File name: ReadMe-ForestLogAssortmentsPhotos-SGAuction-2023-ver1 This (ReadMe) file was created on 2025-10-03 by Vasja Kavčič, Matevž Triplat ------------------- GENERAL INFORMATION ------------------- Name of the dataset: Forest log assortments photo dataset from the 2023 Slovenj Gradec timber auction Authors Name and surname: Matevž Triplat Vasja Kavčič Martin Jež ORCID: 0000-0003-0130-0497 0009-0004-0878-1312 Institution/Affiliation: Slovenian Forestry Institute ROR organization identifier: https://ror.org/0232eqz57 Email: matevz.triplat@gozdis.si vasja.kavcic@gozdis.si martin.jez@bf.uni-lj.si Date of data collection: from 2023-03-09 to 2023-03-16 Geographical location of data collection: Slovenj Gradec, Slovenia (46.49834737983923, 15.070303498836346) Information on the funders/programmes/projects that made the data collection possible: Data collection for this study was co-funded by the Ministry of Agriculture, Forestry and Food of the Republic of Slovenia and the Slovenian Research and Innovation Agency, within the projects “Efficient management of private forests to support wood mobilization” grant number CRP V4-2013” and "New operational e-services in the field of forestry: eForestry" within the framework of the Recovery and Resilience Plan. The study was also supported by the research core group “Forest biology, ecology and technology” (P4-0107), financed by the Slovenian Research and Innovation Agency, and by the Development Fund of the Slovenian Forestry Institute. License: CC BY 4.0 https://creativecommons.org/licenses/by/4.0/legalcode.en ----------------------------- SHARING/ACCESSING INFORMATION ----------------------------- Data licences/restrictions: CC BY 4.0 Links to publications that cite or use the data: Written article supported by the data is not yet published as of 10/3/2025 Links to other publicly available data sites: / Links to ancillary databases: / Data was not obtained from another source. ---------------------- VIEWING DATA AND FILES ---------------------- File list: In zip file there a directories which contain photographies of forest log assortments, grouped by tree species (ForestLogAssortmentsPhotos-sortedBySpecies-SG-2023) and market value (ForestLogAssortmentsPhotos-sortedByMarketValue-SG-2023). ForestLogAssortmentsPhotos-sortedBySpecies-SG-2023 content: Common ash Common walnut European larch Norway spruce Pear Pedunculate Oak Sessile oak Sycamore Wych elm File ratio (number of images per directory): Common ash: 382 Common walnut: 137 European larch: 162 Norway spruce: 838 Pear: 124 Pedunculate Oak: 276 Sessile oak: 2634 Sycamore: 897 Wych elm: 105 ForestLogAssortmentsPhotos-sortedByMarketValue-SG-2023 content is first grouped by tree species: European Larch, Norway spruce, Sessile oak and Sycamore. Each of the directory contains photos sorted by the SG auction market value in quartiles: q1 - first quartile (assortments with lowest price) q2-q3 - second and third quartile (assortments with avarage price) q4 - fourth quartile (assortments with highest price) Additional related data collected that were not included in this dataset: Full quality photographies currently aren't available online. If needed, please contact the authors. Auction organizer's database isn't public due to confidentiality of information. This is currently the only version of the dataset. -------------------------- METHODOLOGICAL INFORMATION -------------------------- Description of the methods used to collect/obtain the data: Images were captured with different cameras (Nikon D7000, iPhone 11 Pro, and Canon PowerShot G3 X), producing resolutions of 4928 x 3264, 3024 x 4032, and 3648 x 5472 pixels, respectively. The images were taken while assortments were on display on auction in Slovenj Gradec. Image capture was completed over two days in March 2023, resulting in a dataset of 5,549 assortment images. Data processing methods: Each assortment at the auction was marked with an identification plate, visible in the photographs. Images were processed in Adobe Lightroom, where the file names were renamed to correspond to the identification codes on the plates. Images were also cropped horizontally to cover only the individual assortment, minimizing background. The auction organisers provided a database containing assortment characteristics, including diameter, length, volume, tree species, number of offers, and winning offer price. File sorting was performed in Python (version 3.11.1). The first four digits of each file name were parsed and converted to integer numbers. Using the pandas module (version 1.5.3), information from the auction database was merged with the assortment images by ID, adding tree species information. The database was then filtered to exclude tree species represented by fewer than 100 images. Using the Python os and shutil modules, files were copied in folders corresponding to assortment tree species. Photos were processed with Pillow (version 9.5.0): resolution of the photos was reduced to fixed height of 299px, resampled using bilinear resampling interpolation method and compressed using JPEG compression. For the market value sorted dataset, descriptive statistic by the tree specie was computed for the highest offered auction prices - images were sorted accordingly to corespond with first, second-third and fourth quartile. Photos of this dataset are processed by reducing resolution of original photos to fixed height of 600px, images were resized using Laczos resampling and compressed with Pillow (version 9.5.0) using JPEG compression. Software information: Adobe Lightroom classic Python (version 3.11.1) Pandas (version 1.5.3) Pillow (version 9.5.0)