• Journal of the Korean Wood Science and Technology
• /
• v.45 no.6
• /
• pp.797-808
• /
• 2017

Automatic wood species identification systems have enabled fast and accurate identification of wood species outside of specialized laboratories with well-trained experts on wood species identification. Conventional automatic wood species identification systems consist of two major parts: a feature extractor and a classifier. Feature extractors require hand-engineering to obtain optimal features to quantify the content of an image. A Convolutional Neural Network (CNN), which is one of the Deep Learning methods, trained for wood species can extract intrinsic feature representations and classify them correctly. It usually outperforms classifiers built on top of extracted features with a hand-tuning process. We developed an automatic wood species identification system utilizing CNN models such as LeNet, MiniVGGNet, and their variants. A smartphone camera was used for obtaining macroscopic images of rough sawn surfaces from cross sections of woods. Five Korean softwood species (cedar, cypress, Korean pine, Korean red pine, and larch) were under classification by the CNN models. The highest and most stable CNN model was LeNet3 that is two additional layers added to the original LeNet architecture. The accuracy of species identification by LeNet3 architecture for the five Korean softwood species was 99.3%. The result showed the automatic wood species identification system is sufficiently fast and accurate as well as small to be deployed to a mobile device such as a smartphone.

• Journal of the Korean Wood Science and Technology
• /
• v.45 no.2
• /
• pp.202-212
• /
• 2017

This study was to establish the interrelation between chemical compositions and near infrared (NIR) spectra for the classification on distinguishability of domestic gymnosperms. Traditional wet chemistry methods and infrared spectral analyses were performed. In chemical compositions of five softwood species including larch (Larix kaempferi), red pine (Pinus densiflora), Korean pine (Pinus koraiensis), cypress (Chamaecyparis obtusa), and cedar (Cryptomeria japonica), their extractives and lignin contents provided the major information for distinction between the wood species. However, depending on the production region and purchasing time of woods, chemical compositions were different even though in same species. Especially, red pine harvested from Naju showed the highest extractive content about 16.3%, whereas that from Donghae showed about 5.0%. These results were expected due to different environmental conditions such as sunshine amount, nutrients and moisture contents, and these phenomena were also observed in other species. As a result of the principal component analysis (PCA) using NIR between five species (total 19 samples), the samples were divided into three groups in the score plot based on principal component (PC) 1 and principal component (PC) 2; group 1) red pine and Korean pine, group 2) larch, and group 3) cypress and cedar. Based on the chemical composition results, it was concluded that extractive content was highly relevant to wood classification by NIR analysis.

• Journal of the Korean Wood Science and Technology
• /
• v.19 no.4
• /
• pp.34-42
• /
• 1991

Protection of tropical forest affects on significant reduce of tropical hardwood supply, and softwood resources will be increasingly important for the timber security in Korea. U.S. softwood log was most favorite species for Korean softwood log importers in overall import conditions except price stablization and consistency of export policy. Reduced export volume from Pacific Northwest to Korean market has been immediately replenished by rediata pine from New Zealand and Chilean plantation. Siberian timber will hardly play major roles in Korean timber market unless budding structure. softwood plywood and softwood furniture uses are enhanced. Recent rapid rise of labor cost and reducing tariff rrate in Korea provided better opportunities for import lumber in building materials market. Dry dimension lumber was relatively profitable when processed from import U.S. soft-wood log while green lumber was favorable products processed from radiata pine log in Korean lumber market. This means U.S. softwood lumber would have better opportunity to market for '2${\times}$'4 studs when wood frame housing is introduced. On the other hand while radiata pine is competitive on temporary construction lumber such as supporter and concrete forming frame in Korea. Shortage of raw material for the new capacity of board plants in Korea will be it bottle neck. Major log export countries to Korea as U.S. New Zealand and Chile showed high trade intensity indices of composite hoard produces for Korean market. As Korea efforts to diversify import sources, and tariffs are reduced to 8% as scheduled by 1994. countries of scoring higher comparative advantages as Portugal. Brazil, Austria as well as New Zealand will have better opportunity to penetrate into promised Korean composites hoard market.

