• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.12
• /
• pp.1588-1594
• /
• 2020

In this paper, we discuss a data balancing method for learning a neural network that generates digital holograms using a deep neural network (DNN). Deep neural networks are based on deep learning (DL) technology and use a generative adversarial network (GAN) series. The fringe pattern, which is the basic unit of a hologram to be created through a deep neural network, has very different data types depending on the hologram plane and the position of the object. However, because the criteria for classifying the data are not clear, an imbalance in the training data may occur. The imbalance of learning data acts as a factor of instability in learning. Therefore, it presents a method for classifying and balancing data for which the classification criteria are not clear. And it shows that learning is stabilized through this.

• Journal of Internet of Things and Convergence
• /
• v.8 no.3
• /
• pp.69-74
• /
• 2022

Big data related analysis techniques, the prescriptive analytics methodology improves the performance of passive learning models by ensuring that active learning secures high-quality learning data. Prescriptive analytics is a performance maximizing process by enhancing the machine learning models and optimizing systems through active learning to secure high-quality learning data. It is the best subscription value analysis that constructs the expensive category data efficiently. To expand the value of data by collecting research field, research propensity, and research activity information, customized researcher through prescriptive analysis such as predicting the situation at the time of execution after data pre-processing, deriving viable alternatives, and examining the validity of alternatives according to changes in the situation Provides research information service.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.408-408
• /
• 2023

하천공간은 하도, 사주, 식생, 하천구조물 등에 대한 특성을 지니고 있으며, 현장조사를 통해 하천공간에 대한 자료를 분석하여 기초자료를 생산한다. 기존에는 현장에서 육안조사나 지상에서 사진촬영, 스케치방법으로 하천공간특성에 대한 조사를 수행하였으나, 지상에서 조사한 자료은 하천특성에 대한 물리적·공간적 특성을 파악하기 어렵고 자료의 활용성이 낮은 한계점이 존재한다. 이와 같은 한계를 극복하기 위해 GIS 및 RS 기술을 활용한 고도화된 첨단조사 기술 및 장비가 도입되어 활용되고 있다. 본 연구에서는 하천공간특성을 GIS 기반으로 객체지향 분류 적용 연구와 분류 항목에 따른 공간분석 연구를 수행하였다. 연구를 위한 대상지역은 섬진강권역의 지석천 유역 하류부에 위치하고 있는 지석천 친수공원을 대상으로 선정하였다. 대상지역의 고해상도 항공영상을 수집 및 정합한 후 QGIS에서 제공하는 Orfeo ToolBox(OTB)의 LSMS(Large Scale Mean-Shift) 기법으로 정합한 항공영상의 객체지향 영상분할을 실시하여 벡터 레이어를 생성하였고, 하천공간특성에 따른 항목을 선정하여 각 항목의 영역에 대한 선별을 통해 훈련데이터를 생성하였다. 훈련데이터는 랜덤 포레스트를 이용하여 각 항목에 대한 자동 분류를 확인하였으며, 하천공간특성의 정량적 평가를 위해 분류된 각 항목별 공간분석을 통해 면적, 위치정보(위도, 경도, 표고)를 산정하였다. 분석 결과, 하천공간특성을 GIS 기반의 벡터 레이어와 각 항목에 대한 정량적 분석을 통해 하천공간의 DB를 구축하였다. 이와 같이 하천공간 DB 구축을 통해 전국 하천관리체계를 위한 기초자료를 구축하고자 하였다.

• Journal of the Korea Convergence Society
• /
• v.13 no.1
• /
• pp.189-199
• /
• 2022

While modern illustration designs require originality, the shape and color of marine life provide rich materials for illustration creation. Illustration has a very high artistic creation value. Based on the knowledge of color science and visual form design, this study takes illustration design and marine biology as the main research object. The purpose of this study is: ① The typical features of Marine life are sorted by color and form. ② Lead out graphic symbols representing Marine life. ③ Combine Marine life with illustration design to make design cases. Through research, it is highlighted that the curve feature is a typical feature for distinguishing marine organisms from terrestrial organisms. The classification and comparative analysis of the color phenomena of tropical fish help to extract and process the color of marine tropical fish on the basis of understanding the color characteristics of tropical fish, and apply it to the illustration design.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.1D
• /
• pp.133-140
• /
• 2006

This paper suggests to reclassify railroad accident types and to standardize them as the standardized code for the railroad safety management system. The existing railroad accident types in both domestic and foreign cases have been carefully analyzed in the beginning. Based on the case studies, the new railroad accident types are classified into 9 classes which are not overlapped one another and 9 classes have been subdivided into 40 different accident patterns. All these patterns are linked with 9 different accident objects and 6 accident locations. Therefore, this study suggested the combination of 4 distinct code factors: accident class, accident pattern, accident object, and accident location to standardize them. In addition, inter-operation between the proposed codes and the existing accident types is suggested. This code will play a major role in the railroad safety management system composed of accident prevention, accident preparedness, accident response, and accident recovery.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.792-799
• /
• 2022

