Earticle

High-resolution land cover maps are essential in fields such as forest resource management, urban green space planning, and environmental protection. In recent years, Unmanned Aerial Vehicles (UAVs) have increasingly become influential in land cover mapping due to their flexibility, low cost, and fast data acquisition capability. However, accurately classifying high-resolution image data collected by UAVs remains a challenge due to the complexity of the data and the substantial computational resources required for processing. To address this problem, this study combines UAV remote sensing data with Object-Based Image Analysis (OBIA) to optimize feature selection to improve the accuracy of land cover classification and provide more reliable data support. In this study, combinations of four feature types were evaluated using a Decision Tree (DT) algorithm in eight scenarios. The results showed that a comparison with spectral features alone and the combination of other feature types can significantly improve the classification accuracy. Height features contribute the most to enhancing the classification results, followed by spectral and geometric features, while the contribution of texture features is relatively limited. In addition, the optimal feature combination selected by the Recursive Feature Elimination (RFE) method further validates its effectiveness in improving land cover classification results. Finally, the best feature combination achieved a classification accuracy of 72.00% and a Kappa coefficient of 0.6543, proving the effectiveness of the feature selection and optimization strategy.

This study aims to estimate the cover classification and carbon respiration of halophytes based on the issues of utilising blue carbon in recent context of climate change. To address the aims, the study classified halophytes(Triglochin maritimum L and Phragmites australis), Intertidal(non-vegetated tidal flats) and Supratidal(sandy tidal flats) to measure carbon respiration and classify cover. The results are revealed that first, the carbon respiration in vegetated areas was less than that in non-vegetated areas. Second, the cover classification could be divided into halophyte communities(Triglochin maritimum L, Phragmites australis), Intertidal and Supratidal by NDWI(Moisture Index, Normalized Difference Water Index) Third, the total carbon respiration of blue carbon was calculated to be –0.0121 Ton km2 hr-1 with halophyte communities at –0.0011 Ton km2 hr-1, Intertidal respiration at –0.0113 Ton km2 hr-1 and Supratidal respiration at 0.0003 Ton km2 hr-1. As this challenge is a fundamental study that calculates the quantitative net carbon storage based on the blue carbon-based marine ecosystem, contributing to firstly, measuring the carbon respiration of cordgrass communities, reed communities, and non-vegetated tidal flats, which are potential blue carbon candidates in the study area, to establish representative values for carbon respiration, secondly, verifying the reliability of cover classification of native halophytes extracted through image classification technology, and thirdly, challenging to create a thematic map of carbon respiration, calculating the area and carbon respiration for each classification category.

Urbanization in Korea has been on the rapid progress, and reached almost 80% in 2000. The phenomenon has caused a reduction in greens and environmental issues. This researcher analyzed land cover maps of the entire region located at Seo-gu, Busan, with the lapse of 10 years in 2002 and 2012 and land cover changes of the two years to look into quantitative and qualitative changes. Therefore, this researcher tries to make a basic investigation on time-series data update of E-GIS DB that will be able to be used as a fundamental material in estaliblishing urban planning.

토지피복도는 환경 관리와 농업 정책 수립에 필수적인 공간정보이지만 국내에서는 여전히 수작업 판독에 의존하고 있어 갱신 주기가 길고 경계 정밀도가 제한적이다. 본 연구는 고해상도 RGB 정사영상과 U-Net 기반 의미 분할 모델을 활용하여 세부 토지피복 분류 및 농경지 경계 검증의 자동화 가능성을 평가하였다. 연구 대상지는 농경 지가 우세하고 인공 구조물이 산재한 경상북도 구미시 해평면이다. 공간해상도 25 cm 정사영상을 512×512 픽셀 단 위로 분할하고 11개 클래스로 수작업 라벨링하여, 총 1,172개 타일을 학습에, 200개를 검증에 사용하였다. 검증 결과, MIoU는 0.3342, 전체 정확도는 0.8157로 나타났다. MIoU 값은 대규모 딥러닝 연구의 일반적 기준보다 낮지만 제한 된 데이터 규모와 클래스 불균형을 고려할 때 의미 있는 결과로 판단된다. 수치지도, 지적도 및 공식 토지피복도와의 시각적 비교를 통해 본 연구의 방법이 필지 단위 농경지 경계와 태양광 시설을 보다 현실적으로 추출함을 확인하였 다. 이러한 결과는 정사영상 기반 딥러닝 기법이 농촌 지역에서 토지피복도 자동 갱신과 농경지 경계 추출에 효과적 으로 활용될 수 있음을 보여준다.

