach is the combination with light in the near-infrared range (NIR). Spectral studies to identify relevant wavelength ranges serve as the basis for development.
Image captures in the visible and near-infrared range were carried out by the Fraunhofer Institute IOSB and analysed using multivariate methods. In combination with the laboratory work of Hochschule Geisenheim, correlations between wavelengths and quality parameters were derived.
Project partners
Fraunhofer Institute IOSB — scientific work on image capture and image processing of the optical sorter, carried out by the staff of the Institute for Optronics, System Technologies and Image Exploitation in Karlsruhe.
Hochschule Geisenheim — collection of quality-relevant data on berries throughout ripening, to relate optical, physical and analytical data and to derive correlations between quality parameters and wavelengths.
ARMBRUSTER Kelterei-Technologie — development and execution of the optical sorting system. Focus areas: mechanical selection by ripeness parameters, reducing air intake and juice misting, calmed material feed, singling and accelerating berries to approx. 3 m/s, cleaning, and user-friendly quality control.
Production-ready 2016
Based on the insights gained in previous trials, the GrapeSort optical sorter was developed to production maturity in 2016 and put into practice in California. Sorting by ripeness parameters was deliberately omitted; the focus was placed on separating foreign objects — never before has there been such high precision.
The high-resolution NIR camera precisely detects defects such as the early stages of shrivelled berries. Illumination via custom LED lights (alternating white and infrared LEDs) and the bespoke software of the Fraunhofer Institute IOSB meet the high market demands for product quality and safety.
Easy access through open housings; belt removal from the conveyor with a pneumatic quick-release mechanism simplifies and accelerates cleaning and maintenance.