• 한국BIM학회 논문집
• /
• 제13권4호
• /
• pp.96-105
• /
• 2023

This study focuses on developing a building object recognition technology for efficient use in the remodeling of buildings constructed without drawings. In the era of the 4th industrial revolution, smart technologies are being developed. This research contributes to the architectural field by introducing a deep learning-based method for automatic object classification and recognition, utilizing point cloud data. We use a TD3D network with voxels, optimizing its performance through adjustments in voxel size and number of blocks. This technology enables the classification of building objects such as walls, floors, and roofs from 3D scanning data, labeling them in polygonal forms to minimize boundary ambiguities. However, challenges in object boundary classifications were observed. The model facilitates the automatic classification of non-building objects, thereby reducing manual effort in data matching processes. It also distinguishes between elements to be demolished or retained during remodeling. The study minimized data set loss space by labeling using the extremities of the x, y, and z coordinates. The research aims to enhance the efficiency of building object classification and improve the quality of architectural plans by reducing manpower and time during remodeling. The study aligns with its goal of developing an efficient classification technology. Future work can extend to creating classified objects using parametric tools with polygon-labeled datasets, offering meaningful numerical analysis for remodeling processes. Continued research in this direction is anticipated to significantly advance the efficiency of building remodeling techniques.

• 한국측량학회:학술대회논문집
• /
• 한국측량학회 2003년도 춘계학술발표회 논문집
• /
• pp.277-281
• /
• 2003

We fused RADARSAT image and SPOT panchromatic image by wavelet transform in order to improve the accuracy of classification on the water area. Fused image in water not only maintained the characteristic of SAR image (low pixel value)but also had boundary information improved. This leads to accurate method to classify water areas.

• 한국환경복원기술학회지
• /
• 제17권2호
• /
• pp.15-30
• /
• 2014

Systematic management of wetlands should be a priority to build the data for the extent and distribution of wetlands all over the country. However there are no clear guidelines for the wetland boundary delineation, so researchers have to determine the boundary of wetlands in each different way. As a result, it is very difficult to identify the extent and distribution of wetlands. This study proposes applicable criteria of setting boundary of wetlands which consider their wetland vegetation and geographical characteristics, according to wetland classification. The proposed site in this study is selected wetlands that represent each wetland type and have been ecologically well preserved like the wetland protected areas. GIS data for setting the boundary of wetlands selected were land-cover maps, aerial photographs, high resolution satellite images, and digital topographic maps. In this study, 'wetland unit determination' of the Washington State Wetlands Rating System(WSDE, 1993) and the concept of 'Wetland and Deep-water Habitats' was suggested by Wetland Delineation Manual(USACE, 1987) were used as criteria for setting the boundary of wetlands. As a result, it was found that the boundary of wetlands could be, in general, set consistently. Also, it seemed possible to set systematic and standardized boundary of wetlands and to provide more objective data for establishing national wetland policies, if maps of wetlands are made and an investigation of wetlands is implemented according to the criteria.

• 대한불안의학회지
• /
• 제1권1호
• /
• pp.18-24
• /
• 2005

The Diagnostic and Statistical Manual 4th edition (DSM-IV) has been widely accepted and used for international classification of mental disorder. The DSM has been changed to improve diagnostic reliability and validity through descriptive and categorical approaches which was undertaken atheoretically. The authors reviewed current studies about the DSM-IV classification system and the diagnostic issues of representative categories of anxiety disorder. The authors concluded that the anxiety disorder classification system in DSM-IV has limitations such as a lack of empirical consideration for overlapping features of anxiety disorders and a lack of discriminant validity. To improve diagnostic validity and revise the current DSM-IV classification system, the authors suggested 1) more longitudinal studies for collecting empirical evidence, 2) decreasing the dependence upon operational criteria, 3) deceasing diagnostic boundary blurring, 4) developing disease specific biological diagnostic techniques and 5) continued collaboration between the DSM and International Classification of Diseases (ICD) systems.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2008년도 International Symposium on Remote Sensing
• /
• pp.141-144
• /
• 2008

Many studies have been conducted on extracting buildings from ALS(Airborne Laser Scanning) data. After segmentation or classification of building points, additional steps such as generalization is required to get straight boundary lines that better approximate the real ones. In much research, orthogonal constraints are used to improve accuracies and qualities. All the lines of the building boundaries are assumed to be either parallel or perpendicular mutually. However, this assumption is not valid in many cases and more complex shapes of buildings have been increased. A new algorithm is presented that is applicable to various complex buildings. It consists of three steps of boundary tracing, grouping, and regularization. The performance of our approach was evaluated by applying the algorithm to some buildings and the results showed that our proposed method has good potential for extracting building boundaries of various shapes.

• 한국지능시스템학회:학술대회논문집
• /
• 한국퍼지및지능시스템학회 2006년도 추계학술대회 학술발표 논문집 제16권 제2호
• /
• pp.251-254
• /
• 2006

본 논문은 multiclass 문제에서 기존에 나와 있는 fuzzy support vector mahchines 이 decision boundary 를 설정하는데 있어 모든 훈련 데이터에 대해서 바람직한 decision boundary 를 만들지 못하므로 그러한 경우를 예로 제시한다. 그리고 그에 대한 개선점으로 밀도를 이용해 decision boundary 를 조정하여 기존 FSVM 의 decision boundary 보다 더 타당한 decision boundary 를 설정하는 것을 보인다.