Land cover maps are essential for environmental management and agricultural policy, but in Korea they are still produced mainly through manual interpretation, limiting update frequency and boundary precision. This study assesses the feasibility of automating detailed land cover classification and farmland boundary verification using high-resolution RGB orthophotos and a U-Net semantic segmentation model. The study area is Haepyeong-myeon, Gumi City, Gyeongsangbuk-do, a rural region dominated by agricultural fields with scattered artificial structures. Orthophotos at 25 cm resolution were tiled into 512 × 512 patches and manually labeled into 11 classes. In total, 1,172 tiles were used for training and 200 for validation. The trained U-Net achieved a Mean Intersection over Union(MIoU) of 0.3342 and an overall accuracy of 0.8157 on the validation set. While the mIoU is below typical benchmarks, the results remain meaningful given the limited dataset size and class imbalance. Comparisons with topographic, cadastral, and official land cover maps indicate that the proposed method delineates parcel-level farmland boundaries and solar facilities more realistically. Overall, the findings suggest that orthophoto-based deep learning can support automated land cover map updates and farmland boundary extraction in rural areas.

In heterogeneous landscapes, high-resolution land cover classification is vital for planning, ecological monitoring, and green infrastructure management. This study evaluates the performance of five machine learning algorithms: Decision Tree (DT), Naïve Bayes, Support Vector Machine (SVM), Random Tree (RT), and K-Nearest Neighbors (KNN), using UAV multispectral imagery and object-based image analysis (OBIA). Five scenarios were designed to compare algorithm accuracy. Additionally, a sixth scenario applied the best-performing algorithm to a feature subset selected through Recursive Feature Elimination (RFE), to examine the effect of feature optimization. RT achieved the highest overall classification accuracy (76.75%) and Kappa coefficient (0.7066), while SVM showed limited performance in complex environments. Height features contributed most to accuracy improvements, followed by spectral and geometric features. In class-specific analysis, the Naïve Bayes algorithm yielded the highest Producer’s Accuracy (90.86%) for forest-type land cover but had a lower User’s Accuracy (70.41%), indicating overclassification. In contrast, RT showed more balanced performance (PA = 87.09%, UA = 85.71%), suggesting greater reliability. The results demonstrate the benefits of integrating algorithm selection with feature optimization to improve classification accuracy in complex settings. This approach provides methodological insights for fine-scale mapping of vegetated areas and supports future applications in landscape monitoring and urban green space assessment.

연구목적: UAV기반의 사진측량은 기존 항공촬영에 비해 비용이 절감될 뿐만 아니라 원하는 시간과 장소 에 대한 고해상도의 데이터를 취득하기 용이하기 때문에, 공간정보 분야에서도 UAV를 활용한 연구가 진행 되고 있다. 본 연구에서는 UAV 기반의 고해상도 영상을 활용하여 토지피복 분류를 수행하고자 하였다. 연구방법: 고해상도 영상의 획득을 위하여 RGB카메라를 사용하였으며, 추가적으로 식생지 역을 정확하게 분류하기 위해서 다중분광 카메라를 사용하여 동일 지역을 추가 촬영하였다. 최종적으로 RGB 및 다중분광 카메라를 이용하여 생성된 정사영상, DSM(Digital Surface Model), NDVI(Normalized Difference Vegetation Index), GLCM(Gray-Level Co-occurrence Matrix)을 이용하여 대표적인 감독분류기법인 RF(Random Forest)방법을 이용해 총 7개 클래스 에 대해 토지피복분류를 수행하였다. 연구결과: 분류정확도 평가를 위해 오차행렬을 기반으로 한 정확도 평가를 실시하였으며, 정확도 평가 결과 RGB 영상만을 이용한 감독분류결과와 비교하여 제안 방법이 해당 지역 의 클래스를 효과적으로 분류할 수 있음을 확인하였다. 결론: 본 연구에서 제안한 정사영상, 다중분광영상, NDVI, GLCM을 모두 추가한 경우 기 존의 정사영상만을 이용하였을 때 보다 높은 정확도를 나타냈다. 추후 연구로는 추가적인 입력자료의 개발을 통해 분류 정확도를 향상시키고자 한다.

