Land monitoring involves systematically understanding changes in land use, leveraging spatial information such as satellite imagery and aerial photographs. Recently, the integration of deep learning technologies, notably object detection and semantic segmentation, into land monitoring has spurred active research. This study developed a web service to facilitate such integrations, allowing users to analyze aerial and drone images using CNN models. The web service architecture comprises AI, WEB/WAS, and DB servers and employs three primary deep learning models: DeepLab V3, YOLO, and Rotated Mask R-CNN. Specifically, YOLO offers rapid detection capabilities, Rotated Mask R-CNN excels in detecting rotated objects, while DeepLab V3 provides pixel-wise image classification. The performance of these models fluctuates depending on the quantity and quality of the training data. Anticipated to be integrated into the LX Corporation's operational network and the Land-XI system, this service is expected to enhance the accuracy and efficiency of land monitoring.

This study was conducted to identify the distribution characteristics of the impervious area in urban watersheds and to reduce the deviation of the impervious area ratio that occurs depending on the degree of construction of land surface condition data. The average impervious area ratio by land use that can be applied to the calculation of the urban impervious area ratio was derived by statistically analyzing the distribution characteristics of the impervious area ratio by land use according to the urban watershed conditions. In urban watersheds, the change in impervious area ratio over the past 20 years has continuously increased in watersheds with an impervious area ratio of less than 60%, and decreased in watersheds with a high impervious area ratio of 60% or more. The average impervious area ratio by land use applicable to the land use technique is “Residential area” 84.0%, “Residential and commercial mix” 93.6%, “Commercial and business facilities” 89.8%, “Industrial land” 84.8%, “Public land” 47.3%, “Transportation facility” 93.3%, “Urban revitalization facility” 61.1%, “Bare land” 17.6%, “Special area” 11.4%, “Forest and open space” 3.5%, “Rivers and lakes” 9.2%. As a result of examining the adequacy of the average impervious area ratio by land use, the difference between the calculated value of the impervious area ratio using land use techniques and the actual impervious area ratio of the biotope map ranged from -3.0%p to 2.6%p at the significance level of 95%. In addition, when the watershed condition is applied, the difference ranged from -2.3%p to 1.7%p. By applying the average impervious area ratio by land use derived in this study, it was found that the impervious area ratio of the target urban watershed could be calculated within a deviation of ±3%p.

Heatwaves are one of the most common phenomena originating from changes in the urban thermal environment. They are caused mainly by the evapotranspiration decrease of surface impermeable areas from increases in temperature and reflected heat, leading to a dry urban environment that can deteriorate aspects of everyday life. This study aimed to calculate daily maximum ground surface temperature affecting heatwaves, to quantify the effects of urban thermal environment control through water cycle restoration while validating its feasibility. The maximum surface temperature regression equation according to the impermeable area ratios of urban land cover types was derived. The estimated values from daily maximum ground surface temperature regression equation were compared with actual measured values to validate the calculation method’s feasibility. The land cover classification and derivation of specific parameters were conducted by classifying land cover into buildings, roads, rivers, and lands. Detailed parameters were classified by the river area ratio, land impermeable area ratio, and green area ratio of each land-cover type, with the exception of the rivers, to derive the maximum surface temperature regression equation of each land cover type. The regression equation feasibility assessment showed that the estimated maximum surface temperature values were within the level of significance. The maximum surface temperature decreased by 0.0450˚C when the green area ratio increased by 1% and increased by 0.0321˚C when the impermeable area ratio increased by 1%. It was determined that the surface reduction effect through increases in the green area ratio was 29% higher than the increasing effect of surface temperature due to the impermeable land ratio.

