• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.482-484
• /
• 2019

Learning and analyzing 3D point clouds with deep networks is challenging due to the limited and irregularity of the data. In this paper, we present a data-driven point cloud augmentation technique. The key idea is to learn multilevel features per point and to reconstruct to a similar point set. Our network is applied to a projection loss function that encourages the predicted points to remain on the geometric shapes with a particular target. We conduct various experiments using ShapeNet part data to evaluate our method and demonstrate its possibility. Results show that our generated points have a similar shape and are located closer to the object.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.36C no.6
• /
• pp.44-51
• /
• 1999

The HMM-Net is a neural network architecture that implements the computation of output probabilities of a hidden Markov model (HMM). The architecture is developed for the purpose of combining the discriminant power of neural networks with the time-domain modeling capability of HMMs. Criteria of maximum likehood(ML) and minimization of mean squared error(MMSE) are used for learning HMM-Net classifiers. The criterion MMSE is better than ML when initial learning condition is well established. However Ml is more useful one when the condition is incomplete[3]. Therefore we propose an efficient learning method of HMM-Net classifiers using a hybrid criterion(ML/MMSE). In the method, we begin a learning with ML in order to get a stable start-point. After then, we continue the learning with MMSE to search an optimal or near-optimal solution. Experimental results for the isolated numeric digits from /0/ to /9/, a training and testing time-series pattern set, show that the performance of the proposed method is better than the others in the respects of learning and recognition rates.

• Proceedings of the KIEE Conference
• /
• 1995.07c
• /
• pp.1350-1352
• /
• 1995

This paper discribed the characteristic of I-t cross-over-point between current limited-fuse and explusion fuse(Bay-O-Net Fuse) and fuse protection in pad-mounted transformer that was generated internal faults and the short circuit of secondary side(load side). In the I-t cross-over-point, current limited fuse was melted when transient recovery voltage was raised rapidly.

• Journal of The Korean Association of Information Education
• /
• v.5 no.2
• /
• pp.240-257
• /
• 2001

Although many attempts have been made to use the Internet for educational purposes, not many attempts have achieved their goals. Such failure is mainly due to the lack of understanding on the way to use the Internet. The goal of this research is to investigate the potentiality of the Project-Based Learning approach using the Internet(NetPBL) and the ways to utilize the NetPBL. The NetPBL can be utilized through various activities such as keypals, mentoring, use of resources, cooperative learning, publishing, survey and data analysis, cooperative problem solving, simulation, and social action. Such diversity of the NetPBL can create a problem-based, context-based and learner-centered environment, which takes various types of the Internet use. In spite of such potentiality, little is known on how to implement the NetPBL. On this point, this research attempts to synthesize instructional strategies to implement the NetPBL at the macro and the micro level. At the macro level, instructional process is divided into four steps such as plan, preparation, implementation and closure, and some instructional suggestions are made for each step. At the micro level, detailed instructional strategies are suggested for the facilitation of self-directed learning and cooperative learning.

• Journal of Korea Water Resources Association
• /
• v.54 no.5
• /
• pp.301-309
• /
• 2021

In this study, we tried to improve the performance of the existing U-net-based deep learning rainfall prediction model, which can weaken the meaning of time series order. For this, ConvLSTM2D U-Net structure model considering temporal consistency of data was applied, and we evaluated accuracy of the ConvLSTM2D U-Net model using a RainNet model and an extrapolation-based advection model. In addition, we tried to improve the uncertainty in the model training process by performing learning not only with a single model but also with 10 ensemble models. The trained neural network rainfall prediction model was optimized to generate 10-minute advance prediction data using four consecutive data of the past 30 minutes from the present. The results of deep learning rainfall prediction models are difficult to identify schematically distinct differences, but with ConvLSTM2D U-Net, the magnitude of the prediction error is the smallest and the location of rainfall is relatively accurate. In particular, the ensemble ConvLSTM2D U-Net showed high CSI, low MAE, and a narrow error range, and predicted rainfall more accurately and stable prediction performance than other models. However, the prediction performance for a specific point was very low compared to the prediction performance for the entire area, and the deep learning rainfall prediction model also had limitations. Through this study, it was confirmed that the ConvLSTM2D U-Net neural network structure to account for the change of time could increase the prediction accuracy, but there is still a limitation of the convolution deep neural network model due to spatial smoothing in the strong rainfall region or detailed rainfall prediction.