Purpose: UAV-based photo measurements are being researched using UAVs in the space information field as they are not only cost-effective compared to conventional aerial imaging but also easy to obtain high-resolution data on desired time and location. In this study, the UAV-based high-resolution images were used to perform the land cover classification. Method: RGB cameras were used to obtain high-resolution images, and in addition, multi-distribution cameras were used to photograph the same regions in order to accurately classify the feeding areas. Finally, Land cover classification was carried out for a total of seven classes using created ortho image by RGB and multispectral camera, DSM(Digital Surface Model), NDVI(Normalized Difference Vegetation Index), GLCM(Gray-Level Co-occurrence Matrix) using RF (Random Forest), a representative supervisory classification system. Results: To assess the accuracy of the classification, an accuracy assessment based on the error matrix was conducted, and the accuracy assessment results were verified that the proposed method could effectively classify classes in the region by comparing with the supervisory results using RGB images only. Conclusion: In case of adding orthoimage, multispectral image, NDVI and GLCM proposed in this study, accuracy was higher than that of conventional orthoimage. Future research will attempt to improve classification accuracy through the development of additional input data.

본 연구는 원격탐사 기술을 활용하여 몽골의 항헹티 특별보호지역과 한국의 설악산국립공원 내 산림식생의 탄소 저장량을 평가하였다. 연구 결과, 항헹티 특별보호지역의 2021년 지상부 바이오매스는 약 6천 5백만 톤으로 산정되었으며, 이는 탄소로 환산하면 약 4천 6백만 탄소톤에 해당한다. 반면, 설악산국립공원은 같은 년도에 지상부 바이오매스 약 607만 톤, 탄소 저장량 약 434만 탄소톤으로 추정된다. 단위 면적 당 탄소저장량은 설악산국립공원이 ha 당 104.23 탄소톤, 항헹티 특별보호지역이 51.78 탄소톤으로 설악산국립공원의 탄소 저장량이 더 높은 것으로 나타났다. 항헹티 특별보호지역의 바이오매스는 2010년에 비해 약 4백만 톤의 바이오매스가 감소했으나 설악산국립공원은 큰 차이가 없는 것으로 나타났다. 이러한 결과는 항헹티 특별보호지역에서 연간 약 93만 9천 톤의 이산화탄소 배출로 이어질 수 있음을 시사한 다. 본 연구는 양국의 산림 생태계 관리 및 기후변화 대응 정책 수립에 중요한 참고 자료가 될 것으로 기대된다.

In this study, remote sensing technology was used to assess the carbon stocks of forest vegetation in the Khan Khentii Strictly Protected Area in Mongolia and Seoraksan National Park in South Korea. The aboveground biomass of the Khan Khentii Strictly Protected Area in 2021 was estimated to be approximately 65 million tons, which is equivalent to approximately 46 million tons of carbon. By contrast, the aboveground biomass of Seoraksan National Park in the same year was approximately 6.07 million tons, with a carbon stock of approximately 4.34 million tons. The carbon stock per unit area was higher in Seoraksan National Park (104.23 ton C/ha vs. 51.78 ton C/ha in the Khan Khentii Strictly Protected Area). The biomass of the Khan Khentii Strictly Protected Area had decreased by approximately 4 million tons since 2010, whereas that at Seoraksan National Park showed no significant difference over the same period. These results suggest that the Khan Khentii Strictly Protected Area could emit approximately 939,000 tons of carbon dioxide annually. The findings of this study are expected to serve as an important reference for management of the forest ecosystem and the formulation of climate change response policies in both countries.