본 연구는 택지개발사업으로 조성된 단독주택용지가 ‘주택의 노후화와 토지이용의 혼재, 비주거시설의 확산, 공실(빈집) 증가 등으로 주거환경이 악화됨에 따라 범죄가 증가할 것이다’라는 가설 아래 진행되었다. 조성된 지 27년이 지난 단독주택용지를 대상으로 1996년 토지이용실태에 관한 연구 자료와 비교·분석하고, 주거환경과 폭력·절도범죄 발생(2006～2018년)과의 관계를 분석․고찰하였다. 연구 결과, 첫째, 1996년에 비해 2016년 비주거시설이 2배 이상 증가하며 지구 내 집산도로에 접한 주거지 내부로 확산되었고, 업무와 종교, 공실의 비율이 증가하였다. 둘째, 폭력범죄는 통과교통이 발생하는 보조간선과 집산도로에 접한 용도혼재 필지에서 발생비율이 높았고, 제2종근린생활시설 중 주류 판매시설이 밀집한 곳에서 집중되어 발생하였다. 셋째, 절도범죄 중 침입절도범죄율이 높게 나타났고 비주거건축물(39.7%) 보다 주거건축물(60.3%)에서 발생률이 높았다. 넷째, 공실이 있는 건축물에서의 범죄율은 높지 않지만, 공실이 입지한 가로의 인접 건축물과 노상에서 중복적으로 범죄가 발생되었다. 이러한 분석결과는 노후주거지의 주거환경 향상을 위해 주민과 행정 및 경찰이 협력하여 CPTED 사업 추진에 활용할 수 있고, 향후 단독주택용지를 조성함에 있어 ‘안전성’을 확보할 수 있는 가로 및 필지체계 계획과 지구단위계획에 반영할 수 있다.

오늘날 원격탐지기술의 발달로 인해, 산림지역과 같이 피복 분류작업이 난해한 지역을 비롯한 광범위한 지역에서의 세밀한 변화탐지를 위한 고해상도 위성영상 취득이 가능해졌다. 하지만, 고해상도 영상에 대한 시계열분석의 과정에서 많은 양의 지상 관측 데이터가 요구된다. 본 연구에서는 토지피복도를 지상 관측데이터로 활용한 위성영상 분류 방법의 가능성을 시험하였다. 연구대상지는 강원도 원주시이며, 산림지역과 시가화지역이 공존하는 공간이다. 연구 자료는 2015년 3월에 촬영된 KOMPSAT-3A 영상과 2017년도 토지피복도를 이용하여 분류를 시도하였다. 서포트벡터머신 (SVM)과 랜덤포레스트(RF)의 두 가지 상이한 화소기반 분류기법을 적용하여 대상지에 대한 피복분류의 분류정확도를 비교･분석하였으며, SVM 분석의 경우 다수 분석(Majority analysis)을 후속 진행하였다. 분석대상은 산림식생만 포함 한 지역과 연구대상지 전지역으로 구분하였고, 대상 면적이 협소한 습지는 분석과정에서 제외하였다. 분류 결과는 오차 행렬의 전체 정확도가 두 가지 분류대상에 대해 RF 기법이 SVM 기법보다 더 나은 것으로 나타났다. 산림지역만을 대상으로 한 경우, RF 기법이 SVM 기법에 비해 18.3% 높은 값을 나타낸 반면, 전체지역을 대상으로 한 경우는 둘 사이의 간격이 5.5%로 줄어들었다. SVM 기법에 다수 분석 (Majority analysis)을 추가로 실시한 경우, 1% 정도의 정확도 향상이 나타났다. RF 기법은 산림지역의 활엽수를 분석해 내는데 상당히 효과적이었지만, 다른 대상에 대해서는 SVM 기법이 더 나은 결과를 나타내었다. 본 연구는 고해상도 단일시기 영상에 대한 화소 기반의 분류기법을 시험한 것으로, 추후 시계열분석 및 객체기반 분류기법의 추가적인 적용으로 향상된 정확도와 신뢰도를 얻을 수 있을 것으로 판단된다. 이 연구의 방법론은 시공간적으로 고해상도 분석결과를 제공함으로써, 대면적의 토지계획에 유용할 것으로 기대된다.

