Address: 17 Naberezhnaya Severnoy Dviny, Arkhangelsk 163002 Russian Federation. Northern (Arctic) Federal University named after M.V.Lomonosov. Office 1425

Phone / Fax: (818-2) 21-61-18



Algorithmization of Wood Flaw Detection

Версия для печати
Creative Commons License
These works are licensed under a Creative Commons Attribution 4.0 International License.

A.E. Alekseev, I.A. Toloknov

Complete text of the article:

Download article (pdf, 0.9MB )





The use of the optical form and image scanning of the uncut timber surface allows to  define the most profitable cutting variant. The form scanning is using to determine the maximum dimensions of sawn timber and future wane, and the reflectivity scanning of the lumber surface is using to determine the presence and position of the main defects. To implement the wood flaw detection based on the analysis of form and image of the board surface, it is necessary to develop relevant algorithms and techniques. Firstly we need to determine the principle of action and the functional diagram of the scanning device. After that we need to create the control algorithms of the scanner.  Defining the primary type of obtained during the scanning process data, we can begin developing analysis algorithms. Previously, the scanned data must be led to the suitable form for the defective analysis - the geometric model and whole image of the timber surface. The image should be matched with the shape so that for the every  point of the image its spatial position on the surface have been determined. The experiment resulted in the detection algorithms of the uncut timber and the scanning device with the software. The paper describes the concept and algorithms of the photometric shape and surface reflectance of uncut timber scanner. An algorithm for constructing a three-dimensional shape model of boards, based on data received from the device, as well as algorithms face and some mechanical damage allocation are  developed. The formation method of lumber surface linearized image, and algorithms for allocating major defects are also described. As a result of the described algorithms we can identify possible options for timber processing and predict the future quality of sawn timber for each option before processing. Obtained as a result scan data can also be further used for the development and testing of various algorithms to analyze the quality of sawn timber, as well as the collection of statistical data.


A.E. Alekseev, Doctor of Engineering, Professor, I.A. Toloknov, Postgraduate Student


Northern (Arctic) Federal University named after M.V. Lomonosov ::: Naberezhnaya Severnoy Dviny, 17, Arkhangelsk, 163002, Russia; е-mail:


wood flaw detection, algorithmization woodworking, technical vision, three-dimensional scanning, optical scanning.


1. Kopeikin A.M., Zadrauskaite N.O., Turushev V.G., Gelfand E.D. K voprosu avtomatizirovanija opredelenija defektnyh uchastkov na pilomateriale [The Question of the Automate Determining of the Defective Sites at the Lumber]. Izv. vissh. ucheb. zavedeniy. Lesnoy zhurnal 2012, no. 3, pp. 72-75. 2. Akushenkov Y.G. Tehnicheskoe zrenie robotov [Technical Robot Vision]. Moscow, 1990, pp. 162-164. 3. Szeliski R., Computer Vision: Algorithms and Applications. Springer, 2010, pp. 578-616.

Algorithmization of Wood Flaw Detection


Make a Submission


Lesnoy Zhurnal (Russian Forestry Journal) was awarded the "Seal of Recognition for Active Data Provider of the Year 2024"