• Journal of the Korean Wood Science and Technology
• /
• v.30 no.4
• /
• pp.58-65
• /
• 2002

This research was carried out to develop field treatment techniques of thinned small diameter softwood logs and less utilized hardwood logs using sap displacement method. In this paper, we report the feasibility of using transpiration method and butt-end method for the treatment of three softwood species and three hardwood species with preservatives, fire-retardant chemicals, and dimensional stabilizer. Butt-end method was effective as a field treatment technique compared to transpiration method when considered the treatability, easiness of treatment, productivity of treated wood, and environmental aspects related to chemical treatment, regardless of the combination of wood species and chemicals.

• Journal of the Korean Wood Science and Technology
• /
• v.47 no.3
• /
• pp.265-276
• /
• 2019

In our previous study, the LeNet3 model successfully classified images from the transverse surfaces of five Korean softwood species (cedar, cypress, Korean pine, Korean red pine, and larch). However, a practical limitation exists in our system stemming from the nature of the training images obtained from the transverse plane of the wood species. In real-world applications, it is necessary to utilize images from the longitudinal surfaces of lumber. Thus, we improved our model by training it with images from the longitudinal and transverse surfaces of lumber. Because the longitudinal surface has complex but less distinguishable features than the transverse surface, the classification performance of the LeNet3 model decreases when we include images from the longitudinal surfaces of the five Korean softwood species. To remedy this situation, we adopt ensemble methods that can enhance the classification performance. Herein, we investigated the use of ensemble models from the LeNet and MiniVGGNet models to automatically classify the transverse and longitudinal surfaces of the five Korean softwoods. Experimentally, the best classification performance was achieved via an ensemble model comprising the LeNet2, LeNet3, and MiniVGGNet4 models trained using input images of $128{\times}128{\times}3pixels$ via the averaging method. The ensemble model showed an F1 score greater than 0.98. The classification performance for the longitudinal surfaces of Korean pine and Korean red pine was significantly improved by the ensemble model compared to individual convolutional neural network models such as LeNet3.

• Journal of the Korean Wood Science and Technology
• /
• v.21 no.3
• /
• pp.53-60
• /
• 1993

Substrates used were hardwood-Suwon poplar-(Populus alba${\times}$glandulosa L.) and softwood-pitch pine-(Pinus rigida M.). And these substrates were steam exploded then treated with sodium chlorite at 75$^{\circ}C$ with occasional stirring in order to obtain samples which had different lignin contents and crystallinity. And then this resulting samples incubated with a commercial cellulase derived from Trichoderma ressei. The contents of Klason lignin were decreased as the increasing of the ratio of sodium chlorite in the two species. The effect of hardwood was more effective than that of softwood in the same ratio of sodium chlorite. The minimum contents of Klason lignin were 0.8% and 5.1% respectively. And the crystallinities of cellulose were increased very little as increasing of the ratio of sodium clorite. The hydrolysis extent of the two species were increased as the increasing of delignification. Especially, the hydrolysis extent of hardwood was more higher than that of softwood. The maximum hydrolysis extent were 89.8% and 71.1%, respectively.

• Journal of Korean Society of Forest Science
• /
• v.95 no.2
• /
• pp.203-208
• /
• 2006

This study compared the roundwood demand prediction accuracy of econometric and time-series models using Korean data. The roundwood was divided into softwood and hardwood by species. The econometric model of roundwood demand was specified with four explanatory variables; own price, substitute price, gross domestic product, dummy. The time-series model was specified with lagged endogenous variable. The dummy variable reflected the abrupt decrease in roundwood demand in the late 1990's in the case of softwood roundwood, and the boom of plywood export in the late 1970's in the case of hardwood roundwood. On the other hand, the prediction accuracy was estimated on the basis of Residual Mean Square Errors(RMSE). The results showed that the softwood roundwood demand prediction can be performed more accurately by econometric model than by time-series model. However, the hardwood roundwood demand prediction accuracy was similar in the case of using econometric and time-series model.

