Overview
This dataset is part of the project Neuralwood: Neural Networks for the Prediction and Utilization of Natural Material Variations in Timber Construction by the Department for Computing in Architecture, Institute for Computational Design and Construction (ICD/CA) and Materials Testing Institute (MPA), University of Stuttgart.
The dataset contains raw data from different scanning, measurement and testing data related to the material properties and wood anatomy of European Spruce (Picea Abies) boards for producing self-shaping double layered plates (bilayers).
Note on Data Intelectual Property: To protect industrial intellectual property while maintaining Open Science standards, the dataset is divided into two distinct packages based on intellectual property rights and access requirements. The data internally produced by
ICD/CA and
ICD/CA is fully publicly available, while the data produced using
Microtec Srl. support remains
protected IP by Microtec Srl.. This dataset only serves as data index to the actual scan data
Data Sources
Fiber Orientation Matrix
Scans of fiber orientation from individual spruce boards were acquired using a GoldenEye scanner from Microtec Srl., Italy.
The scans were conducted in two different sites: (1) the Microtec Lab in Brixen, Italy and (2) HBS Berga Sawmill Facilty in Berga, Germany. This data will only contain shareable metadata.
End Grain Image
End grain images taken by ICD/CA from some spruce board samples were documented using mobile-phone cameras and processed for image cleaning using custom Python script. This data will be used to train an artificial neural network based growth ring extraction model.
Hygro Measurement Data
Hygromeasurement data from small scale spruce samples (taken from the project's material stock) was acquired by MPA.
Destructive Test Data
Experimental destructive test data from a small amount of spruce samples (taken from the project's material stock) was acquired by ICD/CA using material testing lab at the Faculty of Architecture, University of Stuttgart, operated by ITKE.
Rights and Licensing
Folder / Component: fiber_orientation_matrix/
- Access Level: Public
- License Type: Data Index to the actual fiber orientation matrix data (Protected IP by Microtec Srl.)
Folder / Component: end_grain_images/, hygro_measurement/, destructive_test/
- Access Level: Public
- License Type: Creative Commons Attribution 4.0 International (CC BY 4.0)
Data Structure
destructive_test/: Contains experimental results for 4-point bending tests performed on spruce board samples (with standardized size).
250722_4-Pt-Bending_Destructive.xls250819_4-Pt-Bending_Destructive.xls
end_grain_images/: A large dataset of 501 image files capturing the end-grain surfaces of the wood samples.fiber_orientation_matrix/: This section contains metadata (data indexing) for the actual fiber orientation matrix from different scanning sites as JSON files.
berga/: A collection of JSON files (SP_B_XXX.json) representing individual board fiber data for the HBS Berga site.microtec/: A collection of JSON files (SP_M_XXX.json) representing individual board fiber data for the Microtec Lab site
hygro_measurement/: Stores moisture induced shrinkage-related experimental data using small scale samples.
spruce_sample_hygro_measurement.xlsx
Note: The JSON naming convention SP_B_XXX.json or SP_M_XXX.jsoncorresponds to the specific board ID numbering used throughout the project.
Requesting Access to Actual Fiber Orientation Matrix Data
Researchers requiring access to the actual fiber orientation matrix data (protected data) for purposes of academic verification, methodological comparison, or replication may submit an access request to the project team (please use the Contact Owner button above).
Access to the protected data can only be granted following a coordinated communication and approval process involving:
- The requesting researcher or institution
- The project research team
- Microtec Srl.
Approval is subject to agreement by all parties and requires the signing of a Data Use Agreement (DUA). Although the scanning services were commissioned by the project, the intellectual property embedded in the raw dataset, including proprietary scanner parameters, industrial workflows, and trade secrets, remains the property of Microtec Srl. and must not be disclosed, redistributed, published, or reverse-engineered.
Requests that cannot be approved by Microtec Srl. will not be granted access, regardless of academic intent. Actual data distribution might be conducted by Microtec Srl. directly.
