• Journal of Korea Multimedia Society
• /
• v.25 no.8
• /
• pp.1049-1058
• /
• 2022

HAR(Human Action Recognition) such as anomaly and object detection has become a trend in research field(s) that focus on utilizing Artificial Intelligence (AI) methods to analyze patterns of human action in crime-ridden area(s), media services, and industrial facilities. Especially, in real-time system(s) using video streaming data, HAR has become a more important AI-based research field in application development and many different research fields using HAR have currently been developed and improved. In this paper, we propose and analyze a deep-learning-based HAR that provides more efficient scheme(s) using an intelligent AI models, such system can be applied to media services using RGB video streaming data usage without feature extraction pre-processing. For the method, we adopt Slowfast based on the Deep Neural Network(DNN) model under an open dataset(HMDB-51 or UCF101) for improvement in prediction accuracy.

• Journal of KIBIM
• /
• v.14 no.3
• /
• pp.22-30
• /
• 2024

Recently, LLM (Large Language Model), a rapidly developing generative AI technology, is receiving much attention in the smart construction field. This study proposes a methodology for implementing an knowledge expert system by linking BIM (Building Information Modeling), which supports data hub functions in the smart construction domain with LLM. In order to effectively utilize LLM in a BIM expert system, excessive model learning costs, BIM big data processing, and hallucination problems must be solved. This study proposes an LLM-based BIM expert system architecture that considers these problems. This study focuses on the RAG (Retrieval-Augmented Generation) document generation method and search algorithm for effective BIM data retrieval, with the goal of implementing an LLM-based BIM expert system within a small GPU resource. For performance comparison and analysis, a prototype of the designed system is developed, and implications to be considered when developing an LLM-based BIM expert system are derived.

• Journal of Convergence for Information Technology
• /
• v.12 no.1
• /
• pp.31-38
• /
• 2022

This paper applies an expert independent unsupervised neural network learning-based multivariate time series data analysis model, MSCRED(Multi-Scale Convolutional Recurrent Encoder-Decoder), and to overcome the limitation, because the MCRED is based on Auto-encoder model, that train data must not to be contaminated, by using learning data sampling technique, called Subset Sampling Validation. By using the vibration data of power plant equipment that has been labeled, the classification performance of MSCRED is evaluated with the Anomaly Score in many cases, 1) the abnormal data is mixed with the training data 2) when the abnormal data is removed from the training data in case 1. Through this, this paper presents an expert-independent anomaly diagnosis framework that is strong against error data, and presents a concise and accurate solution in various fields of multivariate time series data.

• International journal of advanced smart convergence
• /
• v.12 no.3
• /
• pp.221-232
• /
• 2023

Recently, due to the expansion of the logistics industry, demand for logistics automation equipment is increasing. The modern logistics industry is a high-tech industry that combines various technologies. In general, as various technologies are grafted, the complexity of the system increases, and the occurrence rate of defects and failures also increases. As such, it is time for a predictive maintenance model specialized for logistics automation equipment. In this paper, in order to secure the operational reliability of the ICT/Cold-Chain Unmanned Storage, a predictive maintenance system was implemented based on the LSTM model. In this paper, a server for data management, such as collection and monitoring, and an analysis server that notifies the monitoring server through data-based failure and defect analysis are separately distinguished. The predictive maintenance platform presented in this paper works by collecting data and receiving data based on RabbitMQ, loading data in an InMemory method using Redis, and managing snapshot data DB in real time. The predictive maintenance platform can contribute to securing reliability by identifying potential failures and defects that may occur in the operation of the ICT/Cold-Chain Unmanned Storage in the future.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.6
• /
• pp.44-55
• /
• 2009

Diagnosis of diseases using gene expression data obtained from microarray chip is an active research area recently. It has been done by general machine learning algorithms, because it is difficult to analyze directly. However, recent research results about the analysis based on the interaction between genes is essential for the gene expression analysis, which means the analysis using the traditional machine learning algorithms has limitations. In this paper, we classify the gene expression data using the hyper-network model that considers the higher-order correlations between the features, and then compares the classification accuracies. And also, we present the new hypo-network model that improve the disadvantage of existing model, and compare the processing performances of the existing hypo-network model based on general sequential processor and the improved hypo-network model implemented on parallel processors. In the experimental results, we show that the performance of our model shows improved and competitive classification performance than traditional machine learning methods, as well as, the existing hypo-network model. We show that the performance is maximized when the hypernetwork model is implemented on our parallel processors.

