• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.1243-1244
• /
• 2022

In recent years, the growing interest in off-site construction has led to factories scaling up their manufacturing and production processes in the construction sector. Consequently, continuous large-scale site monitoring in low-variability environments, such as prefabricated components production plants (precast concrete production), has gained increasing importance. Although many studies on computer vision-based site monitoring have been conducted, challenges for deploying this technology for large-scale field applications still remain. One of the issues is collecting and transmitting vast amounts of video data. Continuous site monitoring systems are based on real-time video data collection and analysis, which requires excessive computational resources and network traffic. In addition, it is difficult to integrate various object information with different sizes and scales into a single scene. Various sizes and types of objects (e.g., workers, heavy equipment, and materials) exist in a plant production environment, and these objects should be detected simultaneously for effective site monitoring. However, with the existing object detection algorithms, it is difficult to simultaneously detect objects with significant differences in size because collecting and training massive amounts of object image data with various scales is necessary. This study thus developed a large-scale site monitoring system using edge computing and a small-object detection system to solve these problems. Edge computing is a distributed information technology architecture wherein the image or video data is processed near the originating source, not on a centralized server or cloud. By inferring information from the AI computing module equipped with CCTVs and communicating only the processed information with the server, it is possible to reduce excessive network traffic. Small-object detection is an innovative method to detect different-sized objects by cropping the raw image and setting the appropriate number of rows and columns for image splitting based on the target object size. This enables the detection of small objects from cropped and magnified images. The detected small objects can then be expressed in the original image. In the inference process, this study used the YOLO-v5 algorithm, known for its fast processing speed and widely used for real-time object detection. This method could effectively detect large and even small objects that were difficult to detect with the existing object detection algorithms. When the large-scale site monitoring system was tested, it performed well in detecting small objects, such as workers in a large-scale view of construction sites, which were inaccurately detected by the existing algorithms. Our next goal is to incorporate various safety monitoring and risk analysis algorithms into this system, such as collision risk estimation, based on the time-to-collision concept, enabling the optimization of safety routes by accumulating workers' paths and inferring the risky areas based on workers' trajectory patterns. Through such developments, this continuous large-scale site monitoring system can guide a construction plant's safety management system more effectively.

• Smart Structures and Systems
• /
• v.32 no.4
• /
• pp.267-279
• /
• 2023

Data modelling and interpretation for structural health monitoring (SHM) field data are critical for evaluating structural performance and quantifying the vulnerability of infrastructure systems. In order to improve the data modelling accuracy, and extend the application range from data regression analysis to out-of-sample forecasting analysis, an improved most likely heteroscedastic Gaussian process (iMLHGP) methodology is proposed in this study by the incorporation of the outof-sample forecasting algorithm. The proposed iMLHGP method overcomes this limitation of constant variance of Gaussian process (GP), and can be used for estimating non-stationary typhoon-induced response statistics with high volatility. The first attempt at performing data regression and forecasting analysis on structural responses using the proposed iMLHGP method has been presented by applying it to real-world filed SHM data from an instrumented cable-stay bridge during typhoon events. Uncertainty quantification and correlation analysis were also carried out to investigate the influence of typhoons on bridge strain data. Results show that the iMLHGP method has high accuracy in both regression and out-of-sample forecasting. The iMLHGP framework takes both data heteroscedasticity and accurate analytical processing of noise variance (replace with a point estimation on the most likely value) into account to avoid the intensive computational effort. According to uncertainty quantification and correlation analysis results, the uncertainties of strain measurements are affected by both traffic and wind speed. The overall change of bridge strain is affected by temperature, and the local fluctuation is greatly affected by wind speed in typhoon conditions.

• Journal of Navigation and Port Research
• /
• v.47 no.6
• /
• pp.367-375
• /
• 2023

This study developed a new distance metric for vessel trajectories, applicable to marine traffic control services in the Korean coastal waters. The proposed metric is designed through the weighted summation of the traditional Hausdorff distance, which measures the similarity between spatiotemporal data and incorporates the differences in the average Speed Over Ground (SOG) and the variance in Course Over Ground (COG) between two trajectories. To validate the effectiveness of this new metric, a comparative analysis was conducted using the actual Automatic Identification System (AIS) trajectory data, in conjunction with an agglomerative clustering algorithm. Data visualizations were used to confirm that the results of trajectory clustering, with the new metric, reflect geographical distances and the distribution of vessel behavioral characteristics more accurately, than conventional metrics such as the Hausdorff distance and Dynamic Time Warping distance. Quantitatively, based on the Davies-Bouldin index, the clustering results were found to be superior or comparable and demonstrated exceptional efficiency in computational distance calculation.

