• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2003년도 Proceedings of ACRS 2003 ISRS
• /
• pp.760-762
• /
• 2003

This study confirmed the usefulness of short-wavelength infrared (SWIR) in the discrimination and classification of evergreen forest types. A forested area near Hisayama and Sasaguri in Fukuoka Prefecture, Japan, served as the study area. Warm-temperate forest vegetation dominates the study site vegetation. Coniferous plantation forest, natural broad-leaved forest, and bamboo forest were analyzed using LANDSAT5/TM and SPOT4/HRVIR remote sensing data. Samples were extracted for the three forest types, and reflectance factors were compared for each band. Kappa coefficients of various band combinations were also compared by classification accuracy. For the LANDSAT5/TM data observed in April, October, and November, Bands 5 and 7 showed significant differences between bamboo, broad-leaved, and coniferous forests. The same significant difference was not recognized in the visible or near-infrared regions. Classification accuracy, determined by supervised classification, indicated distinct improvements in band combinations with SWIR, as compared to those without SWIR. Similar results were found for both LANDSAT5/TM and SPOT4/HRVIR data. This study identified obvious advantages in using SWIR data in forest-type discrimination and classification.

• 한국환경복원기술학회지
• /
• 제18권2호
• /
• pp.89-104
• /
• 2015

In order to effectively manage forested areas in South Korea on a national scale, using remotely sensed data is considered most suitable. In this study, utilizing Land coverage maps and Forest type maps of national geographic information instead of collecting field data was tested for conducting supervised classification on SPOT-5 and KOMPSAT-2 imagery focusing on forested areas. Supervised classification were conducted in two ways: analysing a whole area around the study site and/or only forested areas around the study site, using Support Vector Machine. The overall accuracy for the classification on the whole area ranged from 54.9% to 68.9% with kappa coefficients of over 0.4, which meant the supervised classification was in general considered moderate because of sub-classifying forested areas into three categories (i.e. hardwood, conifer, mixed forests). Compared to this, the overall accuracy for forested areas were better for sub-classification of forested areas probably due to less distraction in the classification. To further improve the overall accuracy, it is needed to gain individual imagery rather than mosaic imagery to use more spetral bands and select more suitable conditions such as seasonal timing. It is also necessary to obtain precise and accurate training data for sub-classifying forested areas. This new approach can be considered as a basis of developing an excellent analysis manner for understanding and managing forest landscape.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2002년도 Proceedings of International Symposium on Remote Sensing
• /
• pp.708-708
• /
• 2002

Recent developments in sensor technologies have provided remotely sensed data with very high spatial resolution. In order to fully utilize the potential of high resolution images, new image classification strategies are necessary. Unfortunately, the high resolution images increase the spectral within-field variability, and the classification accuracy of traditional methods based on pixel-based classification algorithms such as Maximum-Likelihood method may be decreased (Schiewe 2001). Recent development in Object Oriented Classification based on image segmentation algorithms can be used for the classification of forest patches on rugged terrain of Korea. The objectives of this paper are as follows. First, to compare the pros and cons of image classification methods based on pixel-based and object oriented classification algorithm for the forest patch classification. Landsat ETM＋ data and IKONOS data will be used for the classification. Second, to investigate ways to increase classification accuracy of forest patches. Supplemental data such as DTM and Forest Type Map of 1:25,000 scale are used for topographic correction and image segmentation. Third, to propose the best classification strategy for forest patch classification in terms of accuracy and data requirement. The research site for this paper is Namhansansung Provincial Park located at the eastern suburb of Seoul Metropolitan City for its diverse forest patch types and data availability. Both Landsat ETM＋ and IKONOS data are used for the classification. Preliminary results can be summarized as follows. First, topographic correction of reflectance is essential for the classification of forest patches on rugged terrain. Second, object oriented classification of IKONOS data enables higher classification accuracy compared to Landsat ETM＋ and pixel-based classification. Third, multi-stage segmentation is very useful to investigate landscape ecological aspect of forest communities of Korea.

• 한국화재소방학회논문지
• /
• 제19권3호
• /
• pp.64-69
• /
• 2005

산불은 경제적 손실뿐만 아니라 인명을 위협할 수 있는 국가적 재해다. 이러한 산불을 미연에 방지하고 피해를 저감하기 위해서는 산불발생위험지역을 사전에 판단하여 효율적으로 관리하는 것이 필요하다. 본 연구에서는 입지환경에서 중요한 부분을 차지하는 산림토양특성 중 토양형, 지형, 토성, 경사, 배수 등과 산불발생지점을 가지고 각 지점별 산불발생위험을 예측할 수 있는 산불발생확률 모형을 개발하였다. 개발 시 조건부확률과 GIS를 이용하였다 개발된 산불발생확률 모형의 적합성 검정을 위하여 추정모형의 예측력 비율을 검토할 수 있는 예측비곡선에 적용한 결과 실효성이 있는 것으로 나타났다. 이러한 결과를 적용하여 산불관리자가 손쉽게 산불발생위험지역을 파악할 수 있도록 위험지역을 구분하였다.