The Terra/MODIS data set over Yellow River Basin, China is generated for the purpose of an input parameter into the water resource management model, which has been developed in the Research Revolution 2002 (RR2002) project. This dataset is mainly utilized for the land cover classification and radiation budget analysis. In this paper, the outline of the dataset generation, and a simple land cover classification method, which will be developed to avoid the influence of cloud contamination and missing data, are introduced.

The objectives of this research are as follows. First, to investigate methods for a national-scale land cover map based on multi-temporal classification of MODIS data and multi-spectral classification of Landsat TM data. Second, to investigate methods to p roduce ecological zone maps of Korea based on vegetation, climate, and topographic characteristics. The results of this research can be summarized as follows. First, NDVI and EVI of MODIS can be used to ecological mapping of the country by using monthly phenological characteris tics. Second, it was found that EVI is better than NDVI in terms of atmospheric correction and vegetation mapping of dense forests of the country. Third, several ecological zones of the country can be identified from the VI maps, but exact labeling requires much field works, and sufficient field data and macro-environmental data of the country. Finally, relationship between land cover types and natural environmental factors such as temperature, precipitation, elevation, and slope could be identified.

There are several statistical classification algorithms available for land use/land cover classification. However, each has a certain bias or compromise. Some methods like the parallel piped approach in supervised classification, cannot classify continuous regions within a feature. On the other hand, while unsupervised classification method takes maximum advantage of spectral variability in an image, the maximally separable clusters in spectral space may not do much for our perception of important classes in a given study area. In this research, the output of an ANN algorithm was compared with the Possibilistic c-Means an improvement of the fuzzy c-Means on both moderate resolutions Landsat8 and a high resolution Formosat 2 images. The Formosat 2 image comes with an 8m spectral resolution on the multispectral data. This multispectral image data was resampled to 10m in order to maintain a uniform ratio of 1:3 against Landsat 8 image. Six classes were chosen for analysis including: Dense forest, eucalyptus, water, grassland, wheat and riverine sand. Using a standard false color composite (FCC), the six features reflected differently in the infrared region with wheat producing the brightest pixel values. Signature collection per class was therefore easily obtained for all classifications. The output of both ANN and FCM, were analyzed separately for accuracy and an error matrix generated to assess the quality and accuracy of the classification algorithms. When you compare the results of the two methods on a per-class-basis, ANN had a crisper output compared to PCM which yielded clusters with pixels especially on the moderate resolution Landsat 8 imagery.

Land cover (LC) is an important factor in socioeconomic and environmental studies. According to various studies, a number of LC maps, including global land cover (GLC) datasets, are made using polar orbit satellite data. Due to the insufficiencies of reference datasets in Northeast Asia, several LC maps display discrepancies in that region. In this paper, we performed a feasibility assessment of LC mapping using Geostationary Ocean Color Imager (GOCI) data over Northeast Asia. To produce the LC map, the GOCI normalized difference vegetation index (NDVI) was used as an input dataset and a level-2 LC map of South Korea was used as a reference dataset to evaluate the LC map. In this paper, 7 LC types(urban, croplands, forest, grasslands, wetlands, barren, and water) were defined to reflect Northeast Asian LC. The LC map was produced via principal component analysis (PCA) with K-means clustering, and a sensitivity analysis was performed. The overall accuracy was calculated to be 77.94%. Furthermore, to assess the accuracy of the LC map not only in South Korea but also in Northeast Asia, 6 GLC datasets (IGBP, UMD, GLC2000, GlobCover2009, MCD12Q1, GlobeLand30) were used as comparison datasets. The accuracy scores for the 6 GLC datasets were calculated to be 59.41%, 56.82%, 60.97%, 51.71%, 70.24%, and 72.80%, respectively. Therefore, the first attempt to produce the LC map using geostationary satellite data is considered to be acceptable.