Building Information Modeling (BIM) technology is a key component of modern construction engineering and project management workflows. As-is BIM models that represent the spatial reality of a project site can offer crucial information to stakeholders for construction progress monitoring, error checking, and building maintenance purposes. Geometric methods for automatically converting raw scan data into BIM models (Scan-to-BIM) often fail to make use of higher-level semantic information in the data. Whereas, semantic segmentation methods only output labels at the point level without creating object level models that is necessary for BIM. To address these issues, this research proposes a hybrid semantic-geometric approach for clutter-resistant floorplan generation from laser-scanned building point clouds. The input point clouds are first pre-processed by normalizing the coordinate system and removing outliers. Then, a semantic segmentation network based on PointNet++ is used to label each point as ceiling, floor, wall, door, stair, and clutter. The clutter points are removed whereas the wall, door, and stair points are used for 2D floorplan generation. A region-growing segmentation algorithm paired with geometric reasoning rules is applied to group the points together into individual building elements. Finally, a 2-fold Random Sample Consensus (RANSAC) algorithm is applied to parameterize the building elements into 2D lines which are used to create the output floorplan. The proposed method is evaluated using the metrics of precision, recall, Intersection-over-Union (IOU), Betti error, and warping error.

• Journal of the Society of Disaster Information
• /
• v.19 no.4
• /
• pp.968-975
• /
• 2023

Purpose: We would like to confirm that the false positive rate of flames/smoke is high when detecting fires. Propose a method and dataset to recognize and classify fire situations to reduce the false detection rate. Method: Using the video as learning data, the characteristics of the fire situation were extracted and applied to the classification model. For evaluation, the model performance of Yolov8 and Slowfast were compared and analyzed using the fire dataset conducted by the National Information Society Agency (NIA). Result: YOLO's detection performance varies sensitively depending on the influence of the background, and it was unable to properly detect fires even when the fire scale was too large or too small. Since SlowFast learns the time axis of the video, we confirmed that detects fire excellently even in situations where the shape of an atypical object cannot be clearly inferred because the surrounding area is blurry or bright. Conclusion: It was confirmed that the fire detection rate was more appropriate when using a video-based artificial intelligence detection model rather than using image data.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.1
• /
• pp.77-84
• /
• 2024

In this paper, we propose a method for accurately and rapidly detecting defects in wire harnesses by utilizing computer vision to calculate six crucial measurement values: the length of crimped terminals, the dimensions (width) of terminal ends, and the width of crimped sections (wire and core portions). We employ Harris corner detection to locate object positions from two types of data. Additionally, we generate reference points for extracting measurement values by utilizing features specific to each measurement area and exploiting the contrast in shading between the background and objects, thus reflecting the slope of each sample. Subsequently, we introduce a method using the Euclidean distance and correction coefficients to predict values, allowing for the prediction of measurements regardless of changes in the wire's position. We achieve high accuracy for each measurement type, 99.1%, 98.7%, 92.6%, 92.5%, 99.9%, and 99.7%, achieving outstanding overall average accuracy of 97% across all measurements. This inspection method not only addresses the limitations of conventional visual inspections but also yields excellent results with a small amount of data. Moreover, relying solely on image processing, it is expected to be more cost-effective and applicable with less data compared to deep learning methods.

• Journal of Korean Society for Geospatial Information Science
• /
• v.21 no.2
• /
• pp.27-34
• /
• 2013

Due to the increased interest in global warming, interest in forest resources aimed towards reducing greenhouse gases have subsequently increased. Thus far, data related to forest resources have been obtained, through the employment of aerial photographs or satellite images, by means of plotting. However, the use of imaging data is disadvantageous; merely, due to the fact that recorded measurements such as the height of trees, in dense forest areas, lack accuracy. Within such context, the authors of this study have presented a method of data processing in which an individual tree is isolated within forested areas through the use of LIDAR data and ortho-images. Such isolation resulted in the provision of more efficient and accurate data in regards to the height of trees. As for the data processing of LIDAR, the authors have generated a normalized digital surface model to extract tree points via local maxima filtering, and have additionally, with motives to extract forest areas, applied object oriented image classifications to the processing of data using ortho-images. The final tree point was then given a figure derived from the combination of LIDAR and ortho-images results. Based from an experiment conducted in the Yongin area, the authors have analyzed the merits and demerits of methods that either employ LIDAR data or ortho-images and have thereby obtained information of individual trees within forested areas by combining the two data; thus verifying the efficiency of the above presented method.

• Korean Journal of Heritage: History & Science
• /
• v.50 no.1
• /
• pp.58-77
• /
• 2017

This study applied the index called Visual Perception Intensity (VPI) which quantitatively deals with landscape values and viewpoints to designate the cultural heritage areas in the Scenic Sites. The results of the study are as follows. First, a VPI selection index was presented for designating the cultural heritage areas in the Scenic Sites. The index was applied in consideration of the distance from the viewing point to the object and its incident angle. In addition, the process of the VPI analysis was implemented with GIS and the analysis algorithm was constructed. Second, the possibility of VPI was examined by comparing the simple frequency of the cumulative visibility with the results of the VPI. The VPI was analyzed to be more influenced by the incidence angle than the distance between the viewpoint and the object within a 4.74 km area. Third, a field survey was performed to investigate the effectiveness of the VPI classification. The survey was implemented based on the results of the investigation into the VPI to examine whether human visual perception was fully reflected. It was confirmed through the field survey that an area with high VPI was also an important area. Fourth, a plan for the cultural heritage area adjustment was constructed by applying the VPI to the areas already designated as Scenic Sites. As a result of classifying the VPI into three classes, it was found that the areas with the second class or higher were needed to be designated as cultural heritage areas and the areas with the third class as the Historical and Cultural Environments Preservation Area.