Technology for Wood and Natural Fiber-Based Materials

Research Project

ASKIVIT – Material-efficient lightweight materials made from hardwood through wood-species combination and fiber reinforcement

Bulky waste contains valuable raw materials. Due to the quantity and variety of the bulky waste that is accumulated, manual sorting is very laborious. In collaboration with project partners, we are developing a solution for the automated sorting of bulky waste in order to recover wood, wood-based materials and non-ferrous metals which is based on various image-acquisition and image-processing methods as well as artificial intelligence. We are thereby helping to ensure that a higher proportion of raw materials from bulky waste can be recycled. This conserves resources and improves economic efficiency.

The process development is initially being focused on wood and wood-based materials. Furthermore, tests are being carried out to determine whether the same process can be used to pick out non-ferrous metals parallel to wood. The reason is that whereas magnetic metals are relatively easy to separate from bulky waste, this does not function in the case of non-ferrous metals.

In order for the sorting processes to work in practice, the initial situation is first clarified at the Fraunhofer WKI: We research the processes currently available in German municipalities for the collection of bulky waste, derive – for reasons of hygiene and manageability – a representative but artificial bulky waste assortment from this, and produce corresponding sample parts. Furthermore, we identify and procure a representative assortment of “real” bulky waste samples with the assistance of two industry partners from the waste-management sector.

Using these samples, we and the project partners perform measurements using various image-acquisition and image-processing techniques which are specifically adapted to the environment in which the bulky waste is sorted. The methods deployed are:

  • Conventional image-acquisition technology in the visible spectral range (Fraunhofer IOSB)
  • Near-infrared spectroscopy (Fraunhofer IOSB and Fraunhofer WKI)
  • Active heat-flow thermography (Fraunhofer WKI)
  • Terahertz imaging (Fraunhofer ITWM)

The sensor data is processed, fused and characterized by the Institute of Industrial Information Technology (IIIT). For this purpose, artificial intelligence methods are applied, in particular deep artificial neural networks (ANN). The training of the ANN is carried out using sample data at the IIIT as well as with bulky waste samples on a demonstrator at the Fraunhofer WKI.

To prove the practicality and marketability of the overall system, the Fraunhofer IOSB is conducting a field test at a sorting facility.

Social relevance

From an ecological point of view, it makes fundamental sense to recycle raw materials such as wood and non-ferrous metals instead of burning or discarding them at the end of their initial useful life. In addition, the demand for wood is rising - partially due to the fact that products of all kinds are increasingly being made from renewable raw materials. At the same time, efforts are being made to restrict the utilization of forests to some extent (forest conversion) or to prevent it completely (protected areas). As a result, the exploitation of waste wood is gaining in importance.

In Germany, more than two million tons of bulky waste are generated annually. Depending on the regional disposal concept, up to 50 percent thereof consists of wood, predominantly old furniture. Manual sorting creates, on the one hand, employment opportunities for low-skilled personnel; on the other hand, however, it can be cost-intensive. This can have a negative impact on the actual recycling ratio.  

Recyclability is furthermore dependent on sorting quality. Pre-sorting by the consumer only helps to a limited extent, as a lack of knowledge concerning the precise material composition leads to many incorrect discards.

Trained employees in waste-management companies can easily recognize wood-containing parts from pre-crushed bulky waste, but they are not infallible. Until now, commercially available sensor-based sorting processes on the basis of conventional color-camera technology have been less able to reliably detect wood-containing components from chunky waste mixtures. Prior shredding of the entire bulky waste is laborious and the sorting results remain inadequate.

The imaging methods which we utilize generate additional information which a sorting employee does not have. Our vision: By means of an “intelligent” system, bulky waste can be sorted accurately and fatigue-free - even without prior shredding.

Economic advantages

Waste-disposal companies could sort bulky waste more cost-efficiently and extract more raw materials which could then be sold.

The wood-based materials industry would become less dependent on fresh wood.

Thanks to the latest material developments, including those at the Fraunhofer WKI, waste wood can be increasingly utilized in the production of modern high-performance materials, for example in polymer composites or as an aggregate in concrete (gravel substitute). Through this broadening of the raw-material base and the increase in efficiency in recovering waste wood from bulky waste, economic advantages arise for companies which manufacture, process or utilize materials.

If bulky waste can be transformed into a valuable raw material, consumers could theoretically also profit from a direct financial benefit. Whether the disposal of bulky waste should be free of charge or even profitable for them is, however, a political question with behavioral pedagogical aspects.

Project partners

  • Fraunhofer Institute of Optronics, System Technologies and Image Exploitation IOSB (project coordination)
  • Fraunhofer Institute for Industrial Mathematics ITWM
  • Institute of Industrial Information Technology (IIIT) at the Karlsruhe Institute of Technology KIT
  • ALBA Braunschweig GmbH
  • DIE GRÜNEN ENGEL Entsorgung und Logistik GmbH
  • Dieffenbacher GmbH Maschinen- und Anlagenbau


Official project title: Altholzgewinnung aus Sperrmüll durch künstliche Intelligenz und Bildverarbeitung im VIS-, IR- und Terahertz-Bereich; Teilprojekt 2 (ASKIVIT-Thermo) mit dem Titel: Versuche zur Erkennung von Holz und Holzwerkstoffen in Sperrmüll mittels aktiver Wärmefluss-Thermographie

(Waste-wood recovery from bulky waste by means of artificial intelligence and image processing in the VIS, IR and terahertz range; Subproject 2 (ASKIVIT-Thermo) entitled: Experiments on the detection of wood and wood-based materials in bulky waste by means of active heat-flow thermography)

Funding body: German Federal Ministry of Food and Agriculture (BMEL)

Project management: Fachagentur Nachwachsende Rohstoffe e. V.

Funding reference: 2220HV048A

Duration: 1.7.2021 to 30.6.2024 

Funded by the Federal Ministry of Food and Agriculture by decision of the German Parliament