• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.35 no.5
• /
• pp.423-430
• /
• 2017

Recently, data analysis research has been carried out using the deep learning technique in various fields such as image interpretation and/or classification. Various types of algorithms are being developed for many applications. In this paper, we propose a precipitation prediction algorithm based on deep learning with high accuracy in order to take care of the possible severe damage caused by climate change. Since the geographical and seasonal characteristics of Korea are clearly distinct, the meteorological factors have repetitive patterns in a time series. Since the LSTM (Long Short-Term Memory) is a powerful algorithm for consecutive data, it was used to predict precipitation in this study. For the numerical test, we calculated the PWV (Precipitable Water Vapor) based on the tropospheric delay of the GNSS (Global Navigation Satellite System) signals, and then applied the deep learning technique to the precipitation prediction. The GNSS data was processed by scientific software with the troposphere model of Saastamoinen and the Niell mapping function. The RMSE (Root Mean Squared Error) of the precipitation prediction based on LSTM performs better than that of ANN (Artificial Neural Network). By adding GNSS-based PWV as a feature, the over-fitting that is a latent problem of deep learning was prevented considerably as discussed in this study.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.5
• /
• pp.241-246
• /
• 2015

Although many different methods have been used to solve stereo correspondent problems, the deviation of accuracy is too big. Among those many methods, the one that uses segmentation information of input image has received high attention in academic field since it is very close to vision recognition. In this thesis, the existing method of acquiring a single value by using the segment information and initial disparity value was viewed in NP-hard problem to propose a new method. In order to verify the validity of the proposed method, well-known data were used for experiment and the resulted data was analyzed. Although there were some disadvantages in the time aspect, it showed somewhat useful results in the accuracy aspect.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2000.11a
• /
• pp.272-276
• /
• 2000

Reverse engineering refers to the process that creates a physical part from acquiring the surface data of an existing part using a scanning device. In recent years, as the non-contact type scanning devices become more popular, the huge amount of point data can be obtained with high speed. The point data handling process, therefore, becomes more important since the scan data need to be refined for the efficiency of subsequent tasks such as mesh generation and surface fitting. As one of point handling functions, the cross-sectioning function is still frequently used for extracting the necessary data from the point cloud. The commercial reverse engineering software supports cross-sectioning functions, however, these are only for cross-sectioning the point cloud with the constant spacing and direction. In this paper, adaptive cross-sectioning point cloud which allow the changes of the spacing and directions of cross-sections according to the constant spacing and direction. In this paper, adaptive cross-sectioning algorithms which allow the changes of the spacing and directions of cross-sections according to the curvature difference of the point cloud data are proposed.

• Geomechanics and Engineering
• /
• v.15 no.1
• /
• pp.677-686
• /
• 2018

Aimed at serious casing damage problem during the process of oilfield development by injecting water, based on seepage mechanics, fluid mechanics and the theory of rock mechanics, the multi-physics coupling theory was also taken into account, the mathematical model for production of petroleum with water flooding was established, and the method to solve the coupling model was presented by combination of Abaqus and Eclipse software. The Q9G3 block in Jibei oilfield was taken for instance, the well log data and geological survey data were employed to build the numerical model of Q9G3 block, the method established above was applied to simulate the evolution of seepage and stress. The production data was imported into the model to conduct the history match work of the model, and the fitting accuracy of the model was quite good. The main mechanism of casing damage of the block was analyzed, and some wells with probable casing damage problem were pointed out, the displacement of the well wall matched very well with testing data of the filed. Finally, according to the simulation results, some useful measures for preventing casing damage in Jibei oilfield was proposed.

• Journal of Korea Multimedia Society
• /
• v.23 no.2
• /
• pp.186-203
• /
• 2020

Image-based Virtual Try-On Systems are among the most potential solution for virtual fitting which tries on a target clothes into a model person image and thus have attracted considerable research efforts. In many cases, current solutions for those fails in achieving naturally looking virtual fitted image where a target clothes is transferred into the body area of a model person of any shape and pose while keeping clothes context like texture, text, logo without distortion and artifacts. In this paper, we propose a new improved image-based virtual try-on network system based on keypoints, which we name as KP-VTON. The proposed KP-VTON first detects keypoints in the target clothes and reliably predicts keypoints in the clothes of a model person image by utilizing a dense human pose estimation. Then, through TPS transformation calculated by utilizing the keypoints as control points, the warped target clothes image, which is matched into the body area for wearing the target clothes, is obtained. Finally, a new try-on module adopting Attention U-Net is applied to handle more detailed synthesis of virtual fitted image. Extensive experiments on a well-known dataset show that the proposed KP-VTON performs better the state-of-the-art virtual try-on systems.

• Journal of ILASS-Korea
• /
• v.14 no.2
• /
• pp.77-83
• /
• 2009

In this study, the effect of important parameters such as the pixel size and number of a CCD, the object distance, the wavelength of laser, and the particle diameter on the depth of focus in digital in-line particle holography were investigated. The depth of focus in several different cases was calculated using simulation holograms and detailed description of the depth of focus in digital particle holography was presented. The depth of focus is directly proportional to the object distance and the particle size. With the increase of the wavelength of laser, the depth of focus is decreased. The depth of focus is also inversely proportional to the pixel size and number of a CCD. Using the data of depth of focus from simulation holograms and a data-fitting software, we obtained the prediction equations of depth of focus for typical CCD cameras. Finally, the prediction equations of depth of focus in digital particle holography were verified by investigating real holograms of the calibration target in different cases and satisfied agreement between measured values and predicted values was confirmed.