• Journal of the Korean Wood Science and Technology
• /
• v.31 no.6
• /
• pp.55-59
• /
• 2003

The fixation and leaching characteristics of chromated copper arsenate (CCA Type C) and chromium- copper-fluoride-zinc (CCFZ) in domestic softwood (Japanese red pine, Korean pine, and Japanese larch) sapwood were investigated using the expressate method to follow chromium fixation and the American Wood-Preservers' Association (AWPA) leaching procedure to determine leaching properties after fixation. The rates of fixation were affected by preservative types; CCA was fixed much faster than CCFZ for all species evaluated. There were definite differences in the fixation rates of different species, with Korean pine requiring shorter to fix than the other species evaluated. Chromium fixation was greatly enhanced by elevated temperatures, and fixation time can be estimated according to fixation temperatures applied. The percentage of arsenic and zinc leached from domestic softwoods was relatively high compared to chromium and copper, indicating that there is still a relatively high unfixed arsenic and zinc components after complete chromium fixation in CCA-and CCFZ-treated samples, respectively.

• Journal of the Korean Wood Science and Technology
• /
• v.41 no.4
• /
• pp.337-345
• /
• 2013

Log grading rules are essential tools to ensure the quality of logs in distribution structure. The rules should reflect the long experience and accepted usage practice in the market. A gap between the rules and market should be improved based on analysis of log qualities that produced and market demand. In this study more than ten millions logs which were produced by 5 Regional Forest Services in 2010~2011 period, were analyzed in their qualities including diameters and lengths by species. A proposal was driven to improve the current log grading rules in terms of log diameter classes and length. The followings are the summary of this study. Most of domestic softwood logs are belong to small diameter class of 100~160 mm, which imply the diameter classes of current log grading rules are immoderate. Distributions of log diameter shows distinctive patterns by species, which indicate a necessity of differentiated diameter classes by species in an improved rules. Lengths of logs in productions do not corresponding to the demands and preferences in sawmills. Therefore it is highly recommended to include log length term in an improved log grading system. Based on these findings, 6 log grading systems for 3 species groups of softwood are newly proposed to improve current log grading rules. Limits of log diameter and log length are also proposed for each log grading system.

• Journal of the Korean Wood Science and Technology
• /
• v.18 no.2
• /
• pp.49-66
• /
• 1990

In order to identify an unknown wood sample native to Korea. the softwood databases(KSWCHUN; Korean SoftWood CHUN) and the hardwood databases(KHWCHUN; Korean HardWood CHUN) had been built. and the new computer searching programs(IDINEX; IDentification INformation EXpress) has been written in Turbo Pascal(V.5.0) and in Macro Assembly(V.5.0). The characters of the data were based on the 74 features of softwood and on the 148 features of hardwood which are a part of new "IAWA list of microscopic features for hardwood identification" published in 1989. For the purpose of this investigation the wood anatomical nature of 25 species of softwood(13 genera of 5 families) and of 112 species of hardwood(57 genera of 31 families) were observed under a scanning electron microscope and light microscope. and a lot of literature used. The IDINEX programs are based on edge-punched card keys. with several improvements. The maximum number of features in the IDINEX is 229. but that is fixed for a given database. Large numbers of taxa are handled efficiently and new taxa easily added. A search may be based on sequence numbers of features. Comparisons are made sequentially by feature and taxon using the entire suite of features specified to produce the list of possible matching taxa. The results are followings. (1) The databases of Korean wood and the searching programs(IDINEX) had been built. (2) The databases of Korean wood could be an information to search an unknown wood. (3) The databases would be valuable. for the new features, which were not mentioned in Korean wood up to the present. were observed in details. (4) The ultrastructures of the cell walls(warty layer) and crystals observed under a scanning electron microscope will be helpful to search an unknown wood in particular. (5) The searching process is more quick and accurate than the others. 6) We can obtain the information on the differences of a species from the other and search an unknown wood using probability. in IDINEX, (7) The IDINEX will be utilized to identify and classify an animal life, vegetable world, mineral kingdom, and so on.