With the concerns of global climate change, the significant changes in rainfall patterns have aggravated the occurrence of pollutants from agricultural uplands. Heavy and intensive rainfalls have rinsed a variety of pollutants off the land, sending them into the neighboring water environments which further causes water pollution problems. Recently, the application of PAM(Polyacrylamide) in agricultural lands has gotten much attention for soil conditioning. Therefore, this study aimed to develop the automatic control system applying the chemical flocculant, liquid PAM in order to flocculate the suspended solids and deposit at the bottom of farmland water channel, so eventually minimize the amount of non-point pollutants discharged into receiving water. The optimum feed rate for liquid PAM was calculated through the experiment in artificial water channel and applied to the control system. As the results of the performance test of the automatic turbid water control system installed in the sloped field, the range of turbidity reduction rate was 25 to 33% with the average turbidity of 292 to 498 NTU in water channel.

In this study, the land use is analyzed by using the SWMM-LID (Low Impact Development) program to minimize the environmental damage caused by the development. In order to effectively utilize pre - development hydrological conditions, we analyzed the land use of existing industrial complex. The study areas selected were a completed industrial complex and an ongoing industrial complex in order to effectively identify the characteristics of the industrial complex and the water circulation system. Numerical simulation used SWMM-LID to enable quantitative hydrological impact assessment of penetration, storage facilities and LID planning elements. In the case of natural conditions, the infiltration amount was 16.3% and 1.5% of the total rainfall at B, C point, respectively. However, after applying the existing land use plan, the infiltration amount at point B was 12.1% and at point C was 3.9 %. In the case of point B, the amount of infiltration decreased due to the development of greenery as an impervious site. On the other hand, the amount of infiltration at point C increased as the existing industrial complex was replaced by greenery. Therefore, high infiltration amount can be secured when land use plan is redeveloped in green areas or parks in areas where the permeability coefficient is high according to the ground conditions in the complex. Two types of bio-polymer soil were developed to increase the LID effect and were tested to compare typical soil with these bio-polymer soils.

Climate change is the biggest concern of the 21st century. Greenhouse gas (GHG) emissions from various sectors are attracting attention as a cause of climate change. The DeNitrification-DeComposition (DNDC) model simulates GHG emissions from cropland. To study future GHG emissions using this simulation model, various factors that could change in future need to be considered. Because most problems are from the agricultural sector, DNDC would be unable to solve the factor-changing problem itself. Hence, it is necessary to link DNDC with separate models that simulate each element. Climate change is predicted to cause a variety of environmental disasters in the future, having a significant impact on the agricultural environment. In the process of human adaptation to environmental change, the distribution and management methods of farmland will also change greatly. In this study, we introduce some drawbacks of DNDC in considering future changes, and present other existing models that can rectify the same. We further propose some combinations with models and development sub-models.

최근 기후에 대한 관심이 증가하면서 전국단위 뿐만 아니라 지역 단위에서도 기후 지도가 필요하게 되었다. 본 연구에서는 기상청에서 제공하고 있는 무인자동 기상관측장비(AWS) 자료와 LANDSAT 8호 열적외선 영상을 이용하여 지상 기온 분포도를 제작하는 과정을 제시하였다. 지상 기온 분포도 제작을 위하여 기존에 사용되었던 AWS 자료의 공간 보간 기법, 열적외선 영상으로 부터 지표 온도 추출 기법, AWS 자료와 위성영상을 이용한 지상 기온 추출 기법을 비교하고, 지상 기온 분포도에 적합한 지도 제작 기법을 파악하였다. 본 연구의 결과 지상 기온 분포도 제작을 위해 AWS 자료와 위성영상을 이용한 지상 기온 추출 기법이 가장 적합한 것으로 나타났다. 본 연구에서 제시한 과정을 통하여 다양한 지역단위의 기후 지도를 제작할 수 있을 것으로 예상된다.