• Journal of Construction Engineering and Project Management
• /
• v.5 no.1
• /
• pp.26-31
• /
• 2015

Success of the construction companies is based on the successful completion of projects within the agreed cost and time limits. Artificial neural networks (ANN) have recently attracted much attention because of their ability to solve the qualitative and quantitative problems faced in the construction industry. For the estimation of cost and duration different ANN models were developed. The database consists of data collected from completed projects. The same data is normalised and used as inputs and targets for developing ANN models. The models are trained, tested and validated using MATLAB R2013a Software. The results obtained are the ANN predicted outputs which are compared with the actual data, from which deviation is calculated. For this purpose, two successfully completed highway road projects are considered. The Nftool (Neural network fitting tool) and Nntool (Neural network/ Data Manager) approaches are used in this study. Using Nftool with trainlm as training function and Nntool with trainbr as the training function, both the Projects A and B have been carried out. Statistical analysis is carried out for the developed models. The application of neural networks when forming a preliminary estimate, would reduce the time and cost of data processing. It helps the contractor to take the decision much easier.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.7 no.4
• /
• pp.81-88
• /
• 2011

Mini Double Gears Frame is critical part of PDP and also produces couple hundred thousand every month. In the process of mass production, product inspection is very important process. Double Gear, one of the part of machine, was inspected by human eyes which caused mistakes and slow progress. To achieve the speed and accuracy the system was compensated by vision system which is inspecting automatically. The focus value is measured based on the fact that high contrast images have much high frequency edge information. High frequency term of the image is extracted using the high-pass filter and the sum of the high frequency term is used as the focus value. We used a Gaussian smoothing filter to reduce the noise and then measures the focus value using the modified Laplacian filter called a Sum modified Laplacian Focus values for the various lens positions are calculated and the position with the maximum focus value is decided as the focused position. The focus values calculated in various lens position showed the Gaussian distribution. We proposed a method to estimate the best focus position using the Gaussian curve fitting. Focus values of the uniform interval lens positions are calculated and the values are used to estimate the Gaussian distribution parameters to find the best focus position.

• Bulletin of the Korean Chemical Society
• /
• v.31 no.10
• /
• pp.2903-2908
• /
• 2010

Protein S-nitrosation is common in cells under nitrosative stress. In order to model proteins with S-nitrosocysteine (CysSNO) residues, we first developed an Amber force field for S-nitrosoethanethiol (EtSNO) and then transferred it to CysSNO. Partial atomic charges for EtSNO and CysSNO were obtained by a restrained electrostatic potential approach to be compatible with the Amber-99 force field. The force field parameters for bonds and angles in EtSNO were obtained from a generalized Amber force field (GAFF) by running the Antechamber module of the Amber software package. The GAFF parameters for the CC-SN and CS-NO dihedrals were not accurate and thus determined anew. The CC-SN and CS-NO torsional energy profiles of EtSNO were calculated quantum mechanically at the level of B3LYP/cc-pVTZ//HF/6-$31G^*$. Torsional force constants were obtained by fitting the theoretical torsional energies with those obtained from molecular mechanics energy minimization. These parameters for EtSNO reproduced, to a reasonable accuracy, the corresponding torsional energy profiles of the capped tripeptide ACE-CysSNO-NME as well as their structures obtained from quantum mechanical geometry optimization. A molecular dynamics simulation of myoglobin with a CysSNO residue produced a well-behaved trajectory demonstrating that the parameters may be used in modeling other S-nitrosated proteins.

• The Bulletin of The Korean Astronomical Society
• /
• v.39 no.1
• /
• pp.55-55
• /
• 2014

국립 고흥 청소년 우주 체험 센터는 연구와 학생들의 교육을 위해 12GHz 전파 간섭계를 개발했다. 저비용으로 제작하기 위하여 상용 제품들을 주로 사용하였고, 해변 가에 위치한 센터 특성상 강한 바다 바람과 부식에 견디도록 제작하였다. 고흥 간섭계는 직경 1.8m의 off-axis parabola 안테나 3대로 이루어져 있으며, 각 안테나 사이의 기선길이는 4, 19, 20m로, 해상도가 최대 약 4'인 영상을 얻을 수 있다. 수신기는 중심 주파수가 12.177GHz, 대역폭이 10MHz이며 메탄올 천이선과 연속파를 관측할 수 있는 시스템이다. 시스템온도는 100-200K로 추정된다. 각 수신기에서 나오는 신호는 digitizer로 읽어 들이며, 병렬 처리 프로그램으로 software correlation을 수행한다. 태양, 달, Crab Nebula, 그리고 Cassiopeia A 등을 관측하여 프린지를 검출하는데 성공하였다. 가시함수를 구하기 위한 프린지 fitting model의 파라미터들은 기선벡터의 측량과 점전파원 관측을 통하여 정밀하게 측정하였다. 태양에 대한 영상관측결과를 논의하고자 한다.