Possibilities of image analysis for quality wood sorting

• Main Authors: Gejdoš Miloš, Gergeľ Tomáš, Bucha Tomáš, Vyhnáliková Zuzana
• Format: Article
• Language: ces
• Published: Sciendo 2019-09-01
• Series: Lesnícky Časopis
• Subjects: log quality; wood sorting; image analysis; wood proporties
• Online Access: https://doi.org/10.2478/forj-2019-0015

Wood assessment optimization should be the top priority of the forestry subjects that are fundamentally dependent on the income from its sale. The aim of this paper is to analyse the beech, oak and ash tree logs that were categorized into quality classes according to the size of one of the qualitative characters related to the surface area (false heartwood, rot). The classical methodology used in forestry was compared with the application of ImageJ software. In total, thirty logs were analysed. The characters of false heartwood and rot were chosen and evaluated according to their size on the log end. There were no other characters that obstructed the categorization into quality classes. The ImageJ software application led to improved assessment (transfer to a higher quality class) in 56% of the logs. The volume of the evaluated assortments was 18.43 m3. The total difference in the value of the assortments with the ImageJ software application reached + €70.44 (+ 4.7%). The analysis therefore confirmed that in case of a considerable irregularity in a qualitative character (when the surface area of the character significantly differs from the circumscribed circular surface), the standard STN EN 1309–3 methodology systematically overvalues the surface area of this character. That affects the assessment potential of the specific log.