• 한국컴퓨터정보학회논문지
• /
• 제26권6호
• /
• pp.29-35
• /
• 2021

본 논문에서는 위치 측위의 정확도를 높일 수 있는 방안으로 KNN(K-Nearest Neighbor)과 Local Map Classification 및 Bayes Filter를 융합한 기법을 제안한다. 먼저 이 기법은 Local Map Classification이 실제 지도를 여러 개의 Cluster로 나누고, 다음으로 KNN으로 Cluster들을 분류한다. 그리고 Bayes Filter가 획득한 각 Cluster의 확률을 통하여 Posterior Probability을 계산한다. 이 Posterior Probability으로 로봇이 위치한 Cluster를 검색한다. 성능 평가를 위하여 KNN과 Local Map Classification 및 Bayes Filter을 적용하여서 얻은 위치 측위의 결과를 분석하였다. 분석 결과로 RSSI 신호가 변하더라도 위치 정보는 한 Cluster에 고정되면서 위치 측위의 정확도가 높아진다는 사실을 확인하였다.

• The Korean Journal of Ecology
• /
• 제26권2호
• /
• pp.75-81
• /
• 2003

This study was performed to prove if the high resolution satellite imagery of IKONOS is suitable for preparing digital vegetation map which is becoming increasingly important in ecological science. Seven classes for forest area and five classes for non-forest area were taken for classification. Three methods, such as the pixel based classification, the segment based classification with majority principle, and the segment based classification with maximum likelihood, were applied to classify IKONOS imagery taken in April 2000. As a whole, the segment based classification shows better performance in classifying the high resolution satellite imagery of IKONOS. Through the comparison of accuracies and kappa values of the above 3 classification methods, the segment based classification with maximum likelihood was proved to be the best suitable for preparing the vegetation map with the help of IKONOS imagery. This is true not only from the viewpoint of accuracy, but also for the purpose of preparing a polygon based vegetation map. On the basis of the segment based classification with the maximum likelihood, a digital vegetation map in which each vegetation class is delimitated in the form of a polygon could be prepared.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2003년도 Proceedings of ACRS 2003 ISRS
• /
• pp.198-200
• /
• 2003

This study proved if the high resolution satellite imagery of IKONOS is suitable for preparing digital forest cover map. Three methods, the pixel based classification with maximum likelihood (PML), the segment based classification with majority principle(SMP), and the segment based classification with maximum likelihood(SML), were applied to classify and delimitate forest cover of IKONOS imagery taken in May 2000 in a forested area in the central Korea. The segment-based classification was more suitable for classifying and deliminating forest cover in Korea using IKONOS imagery. The digital forest cover map in which each class is delimitated in the form of a polygon can be prepared on the basis of the segment-based classification.

• 대한원격탐사학회지
• /
• 제34권4호
• /
• pp.671-680
• /
• 2018

Compared to the imagery produced by traditional satellites, PlanetScope satellite imagery has made it possible to easily capture remotely-sensed imagery every day through dozens or even hundreds of satellites on a relatively small budget. This study aimed to detect changed areas and update a land cover map using a PlanetScope image. To generate a classification map, pixel-based Random Forest (RF) classification was performed by using additional features, such as the Normalized Difference Water Index (NDWI) and the Normalized Difference Vegetation Index (NDVI). The classification result was converted to vector data and compared with the existing land cover map to estimate the changed area. To estimate the accuracy and trends of the changed area, the quantitative quality of the supervised classification result using the PlanetScope image was evaluated first. In addition, the patterns of the changed area that corresponded to the classification result were analyzed using the PlanetScope satellite image. Experimental results found that the PlanetScope image can be used to effectively to detect changed areas on large-scale land cover maps, and supervised classification results can update the changed areas.

• 한국환경복원기술학회지
• /
• 제5권1호
• /
• pp.35-50
• /
• 2002

This study aims to understand wetland distribution and type-specific classification features with a focus on Tumen River downstream in China by adjusting and improving the classification system used in Korea with a reference to international wetland classification systems and their criteria & methods. In this study, wetland types were determined based on hydrology, vegetation, and soil conditions, which are the most basic elements of wetlands. Also, topography analytical map, vegetation analytical map, and soil analytical map for wetland classification were developed and used based on currently available topography map, vegetation map, and soil map. In addition, codes were defined based on topography, location, hydrology, and vegetation. The result shows that, in the Tumen River downstream, wetlands are often found near natural revetment and terrace land & river-bed lakes. In the discovered wetlands, riverine, lacustrine, and inland wetlands were mostly found at system level. Riparian and human-made wetlands were also identified. At a sub-system level, perennial and seasonal wetlands were found to a similar degree. At a class level, perennial open water, herbal plants, and shrubs were mostly found and sandy plain, hydrophytes, and forest tree types were also observed. An overall detailed classification shows that a total of 17 wetland types were found and a large distribution of sand dunes and river-bed lakes, which are scarce in Northeast Asia, indicates that other rare wetland types such as palustrine seasonal sand plain wetland and lacustrine seasonal sand plain wetland may be discovered.

• 한국군사과학기술학회지
• /
• 제13권1호
• /
• pp.147-155
• /
• 2010

Landcover information of access-denied area was extracted from low-medium and high resolution satellite image. Training for supervised classification was performed to refer visually by landcover map which is made and distributed from The Ministry of Environment. The classification result was compared by relating data of FACC land classification system. As we rasterize digital military map with same pixel size of satellite classification, the accuracy test was performed by image to image method. In vegetation case, ancillary data such as NDVI and image for seasons are going to improve accuracy. FACC code of FDB need to recognize the properties which can be automated.