필리핀의 잠사업은 그 역사가 매우 짧다. 1974년 일본의 잠업기술을 도입하면서부터 양잠에 관심을 갖기 시작하였다. 그후 1990년 한국의 잠업기술자의 도움을 받아 뽕밭을 조성하고 잠실을 지어 양잠을 본격적으로 시작하게 되었다. 따라서 이 나라의 잠업역사가 짧은 것 만큼 뽕나무 재배기술도 몹시 낙후되어 있는 실정이다. 필리핀 고지대의 뽕밭의 현황 및 그 개선책을 요약하면 다음과 같다. 강우량, 기온, 일조시간 및 습도 등은 뽕나무가 자라는데 큰 무리가 없어 보인다. 다만 12월부터 3월까지의 건기에는 뽕밭에 관수를 하거나 뽕밭 주위에 무진장으로 자생하고 있는 풀을 베어서 뽕밭에 피복하여 토양수분을 보존한다. 대부분의 뽕밭이 급경사지에 조성되었는데 등고선에 따라 계단을 만들거나 승수로의 설치등이 되어 있지 않기 때문에 강우시에는 토양의 침식 및 유실이 심하였다. 따라서 계단사면에는 초생재배를 실시하고 계단면에는 유기물을 피복하여 토양의 유실을 방지해야 할 것이다. 토양의 pH 4.7, 유기물함량 1.6%, 유효인산함량은 6ppm이었다. 즉 유기물 함량이 낮은 강산성토양이며 유효인산함량이 극히 낮은 토양이었다. 석회와 유기물을 다량 사용하여 토양을 개량하고 인산비료를 증시해야한다. 소나무 밑에서 재배하고 있는 뽕밭에서는 햇빛이 잘 쪼일 수 있도록 소나무의 곁가지를 전정해주고 석회를 증시한다. 현재의 ha당 시비량 질소-46kg, 인산-18kg, 칼리-18kg은 너무 적은 양이므로 ha당 질소-250kg, 인산-120kg, 칼리-120kg을 시용한다. 전체 뽕밭의 70%에 해당하는 뽕나무가 뽕나무 잎말이 나방피해를 받고 있는 실정인바 DDVP 및 KAFIL 등의 약제 방제를 실시한다. 우기에 많이 발생하는 적삽병은 이병된 뽕잎과 새순은 조기 제거하고 Topsin-M을 살포한다.

본 연구에서는 소양강댐 유역의 수문기상인자들의 시공간적 변동성을 명확히 파악하기 위하여 지표해석모형을 구축하였다. 지표해석 모형으로 는 Variable Infiltration Capacity (VIC) 모형을 사용하였으며, 모형의 공간 해상도는 10 km, 시간 해상도는 1일로 정하였다. 2007~2010년 기 간의 일유량자료를 바탕으로 Isolated Particle Swarm Optimization 알고리즘을 사용하여 모형의 7개 매개변수를 보정하였고, 2011~2014년 기간의 일유량자료를 사용하여 모형을 검증하였다. 보정된 모형은 보정기간과 검증기간 모두에 대하여 0.90의 Nash-Sutcliffe Coefficient값과 0.95의 상관계수를 보였다. 소양강댐유역에 대하여 산출된 인자들은 여름철에 강우가 집중되어있는 우리나라의 계절적인 특성과 기온변화로 인한 장 ․ 단파 복사량의 변화와, 지표면 온도의 변화, 이로 인해 피복층에서의 증발과 식생 증산의 변화가 고려되어 총 증발산이 변화하는 경향이 잘 반영된 것으로 나타났다. 산출된 수문인자를 검증하기 위하여 지상관측토양수분자료와 비교하였다. 겨울철을 제외한 4~11월의 비교결과 두 자료의 추세선의 기울기는 1.087로 나타났고, 상관계수는 0.723의 값을 나타냈다. 이러한 본 연구의 결과는 지표해석모형이 우리나라 주요 댐 유역의 수문기상인자의 시공간적인 변화를 정확히 파악하는데 활용될 수 있으며, 나아가서는 더욱 정밀하고 효율적인 수자원계획을 수립하는 데에도 활 용될 수 있다는 점을 시사한다.