The RapidEye satellite sensor has various spectral wavelength bands, and it can capture large areas with high temporal resolution. Therefore, it affords advantages in generating various types of thematic maps, including land cover maps. In this study, we applied a supervised classification scheme to generate high-resolution land cover maps using RapidEye images. To improve the classification accuracy, object-based classification was performed by adding brightness, yellowness, and greenness bands by Tasseled Cap Transformation (TCT) and Normalized Difference Water Index (NDWI) bands. It was experimentally confirmed that the classification results obtained by adding TCT and NDWI bands as input data showed high classification accuracy compared with the land cover map generated using the original RapidEye images.

LiDAR technology is a combination of laser ranging, satellite positioning technology and digital image technology for study and determination with high accuracy of the true earth surface features in 3 D. Laser scanning data is typically a points cloud on the ground, including coordinates, altitude and intensity of laser from the object on the ground to the sensor (Wehr & Lohr, 1999). Data from laser scanning can produce products such as digital elevation model (DEM), digital surface model (DSM) and the intensity data. In Vietnam, the LiDAR technology has been applied since 2005. However, the application of LiDAR in Vietnam is mostly for topological mapping and DEM establishment using point cloud 3D coordinate. In this study, another application of LiDAR data are present. The study use the intensity image combine with some other data sets (elevation data, Panchromatic image, RGB image) in Bacgiang City to perform land cover classification using neural network method. The results show that it is possible to obtain land cover classes from LiDAR data. However, the highest accurate classification can be obtained using LiDAR data with other data set and the neural network classification is more appropriate approach to conventional method such as maximum likelyhood classification.

This study proposes a classification method based on an automated training extraction procedure that may be used with very high resolution (VHR) images of non-accessible areas. The proposed method overcomes the problem of scale difference between VHR images and geographic information system (GIS) data through filtering and use of a Landsat image. In order to automate maximum likelihood classification (MLC), GIS data were used as an input to the MLC of a Landsat image, and a binary edge and a normalized difference vegetation index (NDVI) were used to increase the purity of the training samples. We identified the thresholds of an NDVI and binary edge appropriate to obtain pure samples of each class. The proposed method was then applied to QuickBird and SPOT-5 images. In order to validate the method, visual interpretation and quantitative assessment of the results were compared with products of a manual method. The results showed that the proposed method could classify VHR images and efficiently update GIS data.

This study presents the use of multi-temporal JERS-1 SAR images to the land cover classification. So far, land cover classified by high resolution aerial photo and field survey and so on. The study site was located in Non-san area. This study developed on multi-temporal land cover status monitoring and coherence information mapping can be processing by L band SAR image. From July, 1997 to October, 1998 JERS SAR images (9 scenes) coherence values are analyzed and then classified land cover. This technique which forms the basis of what is called SAR Interferometry or InSAR for short has also been employed in spaceborne systems. In such systems the separation of the antennas, called the baseline is obtained by utilizing a single antenna in a repeat pass