• Advances in nano research
• /
• v.16 no.4
• /
• pp.325-340
• /
• 2024

There are several novel uses for dispersing many nanoparticles into a conventional fluid, including dynamic sealing, damping, heat dissipation, microfluidics, and more. Therefore, melting heat and mass transfer characteristics of a 3-D MHD Hybrid Nanofluid flow over a rotating disc with presenting dufour and soret effects are assessed numerically in this study. In this instance, we investigated both ferric sulfate and molybdenum disulfide as nanoparticles suspended within base fluid water. The governing partial differential equations are transformed into linked higher-order non-linear ordinary differential equations by the local similarity transformation. The collection of these deduced equations is then resolved using a Chebyshev spectral collocation-based algorithm built into the Mathematica software. To demonstrate how different instances of hybrid/ nanofluid are impacted by changes in temperature, velocity, and the distribution of nanoparticle concentration, examples of graphical and numerical data are given. For many values of the material parameters, the computational findings are shown. Simulations conducted for different physical parameters in the model show that adding hybrid nanoparticle to the fluid mixture increases heat transfer in comparison to simple nanofluids. It has been identified that hybrid nanoparticles, as opposed to single-type nanoparticles, need to be taken into consideration to create an effective thermal system. Furthermore, porosity lowers the velocities of simple and hybrid nanofluids in both cases. Additionally, results show that the drag force from skin friction causes the nanoparticle fluid to travel more slowly than the hybrid nanoparticle fluid. The findings also demonstrate that suction factors like magnetic and porosity parameters, as well as nanoparticles, raise the skin friction coefficient. Furthermore, It indicates that the outcomes from different flow scenarios correlate and are in strong agreement with the findings from the published literature. Bar chart depictions are altered by changes in flow rates. Moreover, the results confirm doctors' views to prescribe hybrid nanoparticle and particle nanoparticle contents for achalasia patients and also those who suffer from esophageal stricture and tumors. The results of this study can also be applied to the energy generated by the melting disc surface, which has a variety of industrial uses. These include, but are not limited to, the preparation of semiconductor materials, the solidification of magma, the melting of permafrost, and the refreezing of frozen land.

• Korean Journal of Remote Sensing
• /
• v.36 no.5_1
• /
• pp.763-774
• /
• 2020

Wind data forecasted from the numerical weather prediction (NWP) model is generally used as the first-guess of the target tracking process to obtain the atmospheric motion vectors(AMVs) because it increases tracking accuracy and reduce computational time. However, there is a contradiction that the NWP model used as the first-guess is used again as the reference in the AMVs verification process. To overcome this problem, model-independent first guesses are required. In this study, we propose the AMVs derivation from Lucas and Kanade optical flow method and then using it as the first guess. To retrieve AMVs, Himawari-8/AHI geostationary satellite level-1B data were used at 00, 06, 12, and 18 UTC from August 19 to September 5, 2015. To evaluate the impact of applying the optical flow method on the AMV derivation, cross-validation has been conducted in three ways as follows. (1) Without the first-guess, (2) NWP (KMA/UM) forecasted wind as the first-guess, and (3) Optical flow method based wind as the first-guess. As the results of verification using ECMWF ERA-Interim reanalysis data, the highest precision (RMSVD: 5.296-5.804 ms^-1) was obtained using optical flow based winds as the first-guess. In addition, the computation speed for AMVs derivation was the slowest without the first-guess test, but the other two had similar performance. Thus, applying the optical flow method in the target tracking process of AMVs algorithm, this study showed that the optical flow method is very effective as a first guess for model-independent AMVs derivation.