• Structural Engineering and Mechanics
• /
• v.45 no.3
• /
• pp.411-422
• /
• 2013

A well-designed mesh shape of the cable net is of essential significance to achieve high performance of cable-network antenna reflectors. This paper is concerned with the mesh design problem for such antenna reflector structure. Two new methods for creating the topological forms of the cable net are first presented. Among those, the cyclosymmetry method is useful to generate different polygon-faceted meshes, while the topological mapping method is suitable for acquiring triangle-faceted meshes with different mesh grid densities. Then, the desired spatial paraboloidal mesh geometrical configuration in the state of static equilibrium is formed by applying a simple mesh generation approach based on the force density method. The main contribution of this study is that a general technical guide for how to create the connectivities between the nodes and members in the cable net is provided from the topological point of view. With the new idea presented in this paper, multitudes of mesh configurations with different net patterns can be sought by a certain rule rather than by empiricism, which consequently gives a valuable technical reference for the mesh design of this type of cable-network structures in the engineering.

• Journal of Advanced Navigation Technology
• /
• v.16 no.1
• /
• pp.109-115
• /
• 2012

Humanities is very difficult area to use the computer. However, recently, convergence trend has increase widely at all of the academic area. Therefore, in this paper, we propose a technical method to use an IT for the popularity of humanities. For this purpose, we implement a path process that use Petri-net to apply digital storytelling to humanities. We also make a structure to connect an examples and questions from sentences or articles as digital storytelling. The digital storytelling consists of six factors; author, synopsis, background, construction, view point, and user's or reader's review. Proposed method provides a process to analyze the data path of a literary work using Petri-net.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.4
• /
• pp.147-153
• /
• 2018

In this paper, we propose an efficient WBC 14-Diff classification which performs using the WBC-ResNet-152, a type of CNN model. The main point of view is to use Super-pixel for the segmentation of the image of WBC, and to use ResNet for the classification of WBC. A total of 136,164 blood image samples (224x224) were grouped for image segmentation, training, training verification, and final test performance analysis. Image segmentation using super-pixels have different number of images for each classes, so weighted average was applied and therefore image segmentation error was low at 7.23%. Using the training data-set for training 50 times, and using soft-max classifier, TPR average of 80.3% for the training set of 8,827 images was achieved. Based on this, using verification data-set of 21,437 images, 14-Diff classification TPR average of normal WBCs were at 93.4% and TPR average of abnormal WBCs were at 83.3%. The result and methodology of this research demonstrates the usefulness of artificial intelligence technology in the blood cell image classification field. WBC-ResNet-152 based morphology approach is shown to be meaningful and worthwhile method. And based on stored medical data, in-depth diagnosis and early detection of curable diseases is expected to improve the quality of treatment.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.3
• /
• pp.133-139
• /
• 2021

Currently, the process of constructing a high-definition road map has a high proportion of manual labor, so there are limitations in construction time and cost. Research to automate map production with high-definition road maps using artificial intelligence is being actively conducted, but since the construction of training data for the map construction is also done manually, there is a need to automatically build training data. Therefore, in this study, after converting to images using point clouds acquired by a mobile mapping system, the road marking areas were extracted through image reclassification and overlap analysis using thresholds. Then, a methodology was proposed to automatically construct training data for deep learning data for the high-definition road map through the classification of the polygon types in the extracted regions. As a result of training 2,764 lane data constructed through the proposed methodology on a deep learning-based PointNet model, the training accuracy was 99.977%, and as a result of predicting the lanes of three color types using the trained model, the accuracy was 99.566%. Therefore, it was found that the methodology proposed in this study can efficiently produce training data for high-definition road maps, and it is believed that the map production process of road markings can also be automated.

• Journal of the Ergonomics Society of Korea
• /
• v.6 no.1
• /
• pp.33-40
• /
• 1987

To study the body trunk basic to Clothing construction, and study the peculiarities of visual spatial division, necessary items are measured indirectly from 216 unmarried women from 19 to 24 years old by a photographic net-work method. In so doing, the problem of Fashion Design in establishing the location of Darts for Basic Dress is not considered. The following results are obtained. 1) Indirect measuring method, is obtained approximate to actual size, with an error of .+-. 2.8cm. 2) In the modeling plan of Dart location viewed from the visual-spatial division in Basic Dress, it is concluded that Darts are to be placed at the point of 1/3k+1/5k form the waist. From the aesthetic point of wiew, it is more appealling for darts to be placed at the point of 6cm .+-. 0.6cm right or left of center. 3) From direct measurement dart location can be set based on bust point width, and from indirect measurement, dart location can be set based on waist width.