• 대한지리학회지
• /
• 제30권3호
• /
• pp.296-310
• /
• 1995

본 연구는 퍼지이론을 공간분석에 적용하기 위한 이론적인 배경을 고찰하고, 퍼지 분류법의 특성에 대해 살펴본 것이다. 이를 위해 필자는 공간정보의 모호성에 대해 살펴보 고, 퍼지공간분석의 전제를 설정한 다음 퍼지분류법을 소개하였다. 그리고 퍼지분류법의 특 성을 명확히 하기 위해 경상남도 읍급이상 도시의 산업별 고용비율을 대상으로 퍼지분류를 행한 후, 퍼지분류와 전통적인 군집분석의 결과를 비교하였다. 그 결과, 공간정보의 모호성 은 구체성의 부족, 인간행태, 인내치문제, 분류기준의 부족 등에 의해 발생하는데 기존의 공 간분석기법으로는 공간의 모호성을 반영할 수 없으므로 퍼지기법을 도입한 퍼지공간분석의 필요성이 있음을 확인하였다. 퍼지분류법 중, 퍼지이산분류는 계산절차는 상대적으로 간단하 나 분류결과가 집단간의 점이성을 고려하지 못하며, 퍼지중첩분류는 분류집단간의 점이성은 고려하나 분류결과가 지나치게 많아 적절한 분류수준을 선택하기 어렵고 결과해석이 상대적 으로 난해하다는 문제점이 있음이 밝혀졌다, 또 경남의 도시기능분류는 분류기법에 따라 다 르게 이루어졌지만 창원, 울산, 마산, 진해, 김해, 양산, 웅상, 장승포, 신현으로 구성된 제조 업 군집과 단독군집 충무의 존재가 세 가지 분류 모두에서 공통적으로 확인되었다.

• 한국음향학회지
• /
• 제18권4호
• /
• pp.60-64
• /
• 1999

본 논문은 연속음 인식과 합성을 위한 운율 경계강도 예측모델을 위해 최근에 널리 사용되고 있는 방법으로 분류·회귀트리라 불리우는 CART(Classification And Regression Tree) 와 OC1(Oblique Classifier1)을 적용한다. 운율 경계강도 수준을 4로 하고 문법적인 특징으로는 트리구조 방법으로 결정된 오른쪽 가지의 수식의 깊이(Rd)와 link grammar 방법으로 결정된 연결거리(To_Right)를 tri_gram모형과 결합하여 CART와 OC1에 적용해 각각 운율 경계강도를 예측, 비교한다. 실험을 통하여 OC1 방법이 CART 방법에 비해 더 적은 터미널 노드에 더 향상된 예측율을 보임을 확인할 수 있다.

• International journal of advanced smart convergence
• /
• 제7권4호
• /
• pp.57-65
• /
• 2018

Artificial neural network is inspired by the biological neural network. For simplicity, in computer science, it is represented as a set of layers. Many research has been made in evaluating the number of neurons in the hidden layer but still, none was accurate. Several methods are used until now which do not provide the exact formula for calculating the number of thehidden layer as well as the number of neurons in each hidden layer. In this paper model selection approach was presented. Proposed model is based on geographical analysis of decision boundary. Proposed model selection method is useful when we know the distribution of the training data set. To evaluate the performance of the proposed method we compare it to the traditional architecture on IRIS classification problem. According to the experimental result on Iris data proposed method is turned out to be a powerful one.

• 정보처리학회논문지B
• /
• 제16B권6호
• /
• pp.479-488
• /
• 2009

본 논문은 기존의 목표항목만을 위주로 한 학습체계에서 발생하는 오분류 문제의 해결을 위해 기존의 학습체계에 경계항목을 자동으로 탐 색하여 포함시켜 확대시키는 방법을 제안하고 있다. 여러 주제에 걸쳐 다양한 내용을 다루는 복잡한 문서들은 확실히 어느 범주로 분류해야 할 지 판가름하기 어려운 성질인 모호성이 강하다. 이러한 경우 모든 경우들을 정확히 구분할 수 있는 최적의 경계를 찾는 일은 더욱 어려운 일이 다. 복잡하고 불확실성이 높은 데이터들의 특징은 대부분 분류 경계영역에 위치하므로 이러한 분류경계의 데이터들을 새로운 학습 항목으로 인 식시키도록 하는 것이 필요하다. 본 연구에서는 주어진 목표항목 사이의 경계항목을 자동으로 탐색하여 학습체계에 추가하는 학습 체계 확장 알고리즘을 제시하고, 의도적인 학습오류를 발생시킨 후 기존방법과의 비교실험을 수행함으로써 제안방법의 정확성과 안정성을 비교하였다. 실 험결과 경계범주를 포함하여 학습 체계를 확장시켰을 때의 예측력은 기존 0.70에서 0.86으로 약 24% 향상 되었고, 오류를 포함시켰을 때의 예 측력은 기존 0.52에서 0.79로 약 49% 향상되었다.