• Geophysics and Geophysical Exploration
• /
• v.5 no.4
• /
• pp.280-290
• /
• 2002

Three-dimensional (3-D) electromagnetic (EM) modeling algorithm has been developed using finite element method (FEM) to acquire more efficient interpretation techniques of EM data. When FEM based on nodal elements is applied to EM problem, spurious solutions, so called 'vector parasite', are occurred due to the discontinuity of normal electric fields and may lead the completely erroneous results. Among the methods curing the spurious problem, this study adopts vector element of which basis function has the amplitude and direction. To reduce computational cost and required core memory, complex bi-conjugate gradient (CBCG) method is applied to solving complex symmetric matrix of FEM and point Jacobi method is used to accelerate convergence rate. To verify the developed 3-D EM modeling algorithm, its electric and magnetic field for a layered-earth model are compared with those of layered-earth solution. As we expected, the vector based FEM developed in this study does not cause ny vector parasite problem, while conventional nodal based FEM causes lots of errors due to the discontinuity of field variables. For testing the applicability to high frequencies 100 MHz is used as an operating frequency for the layer structure. Modeled fields calculated from developed code are also well matched with the layered-earth ones for a model with dielectric anomaly as well as conductive anomaly. In a vertical electric dipole source case, however, the discontinuity of field variables causes the conventional nodal based FEM to include a lot of errors due to the vector parasite. Even for the case, the vector based FEM gave almost the same results as the layered-earth solution. The magnetic fields induced by a dielectric anomaly at high frequencies show unique behaviors different from those by a conductive anomaly. Since our 3-D EM modeling code can reflect the effect from a dielectric anomaly as well as a conductive anomaly, it may be a groundwork not only to apply high frequency EM method to the field survey but also to analyze the fold data obtained by high frequency EM method.

• Journal of Intelligence and Information Systems
• /
• v.25 no.4
• /
• pp.105-122
• /
• 2019

Dimensionality reduction is one of the methods to handle big data in text mining. For dimensionality reduction, we should consider the density of data, which has a significant influence on the performance of sentence classification. It requires lots of computations for data of higher dimensions. Eventually, it can cause lots of computational cost and overfitting in the model. Thus, the dimension reduction process is necessary to improve the performance of the model. Diverse methods have been proposed from only lessening the noise of data like misspelling or informal text to including semantic and syntactic information. On top of it, the expression and selection of the text features have impacts on the performance of the classifier for sentence classification, which is one of the fields of Natural Language Processing. The common goal of dimension reduction is to find latent space that is representative of raw data from observation space. Existing methods utilize various algorithms for dimensionality reduction, such as feature extraction and feature selection. In addition to these algorithms, word embeddings, learning low-dimensional vector space representations of words, that can capture semantic and syntactic information from data are also utilized. For improving performance, recent studies have suggested methods that the word dictionary is modified according to the positive and negative score of pre-defined words. The basic idea of this study is that similar words have similar vector representations. Once the feature selection algorithm selects the words that are not important, we thought the words that are similar to the selected words also have no impacts on sentence classification. This study proposes two ways to achieve more accurate classification that conduct selective word elimination under specific regulations and construct word embedding based on Word2Vec embedding. To select words having low importance from the text, we use information gain algorithm to measure the importance and cosine similarity to search for similar words. First, we eliminate words that have comparatively low information gain values from the raw text and form word embedding. Second, we select words additionally that are similar to the words that have a low level of information gain values and make word embedding. In the end, these filtered text and word embedding apply to the deep learning models; Convolutional Neural Network and Attention-Based Bidirectional LSTM. This study uses customer reviews on Kindle in Amazon.com, IMDB, and Yelp as datasets, and classify each data using the deep learning models. The reviews got more than five helpful votes, and the ratio of helpful votes was over 70% classified as helpful reviews. Also, Yelp only shows the number of helpful votes. We extracted 100,000 reviews which got more than five helpful votes using a random sampling method among 750,000 reviews. The minimal preprocessing was executed to each dataset, such as removing numbers and special characters from text data. To evaluate the proposed methods, we compared the performances of Word2Vec and GloVe word embeddings, which used all the words. We showed that one of the proposed methods is better than the embeddings with all the words. By removing unimportant words, we can get better performance. However, if we removed too many words, it showed that the performance was lowered. For future research, it is required to consider diverse ways of preprocessing and the in-depth analysis for the co-occurrence of words to measure similarity values among words. Also, we only applied the proposed method with Word2Vec. Other embedding methods such as GloVe, fastText, ELMo can be applied with the proposed methods, and it is possible to identify the possible combinations between word embedding methods and elimination methods.