• Geomechanics and Engineering
• /
• 제37권6호
• /
• pp.539-554
• /
• 2024

This study explores development of prediction model for seismic site classification through the integration of machine learning techniques with horizontal-to-vertical spectral ratio (HVSR) methodologies. To improve model accuracy, the research employs outlier detection methods and, synthetic minority over-sampling technique (SMOTE) for data balance, and evaluates using seven machine learning models using seismic data from KiK-net. Notably, light gradient boosting method (LGBM), gradient boosting, and decision tree models exhibit improved performance when coupled with SMOTE, while Multiple linear regression (MLR) and Support vector machine (SVM) models show reduced efficacy. Outlier detection techniques significantly enhance accuracy, particularly for LGBM, gradient boosting, and voting boosting. The ensemble of LGBM with the isolation forest and SMOTE achieves the highest accuracy of 0.91, with LGBM and local outlier factor yielding the highest F1-score of 0.79. Consistently outperforming other models, LGBM proves most efficient for seismic site classification when supported by appropriate preprocessing procedures. These findings show the significance of outlier detection and data balancing for precise seismic soil classification prediction, offering insights and highlighting the potential of machine learning in optimizing site classification accuracy.

• 한국산림과학회지
• /
• 제90권4호
• /
• pp.535-542
• /
• 2001

이 연구의 목적은 토사붕괴산지재해에 대한 산지의 기능적합도 평가를 위한 공간분석모델링기법을 개발하기 위해 수행되었다. 기능적합도는 산지의 토사붕괴 가능성에 따라 상, 중, 하의 3단계로 구분되었다. 토사붕괴의 가능성온 경사, 모암, 토심, 경사형태, 임상 및 임목의 직경급과 같은 7개의 입지인자들에 대한 측정치를 이용하여 추정되었고, 이 과정에 토사붕괴 발생 요인으로서 각 인자들의 상대적 가중치는 AHP기법에 의해 결정되었다. 공간분석모델링은 7개 입지인자들에 대한 $25m{\times}25m$ grid 분석 혹은 TIN 분석을 통해 기초 layer 작성에서 시작된다. 이를 토대로 재분류 및 점수화 과정을 거쳐 토사붕괴 가능성 추정에 필요한 인자들의 속성 값을 지니는 새로운 layer를 형성한다. 이러한 속성 값에 가중치를 적용하고 지도대수분석을 통해 $25m{\times}25m$ cell 단위의 기능평가도를 작성하고, 마지막으로 cell-grouping을 통해 보다 실무적인 기능도를 작성하게 된다. 이 논문은 이러한 일련의 공간분석모델링 과정을 방법론적 관점에서 제시한다.

• Journal of Forest and Environmental Science
• /
• 제9권1호
• /
• pp.57-66
• /
• 1993

본 연구는 강원도 춘천군에 위치한 임업진흥촉진지역 21,417ha를 대상으로 10개의 지형특성치를 이용, 집락분석과 주성분분석에 의하여 임지를 구분함과 동시에 판별분석에 의하여 임지구분에 영향을 미치는 최소한의 변수를 검토하여, 실질적으로 기계화 집재작업을 위한 작업기종의 선정에 정확한 지형정보를 제공하는데 목적이 있다. 그 결과는 다음과 같다. 1) 고성단지는 총면적 2,252ha로 중에서 57%가 완지형으로 분류되어 트랙터형 집재가 적합하고 43%가 급지형으로 중거리 가선집재형이 적합함을 보여 주었다. 2) 가정단지(2,306ha)와 광판단지(2,627ha)는 각각 65%와 67%가 급지형으로 분류되어 대개 중거리 가선형의 집재가 적합하고, 나머지 35%와 33%가 트랙터형 집재가 적합한 완지형으로 나타났다. 3) 지암단지(4,519ha)는 대부분의 지역이 급지형으로 분류되어 중거리 가선형이 적합하였다. 4) 군자단지(3,400ha), 수동단지(3,894ha), 신포단지(2,430ha)는 총면적중에서 각각 85%, 75%, 75%가 급지형으로 분류되어 중거리 가선집재형이 적합함을 보여 주었다.

