Figure 8 - uploaded by Tomáš Gergeľ
Content may be subject to copyright.
Source publication
This article deals with the testing of a methodology for creating log cutting patterns. Under this methodology, programs were developed to optimize the log yield. Testing was conducted by comparing the values of the proportions of the individual products resulting from an implementation of the proposed cutting pattern of a specific log with the cal...
Context in source publication
Context 1
... value is chosen on the basis of the qualitative sorting of the timber; • Standard timber length: with the entered input parameters, the program calculates the exact lengths of the individual timber pieces. It allows one to select the standard timber lengths (rounded to an integer in tenths of a meter); • Nominal dimensions of timber (Figure 8): the cutting pattern is compiled using two tables. One table is intended for vertical timber and the other for horizontal timber. ...
Similar publications
Citations
... On the one hand it is prone to leakage and mis-detection due to visual fatigue, and on the other hand the manually calibrated defect rejection scheme often does not maximise the use of wood (Lai et al. 2021). At the same time, wood processing is more conservative compared to other industries, so it is necessary to develop technology to work with intelligent algorithms for wood defect identification and rejection (Gergeľ et al. 2020). This has also become a new hot topic in the wood processing industry (de Geus et al. 2021;Zhang et al. 2018). ...
Wood utilisation is an important factor affecting production costs, but the combined utilisation rate of wood is generally only 50 to 70%. During the production process, the rejection scheme of wood defects is one of the most important factors affecting the wood yield. This paper provides an overview of the main wood defects affecting wood quality, introduces techniques for detecting and identifying wood defects using different technologies, highlights the more widely used image recognition-based wood surface defect identification methods, and presents three advanced wood defect detection and identification equipment. In view of the relatively fixed wood defect recognition requirements in wood processing production, it is proposed that wood defect recognition technology should be further developed toward deep learning to improve the accuracy and efficiency of wood defect recognition.
... The recent optimization has been connected with trends leading towards automation of assessing the qualitative features, which can be seen primarily in wood processing plants [14][15][16]. Optimization approaches to assessing the qualitative features represented by expensive technologies have been unavailable in forestry field operations so far. Trunk quality assessment is, therefore, usually carried out visually and manually by a respective employee. ...
Proper quality assessment of timber requires a certain level of knowledge and overview of technical conditions and correct identification and assessment of the qualitative features of trunks. The ratio of the highest quality classes is decreasing. Therefore, increasing the potential financial resources allocated to forest management could lead to the improvement and increase of this ratio. The objective of the study was to identify the frequency and occurrence of limiting features in the group of non-coniferous beech and oak trunks. A further objective was to classify major factors causing and increasing the frequency of occurrence of such limiting marks. Altogether, 969 beech and oak trunks were assessed in the University Forest Enterprise of the Technical University in Zvolen. The dependences of the size and occurrence of individual qualitative features on the selected factors were statistically assessed using the Pearson correlation coefficient and Cramer's V; significance was established using χ 2 test and a significance level α = 0.05. The most frequently occurring features were sweep, knots, and heart shakes. The results of the comparative and statistical analysis indicate that the management of forest stands and interventions carried out in the forest stands affect the occurrence of the negative features being analyzed the most. However, the conditions of the given site (soil, subsoil, and slope) also play a certain role and can affect the technological aspect of the harvest. The obtained results are valid for the conditions of the University Forest Enterprise of the Technical University in Zvolen; however, they can also be applied in a wider range of similar conditions of Central European forest stands.
Accurate qualitative evaluation of grown and harvested wood is a key issue from the point of view of its subsequent economic evaluation. With the current trend of global climate change and large volumes of wood damaged by harmful agents, automated methods of wood quality assessment are becoming more and more important. The work aimed to verify the applicability and significance of the results of using the acoustic tomograph for the qualitative assessment of selected tree species logs. Ten samples of log sections of non-coniferous and coniferous trees were evaluated, on which an image analysis of qualitative features was performed on a cross-section from their digital photograph and the image output of an acoustic tomograph software. The results were compared with each other and the accuracy of qualitative feature identification by acoustic tomograph was evaluated. At the same time, the results of the image analysis of the qualitative feature were compared with its assessment through STN EN 1309-3. It was shown that, when evaluated according to the Standard, qualitative features were overestimated by an average of 29.19% compared to the acoustic tomograph and by 28.22% compared to the digital photograph. The use of the acoustic tomograph confirmed a good level of accuracy in the identification of qualitative features even on logs of harvested wood, although it is primarily intended for the qualitative evaluation of standing trees.
Valuation of timber growing in commercial forests is a key issue for forest owners, forest enterprises, but also a starting point for long-term planning in the field of forest management. The subjective approach of the evaluator can, to a large extent, lead to inconsistencies in the area of wood qualitative evaluation. This paper aimed to perform an objective qualitative analysis on a selected set of 179 logs of hardwood raw-wood assortments in a selected Forest enterprise in Slovakia. Qualitative analysis, which was performed by the Technical Conditions used by Lesy SR, š.p. confirms the identical classification of raw-wood assortments, in comparison with the classification performed by the management of the forestry enterprise, in 65 logs (36.3% of logs). In 114 (63.7%) logs, the log assortment was classified in another quality class. Most of the logs, which were reclassified to lower quality classes, showed limiting qualitative features of multiple sweep (83 logs), resp. significant simple sweep over 8 cm/m (5 logs), soft rot over 20% of the end diameter area (10 logs), and decaved knots (19 logs). Our work confirmed that the qualitative evaluation of raw-wood assortments in forestry enterprises in Slovakia is not optimal and correct. The work should thus contribute to improving the setting of optimization processes in the timber production phase in forestry enterprises.
The timely detection of defects in wood helps optimize operation of sawmills and find effective log processing solutions. This paper aims to develop effective method for automated wood defect detection and recognition in CT images using a multiple-layer convolutional neural network and a reinforcement learning strategy. The data augmentation technique was proposed to increase the volume of training, validation, and test sets. The network can achieve sufficient accuracy up to 98.7% at a total set of 500 images. The study shows a direct non-linear relationship between the dimension of the training set and the recognition rate. The results of calculating performance metrics for the developed method indicate the high accuracy of the ANN prediction model. The study results will be useful in designing software applications for industrial and laboratory CT scanners that are used in lumber production and R&D centers. The proposal can be improved to perform wood defect detection and recognition in color and 3D CT images of logs.