• 한국산학기술학회논문지
• /
• 제14권8호
• /
• pp.3748-3759
• /
• 2013

본 연구의 목적은 DINA(deterministic-input, noisy "and" gate)모형에서 최대우도(maximum likelihood: ML), 최대사후확률(maximum a posteriori: MAP), 사후기대(expected a posteriori: EAP)방법들의 분류 정확성이 어느 정도인가를 알아보는 것이다. 연구 목적을 달성하기 위하여 다양한 모의실험 조건들[인지요소의 수(K= 5, 7), 응시생 능력분포(고능력, 중간능력, 저능력 집단), 검사 길이(J= 15, 30, 45)]에 따라 모의자료를 생성했다. 응시생 분류 정확성을 평가하기 위한 준거로 참 인지요소(true ${\alpha}$)와 ML, MAP, EAP방법으로 추정된 인지요소가 어느 정도 일치하는지를 계산했다. 본 연구의 주요결과를 요약하면 다음과 같다. 첫째, 본 연구에서 설정한 검사 조건에서 ML, MAP방법보다 EAP방법의 정확일치도 평균이 높았다. 둘째, 다른 검사 조건이 동일할 때, 인지요소의 수가 증가하면 ML, MAP, EAP방법 모두에서 정확일치도 평균이 낮아졌다. 셋째, 동일한 검사 길이에서 사전분포로 고능력, 중간능력, 저능력 집단을 각각 가정했을 때 ML, MAP방법보다 EAP방법의 정확일치도 평균이 높았다. 넷째, 동일한 응시생 능력분포에서 검사 길이가 증가하면 ML, MAP, EAP방법 모두에서 정확일치도 평균이 높아졌다. 인지요소의 수에 따라 응시생을 정확하게 분류하기 위한 적절한 검사 길이를 보면, 인지요소의 수가 5, 7개이고 이에 대응하는 검사 길이가 각각 30, 45문항일 때 본 연구에서 설정한 높은 분류 정확성 기준에 부합하는 것으로 나타났다.

• 대한수학회지
• /
• 제35권2호
• /
• pp.315-330
• /
• 1998

Chen and Piccinni [7] have classified all compact surfaces in a Euclidean space $R^{2＋p}$ with 1-type generalized Gauss map. Being motivated by this result, the purpose of this paper is to consider the Lorentz version of the classification theorem and to obtain a complete classification of space-like surfaces in indefinite Euclidean space $R_{p}$ $^{2＋p}$ with 1-type generalized Gauss map.p.

• 한국환경복원기술학회지
• /
• 제18권1호
• /
• pp.127-133
• /
• 2015

Interests in ecosystem services have increased and a number of attempts to perform a quantitative valuation on them have been undertaken. To classify the ecosystem types landcover classification maps are generally used. However, some forest types on landcover classification maps have a number of errors. The purpose of this study is to verify the forest types on the landcover map by using a variety of field survey data and to suggest an improved method for forest type classifications. Forest types are compared by overlaying the landcover classification map with the 4th forest type map, and then they are verified by using National Forest Inventory, 3rd National Ecosystem Survey and field survey data. Misclassifications of forest types are found on the forest on the forest type map and farm and other grassland on the landcover map. Some errors of forest types occur at Daegu, Busan and Ulsan metropolitan cities and Gangwon province. The results of accuracy in comprehensive classification show that deciduous forest is 76.1%; coniferous forest is 54.0%; and mixed forest is 22.2%. In order to increase the classification accuracy of forest types a number of remote sensing images during various time periods should be used and the survey period of NFI and the National Forest Inventory and National Ecosystem Survey should be consistent. Also, examining areas with wide forest patch should be prioritized during the field survey in order to decrease any errors.

• 대한전자공학회논문지SP
• /
• 제44권1호
• /
• pp.40-48
• /
• 2007

이 논문은 다중 스케일 베이지안 관점에서 다층 퍼셉트론과 마코프 랜덤 필드를 사용한 새로운 결 분할 방법을 제안한다. 다층 퍼셉트론의 출력은 사후 확률을 모델링하므로 본 논문에서는 다중 스케일 웨이블릿 계수들을 다층 퍼셉트론의 입력으로 사용한다. 다층 퍼셉트론으로부터 구한 사후 확률과 MAP (maximum a posterior) 분류를 이용하여 각 스케일에서 결 분류를 수행한다. 또한 가장 섬세한 스케일에서 더 개선된 분할 결과를 얻기 위하여 모든 스케일에서 MAP 분류 결과들을 거친 스케일에서 섬세한 스케일까지 차례로 융합한다. 이런 과정은 한 스케일에서의 분류 정보와 그 인접한 보다 거친 스케일에서 얻어지는 문맥과 관련한 연역적 정보를 이용하여 MAP 분류를 행함으로써 이루어진다. 이 융합 과정에서, MRF (Markov random fields) 사전 모델이 평탄화 제한자로서 동작하고, 깁스 샘플러 (Gibbs sampler)는 MAP 분류기로서 동작한다. 제안한 분할 방법은 HMT (Hidden Markov Trees) 모델과 HMTseg 알고리즘을 이용한 결 분할 방법보다 더 좋은 성능을 보인다.