• Journal of Plant Biology
• /
• 제9권1_2호
• /
• pp.7-16
• /
• 1966

In order to determine the factors related to site quality, 13 areas of Larch growing in the Kwangung and its vicinity forest as sample plots, were examined. Sample plots included various site classes as well as age classes. Three were divided into two groups (major and minor trees). Average height of dominant trees was determined through messurement of 5 to 6 dominant tree in each sample plots. Average height of dominant 30 year-old trees was the basis for site index. A Standard Yield Table for the larch produced in Kwangnung forest was made by various data, which included age class 5, ranging from 10 to 45 years. The relationship of the height of the trees, the site conditions, and ground vegetation are investigated in this paper. The site indexes of 40 forest class age in 28-B and 28-G forest classes of the larch associations for ground vegetation had comparatively rarge differences due to the sampled areas. The relation of the direction of forest communities to the height and the diameter of the tree shwoed that its communiteis of northest and northwest parts appeared higher valueof the height and the diameter. The diameter and the height of trees were closely realted to each other. The samller the occupied area per tree and the smaller the average distance among trees, the more density was increased. The larger the density was the lower height of the trees. In the ground vegetation of the larch communities, there seems to be a definite correlation between the height of trees and the occupied area per tree or the average distance among the trees. The height of trees and site index of two larch communities were as follow: 28-B forest class site index 20.8, height 24.0m, 28-G forest class site index 18.4, height 20.9m. The ground layer was analyzed by the method of Quadrat(20/20sq. cm) with an interval of 1M. It set up 40 Quadrats of the larch communiteis. The community structure of the ground vegetation of two larch was analyzed, and important value was calculated and then evaluated. The ground vegetation under the larch had developed Burmannii Beauv stratal society below the 28-B and 28-G the forest class. Accordingly, the first important value of Burmannii Beauv was found in two ground vegetation below the larch. Therefore, this species could be quantitatively considered as the forest indicator species. Common species of each community appeared 18 species out of 34 species in the ground vegetation under two larch communities. The ground vegetation of the 28-B forest class showed more than that of the 28-G forest class. the similarity of the ground vegetation was measrued by the Frequency Index Community Coefficient. The differences between the associations were lcearly manifested by the ground vegetation tested by Gleason's Frequency Index of Community Coefficient for the analysis of each stratal society of all associations. According to F.I.C.C. the ground vegetation under two larch(28-B and 28-G) forest classes showed higher value. An investigation into the relationship of physical and chemical properties of soil and site was considered the next step to be taken in the study of the larch site classification.

• 환경영향평가
• /
• 제22권1호
• /
• pp.65-78
• /
• 2013

Since pursuing the pleasant life for people, there is an increase of desire to appreciate outstanding scenery with the difference in certain level for perception and understanding of human on landscaping, However, the quality of landscaping has become artificial with the pleasance to be declining due to the urbanization. This study was applied at the site around High One Resort area in Gohan-eup, Jeongseon-gun Gangwon-do for analyzing the areas sensitive to the landscaping change as well as degree of requirement for landscape management for forest landscape management with the focus on presenting the zoning method and the management class classification method. Even if the forest is the same, the function of it is different depending on land use or what resource is placed that the forestry function is found out to present the management plan for each forestry function in the subject site and the result of the management grade classification is analyzed in overlapping to the forestry function level. As a result, from the landscaping management requirement and visual absorption analysis, the result formulated for upper, middle and lower zones to classify the final forestry landscape management degree into 1-4 grades and the management plan is presented on the respective 1-4 grade area for each forestry function. By applying the technique to set the management grade, it was possible to formulate the result to provide the means for integrated management in consideration of the forestry function and management of forestry landscape and resources.

• Journal of Forest and Environmental Science
• /
• 제36권3호
• /
• pp.199-206
• /
• 2020

Satchari National Park is one of the most biodiverse forest in Bangladesh and home of many endangered flora and fauna. 206 tons of CO₂ per hectare is sequestrated in this national park every year which helps to mitigate climate issues. As people living near the area are dependent on this forest, degradation has become a regular phenomenon destroying the forest biodiversity by altering its forest cover. So, it is important to map land cover quickly and accurately for the sustainable management of Satchari National Park. The main objective of this study was to obtain information on land cover change using remote sensing data. Combination of unsupervised NDVI classification and supervised classification using maximum likelihood is followed in this study to find out land cover map. The analysis showed that the land cover is gradually converting from one land use type to another. Dense forest becoming degraded forest or bare land. Although it was slowed down by the establishment of 'National Park' on the study site, forecasting shows that it is not enough to mitigate forest degradation. Legal steps and proper management strategies should be taken to mitigate causes of degradation such as illegal felling.