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.36 no.3
• /
• pp.205-214
• /
• 2016

The objective of this study was to construct a forage rye (FR) dry matter yield (DMY) estimation model based on climate data by locations in South Korea. The data set (n = 549) during 29 years were used. Six optimal climatic variables were selected through stepwise multiple regression analysis with DMY as the response variable. Subsequently, via general linear model, the final model including the six climatic variables and cultivated locations as dummy variables was constructed as follows: DMY = 104.166SGD + 1.454AAT + 147.863MTJ + 59.183PAT150 - 4.693SRF + 45.106SRD - 5230.001 + Location, where SGD was spring growing days, AAT was autumnal accumulated temperature, MTJ was mean temperature in January, PAT150 was period to accumulated temperature 150, SRF was spring rainfall, and SRD was spring rainfall days. The model constructed in this research could explain 24.4 % of the variations in DMY of FR. The homoscedasticity and the assumption that the mean of the residuals were equal to zero was satisfied. The goodness-of-fit of the model was proper based on most scatters of the predicted DMY values fell within the 95% confidence interval.

• Journal of the Korean Geographical Society
• /
• v.43 no.1
• /
• pp.117-133
• /
• 2008

Although development density control has received murk attention in urban planning, there has been little research on empirical methods that can examine local changes in development density. Recently, attempts have been made to develop cellular automata (CA) models that can be applied to urban landuse change. This paper aims to develop an extended landuse change model based on urban CA considering development density. The proposed model not only includes density control component in a model framework, but also directly estimates local density changes in land use. The developed model was applied to the study area, which was a part of central Seoul. The calibration of the model was carried out over the period $1900s\sim2000s$ using parcel-based land use data and related variables The results of the calibrated model have been tested by comparison with actual landuse data, and have demonstrated that the developed model can produce realistic simulations of urban landuse changes. But model output is dependent on the spacio-temporal resolution of input data. Further research is necessary to improve the calibration procedure and methods for evaluating model validity.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.4
• /
• pp.222-227
• /
• 2022

In this paper, we implemented a Korean text generation and classification model based on a deep learning algorithm that can be applied to various industries. It consists of two implemented GAN-based Korean handwriting generation models and CNN-based Korean handwriting classification models. The GAN model consists of a generator model for generating fake Korean handwriting data and a discriminator model for discriminating fake handwritten data. In the case of the CNN model, the model was trained using the 'PHD08' dataset, and the learning result was 92.45. It was confirmed that Korean handwriting was classified with % accuracy. As a result of evaluating the performance of the classification model by integrating the Korean cursive data generated through the implemented GAN model and the training dataset of the existing CNN model, it was confirmed that the classification performance was 96.86%, which was superior to the existing classification performance.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.1
• /
• pp.47-55
• /
• 2024

We spend a lot of time in indoor space, and the space has a huge impact on our lives. Interior design plays a significant role to make an indoor space attractive and functional. However, it should consider a lot of complex elements such as color, pattern, and material etc. With the increasing demand for interior design, there is a growing need for technologies that analyze these design elements accurately and efficiently. To address this need, this study suggests a deep learning-based design analysis system. The proposed system consists of a semantic segmentation model that classifies spatial components and an image classification model that classifies attributes such as color, pattern, and material from the segmented components. Semantic segmentation model was trained using a dataset of 30000 personal indoor interior images collected for research, and during inference, the model separate the input image pixel into 34 categories. And experiments were conducted with various backbones in order to obtain the optimal performance of the deep learning model for the collected interior dataset. Finally, the model achieved good performance of 89.05% and 0.5768 in terms of accuracy and mean intersection over union (mIoU). In classification part convolutional neural network (CNN) model which has recorded high performance in other image recognition tasks was used. To improve the performance of the classification model we suggests an approach that how to handle data that has data imbalance and vulnerable to light intensity. Using our methods, we achieve satisfactory results in classifying interior design component attributes. In this paper, we propose indoor space design analysis system that automatically analyzes and classifies the attributes of indoor images using a deep learning-based model. This analysis system, used as a core module in the A.I interior recommendation service, can help users pursuing self-interior design to complete their designs more easily and efficiently.

• Journal of Biomedical Engineering Research
• /
• v.22 no.4
• /
• pp.369-375
• /
• 2001

The statistical regression model is one of the most frequently used clinical analysis methods. It has basic assumption of linearity, additivity and normal distribution of data. However, most of biological data in medical field are nonlinear and unevenly distributed. To overcome the discrepancy between the basic assumption of statistical model and actual biological data, we propose a new analytical method based on artificial neural network. The newly developed multilayer perceptron(MLP) is trained with 120 data set (60 normal, 60 patient). On applying test data, it shows the discrimination power of 0.76. The diabetic risk factors were also identified from the MLP neural network model and the logistic regression model. The signigicant risk factors identified by MLP model were post prandial glucose level(PP2), sex(male), fasting blood sugar(FBS) level, age, SBP, AC and WHR. Those from the regression model are sex(male), PP2, age and FBS. The combined risk factors can be identified using the MLP model. Those are total cholesterol and body weight, which is consistent with the result of other clinical studies. From this experiment we have learned that MLP can be applied to the combined risk factor analysis of biological data which can not be provided by the conventional statistical method.