본 연구에서는 최대우도법과 인공신경망 모형에 의해 카테고리 분류를 수행하고 각각의 분류 성능을 비교 평가하였다. 인공신경망 모형은 오류역전파 알고리즘을 이용한 것으로서 학습을 통한 은닉층의 최적노드수를 결정하여 카테고리 분류를 수행하도록 하였다. 인공신경망 최적 모형은 입력층의 노드수가 7개, 은닉층의 최적노드수가 18개, 그리고 출력층의 노드수가 5개인 것으로 구성하였다. 위성영상은 1996년에 촬영된 Landsat TM-5 영상을 사용하였고, 최대우도법과 인공신경망 모형에 의한 카테고리 분류를 위하여 각각의 카테고리에 대한 분광특성을 대표하는 지역을 절취하였다. 분류 정확도는 인공신경망 모형에 의한 방법이 90%, 최대우도법이 83%로서, 인공신경망 모형의 분류 성능이 뛰어난 것으로 나타났다. 카테고리 분류 항목인 토지 피복 상태에 따른 분류는 두 가지 방법에서 밭과 주거지의 분류오차가 큰 것으로 나타났다. 특히, 최대우도법에 의한 밭에서의 태만오차는 62.6%로서 매우 큰 값을 보였다. 이는 밭이나 주거지의 특성이 위성영상 촬영시기에 따라 나지의 형태로 분류되거나 산림, 또는 논으로도 분류되는 경향이 있기 때문인 것으로 보인다. 차후에 카테고리 분류를 위한 각각의 클래스의 보조적인 정보를 추가한다면, 카테고리 분류 향상이 이루어질 것으로 기대된다.

It is important to analyze the seasonal change profiles of land cover type in large scale for establishing preservation strategy and environmental monitoring. Because the NOAA-AVHRR data sets provide global data with high temporal resolution, it is suitable for the land cover classification of the large area. The objectives of this study were to classify land cover of Korea, to investigate the phenological profiles of land cover. The NOAA-AVHRR data from Jan. 1998 to Dec. 1998 were received by Korea Ocean Research & Development Institute(KORDI) and were used for this study. The NDVI data were produced from this data. And monthly maximum value composite data were made for reducing cloud effect and temporal classification. And the data were classified using the method of supervised classification. To label the land cover classes, they were classified again using generalized vegetation map and Landsat-TM classified image. And the profiles of each class was analyzed according to each month. Results of this study can be summarized as follows. First, it was verified that the use of vegetation map and TM classified map was available to obtain the temporal class labeling with NOAA-AVHRR. Second, phenological characteristics of plant communities of Korea using NOAA-AVHRR was identified. Third, NDVI of North Korea is lower on Summer than that of South Korea. And finally, Forest cover is higher than another cover types. Broadleaf forest is highest on may. Outline of covertype profiles was investigated.

The objective of this research was to investigate the optimal land cover classification algorithm for the monitoring of North Korea with MODIS multi-temporal data based on monthly phenological characteristics. Three frequently used land cover classification algorithms, ISODATA1), SMA2), and SOM3) were employed for this study; the land cover categories were forest, grass, agricultural, wetland, barren, built-up, and water body. The outcomes of the study can be summarized as follows. First, the overall classification accuracy of ISODATA, SMA, and SOM was 69.03%, 64.28%, and 73.57%, respectively. Second, ISODATA and SMA resulted in a higher classification accuracy of forest and agricultural categories, but SOM performed better for the built-up area, bare soil, grassland, and water. A possible explanation for this difference would be related to the difference of sensitivity against the vegetation activity. This would be related to the capability of SOM to express all of their values without any loss of data by maintaining the topology between pixels of primitive data after classification, while ISODATA and SMA retain limited amount of data after normalization process. Third, we can conclude that SOM is the best algorithm for monitoring the land cover change of North Korea.

The objective of this study is to quantitatively evaluate the effects of various SAR speckle reduction methods for multisource land-cover classification by backpropagation neural network, especially over the coastal region. The land-cover classification using neural network has an advantage over conventional statistical approaches in that it is distribution-free and no prior knowledge of the statistical distributions of the classes is needed. The goal of multisource land-cover classification acquired by different sensors is to reduce the classification error, and consequently SAR can be utilized an complementary tool to optical sensors. SAR speckle is, however, an serious limiting factor when it is exploited for land-cover classification. In order to reduce this problem. we test various speckle methods including Frost, Median, Kuan and EPOS. Interpreting the weights about training pixel samples, the “Importance Value” of each SAR images that reduced speckle can be estimated based on its contribution to the classification. In this study, the “Importance Value” is used as a criterion of the effectiveness.