• Journal of Internet Computing and Services
• /
• v.16 no.2
• /
• pp.49-55
• /
• 2015

According to the statistics of traffic accidents over recent 5 years, traffic accidents during the night times happened more than the day times. There are various causes to occur traffic accidents and the one of the major causes is inappropriate or missing street lights that make driver's sight confused and causes the traffic accidents. In this paper, with smartphones, we designed and implemented a lane luminance measurement application which stores the information of driver's location, driving, and lane luminance into database in real time to figure out the inappropriate street light facilities and the area that does not have any street lights. This application is implemented under Native C/C++ environment using android NDK and it improves the operation speed than code written in Java or other languages. To measure the luminance of road, the input image with RGB color space is converted to image with YCbCr color space and Y value returns the luminance of road. The application detects the road lane and calculates the road lane luminance into the database sever. Also this application receives the road video image using smart phone's camera and improves the computational cost by allocating the ROI(Region of interest) of input images. The ROI of image is converted to Grayscale image and then applied the canny edge detector to extract the outline of lanes. After that, we applied hough line transform method to achieve the candidated lane group. The both sides of lane is selected by lane detection algorithm that utilizes the gradient of candidated lanes. When the both lanes of road are detected, we set up a triangle area with a height 20 pixels down from intersection of lanes and the luminance of road is estimated from this triangle area. Y value is calculated from the extracted each R, G, B value of pixels in the triangle. The average Y value of pixels is ranged between from 0 to 100 value to inform a luminance of road and each pixel values are represented with color between black and green. We store car location using smartphone's GPS sensor into the database server after analyzing the road lane video image with luminance of road about 60 meters ahead by wireless communication every 10 minutes. We expect that those collected road luminance information can warn drivers about safe driving or effectively improve the renovation plans of road luminance management.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.12C
• /
• pp.1288-1298
• /
• 2002

To implement elliptic curve cryptosystem in GF(2$\^$m/) at high speed, a fast divider is required. Although bit-parallel architecture is well suited for high speed division operations, elliptic curve cryptosystem requires large m(at least 163) to support a sufficient security. In other words, since the bit-parallel architecture has an area complexity of 0(m$\^$m/), it is not suited for this application. In this paper, we propose a new serial-in serial-out systolic array for computing division operations in GF(2$\^$m/) using the standard basis representation. Based on a modified version of tile binary extended greatest common divisor algorithm, we obtain a new data dependence graph and design an efficient bit-serial systolic divider. The proposed divider has 0(m) time complexity and 0(m) area complexity. If input data come in continuously, the proposed divider can produce division results at a rate of one per m clock cycles, after an initial delay of 5m-2 cycles. Analysis shows that the proposed divider provides a significant reduction in both chip area and computational delay time compared to previously proposed systolic dividers with the same I/O format. Since the proposed divider can perform division operations at high speed with the reduced chip area, it is well suited for division circuit of elliptic curve cryptosystem. Furthermore, since the proposed architecture does not restrict the choice of irreducible polynomial, and has a unidirectional data flow and regularity, it provides a high flexibility and scalability with respect to the field size m.

• Korean Chemical Engineering Research
• /
• v.61 no.4
• /
• pp.523-541
• /
• 2023

As accessibility to 3D printers increases, there is a growing frequency of exposure to chemicals associated with 3D printing. However, research on the toxicity and harmfulness of chemicals generated by 3D printing is insufficient, and the performance of toxicity prediction using in silico techniques is limited due to missing molecular structure data. In this study, quantitative structure-activity relationship (QSAR) model based on data-centric AI approach was developed to predict the toxicity of new 3D printing materials by imputing missing values in molecular descriptors. First, MissForest algorithm was utilized to impute missing values in molecular descriptors of hazardous 3D printing materials. Then, based on four different machine learning models (decision tree, random forest, XGBoost, SVM), a machine learning (ML)-based QSAR model was developed to predict the bioconcentration factor (Log BCF), octanol-air partition coefficient (Log Koa), and partition coefficient (Log P). Furthermore, the reliability of the data-centric QSAR model was validated through the Tree-SHAP (SHapley Additive exPlanations) method, which is one of explainable artificial intelligence (XAI) techniques. The proposed imputation method based on the MissForest enlarged approximately 2.5 times more molecular structure data compared to the existing data. Based on the imputed dataset of molecular descriptor, the developed data-centric QSAR model achieved approximately 73%, 76% and 92% of prediction performance for Log BCF, Log Koa, and Log P, respectively. Lastly, Tree-SHAP analysis demonstrated that the data-centric-based QSAR model achieved high prediction performance for toxicity information by identifying key molecular descriptors highly correlated with toxicity indices. Therefore, the proposed QSAR model based on the data-centric XAI approach can be extended to predict the toxicity of potential pollutants in emerging printing chemicals, chemical process, semiconductor or display process.