Reckless development policy, which has continued during the past several decades, caused landscape damage. In order to solve this problem, it is very urgent to conserve the habitat and to establish systematic database. Recently, various researches related to landscape damage have been conducted actively and the necessity of landscape ecological planning based on the biotop mapping has been issued as one of the solutions. Therefore, the purpose of this study is to select the study area, which is located in Guji-meon, Dalseong-gun, Daegu and will be formed into Daegu Science Park, and then to develop and apply the landscape ecological planning model based on biotop mapping in district unit. The results are as follows. Fist of all, according to the results of biotop classification with field survey and literature review, total 13 biotop and total 63 biotop types were classified. According to the first evaluation with B-VAT, total 19 biotop types were classified I grade including natural river with abundant plants(BA) and so on. While V grade, which has lowest value, was classified 16 biotop types including vegetable garden adjacent to river(BC). Also the second evaluation, we analyzed total 30 areas such as 1a, 1b grades, which had special meaning for the conservation of species and biotop. Next, 2a, 2b, 2c grades were analyzed total 82 areas. Secondly, in the compliance with Ecological landscape planning, we divided landscape planning model into two parts such as setting of improvement goal and detailed implement plan. For instance, setting of improvement goal, we classified 9 parts such as conservation region and parts of detailed implement plan, we classified total 28 planning indicators with 4 view such as the region which should be special protected from nature and landscape. Lastly, with the developed landscape model we applied to research areas, made maps, and differentiate proposals in each region. Specially, the final master plan was made to help understand the contents of detailed landscape planning. This study will be a useful data, which can solve the landscape damage problem systematically and control it landscape-friendly with biotop classification, evaluation, landscape ecological planning model, and application method which we developed.

Land suitability assessment assesses development, farming, and conservation suitability, considering land's soil, location, and possibility for use. It also implement segmentation of management regions into production, conservation, and plan management area. It is evaluated as a very significant system in establishing a land use system of sustainable development and development after planning in the aspect that it can establish proper land use plan. This study developed a recommendation model for development in agent-based model that interacts with surrounding lands. It also tried to summarize the area characteristic analysis and the results of land suitability evaluation, targeting three ri's in Yesan-Gun, and analyze the model's applicability by selection results. In order to recommend area for development that considers the use of the surrounding lands, it calculated development possibility indices that considered the ratings of all the lands in the target areas for each parcel and simulated the model. As a result, selected three areas in target region were suitable areas for development in land suitability assessment. In detail, ratings of the recommended parcels were 3, 4, and 5 ratings. As a result of examining the land status, it showed that all the three areas were plan management areas, thus easy for development. It is judged that the model for recommending area for development suggested in this study can be used as important basic data for setting the direction for development when establishing a regional planning.

The purpose of this study is to develop the landscape resource assessment system(LRAS) to help evaluate the value of landscape resources(Jeju Island's Stonewall fencing farming land) for the introduction of direct payment system of the landscape preservation objectively and to applicate the model in the fields. Delphi survey on Jeju's stonewall experts shows that the order of priority among value evaluation elements on Jeju's stonewall is its harmony with surroundings(34%), the preservation of its original state(34%) and its density and scale(32%). Evaluation system development of landscape resources(Jeju's stonewall fencing filming land) and field observation survey utilizing it consists of the following five steps. Step 1 includes the first Delphi survey on Jeju's stonewall experts to decide its value evaluation elements and their priority. Step 2 is the second Delphi survey on Jeju's stonewall experts to grade pictures of landscape resources(Jeju's stonewall fencing farming land) on the basis of expert-proposed value evaluation standards. Step 3 consists of analysis work using the result of Delphi survey on experts. Step 4 is to select five grade standard pictures according to each of three grading elements of A, B, C belonging to each of the three standards. Then, it is necessary to make panels including five A-grade pictures, five B-grade pictures and five C-grade pictures according to each of the three elements of density, harmony, and original state preservation. Step 5 consists of field observation survey. According to the result of few experts' value evaluation of stonewall fencing farming land with the aid of NRAS developed in this research, the area of Pyeongdae-ri is ranked first, and then the area of Bukcheon-ri, Chocheon-up, the area of Gwakgi-ri, Ewol-up, the area of Shinum-ri, Ewol-up and the area of Yongsu-ri, hankyung-Myun are ranked in the order named. When those areas are graded, A Grade Areas includes the areas of Pyeongdae-ri, the area of Bukcheon-ri, the area of Gwakgi-ri, B Grade Areas consist of the area of Shinum-ri and the area of Yongsu-ri, and the areas of Onpyeong, wimi and youngrak belong to C Grade Area.