• International journal of advanced smart convergence
• /
• v.12 no.4
• /
• pp.88-97
• /
• 2023

Traffic flow prediction is of great significance in urban planning and traffic management. As the complexity of urban traffic increases, existing prediction methods still face challenges, especially for the fusion of spatiotemporal information and the capture of long-term dependencies. This study aims to use the fusion model of graph neural network to solve the spatio-temporal information fusion problem in traffic flow prediction. We propose a new deep learning model Spatio-Temporal Information Fusion using Graph Neural Networks (STFGNN). We use GCN module, TCN module and LSTM module alternately to carry out spatiotemporal information fusion. GCN and multi-core TCN capture the temporal and spatial dependencies of traffic flow respectively, and LSTM connects multiple fusion modules to carry out spatiotemporal information fusion. In the experimental evaluation of real traffic flow data, STFGNN showed better performance than other models.

• Clinical Endoscopy
• /
• v.55 no.5
• /
• pp.594-604
• /
• 2022

Over the past decade, technological advances in deep learning have led to the introduction of artificial intelligence (AI) in medical imaging. The most commonly used structure in image recognition is the convolutional neural network, which mimics the action of the human visual cortex. The applications of AI in gastrointestinal endoscopy are diverse. Computer-aided diagnosis has achieved remarkable outcomes with recent improvements in machine-learning techniques and advances in computer performance. Despite some hurdles, the implementation of AI-assisted clinical practice is expected to aid endoscopists in real-time decision-making. In this summary, we reviewed state-of-the-art AI in the field of gastrointestinal endoscopy and offered a practical guide for building a learning image dataset for algorithm development.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.8
• /
• pp.2067-2081
• /
• 2024

Recently, image analysis research has been actively conducted due to the accumulation of big image data and the development of deep learning. Image analytics research has different characteristics from other data such as data size, real-time, image quality diversity, structural complexity, and security issues. In addition, a large amount of data is required to effectively analyze images with deep-learning models. However, in many fields, the data that can be collected is limited, so there is a need for meta learning based image analysis technology that can effectively train models with a small amount of data. This paper presents a comprehensive survey of meta-learning-based object-tracking techniques. This approach comprehensively explores object tracking methods and research that can achieve high performance in data-limited situations, including key challenges and future directions. It provides useful information for researchers in the field and can provide insights into future research directions.

• Journal of the Ergonomics Society of Korea
• /
• v.35 no.4
• /
• pp.193-204
• /
• 2016

Objective: This study carried out base research to build an agricultural safety monitoring system through ICT convergence to reduce safety accidents and enhance welfare in life in the agricultural field. Background: The functions and values of rural villages as the space of living are recognized anew, but occupational accident rate due to farm work accidents is on the rise each year. Therefore, the seriousness of such a problem emerges. The convergence technology combining ICT is recently applied to industries overall, and therefore better services are offered. However, studies on ICT convergence has not yet been applied to the agricultural safety field. Method: This study identified ICT convergence service technology trends and representative serious accident types mainly occurring in agricultural activities. This study defined the major factors of farm work accidents and ICT to solve those accident factors including the sensor technology, wired/wireless communication technology and location information service, and applied them to prototype PCB for the development of an agricultural safety monitoring system. Results: This study proposed an emergency monitoring system for farmers and a harmful environment monitoring system. The ICT technology to prevent farm work accidents can be summarized as sensing technology, ICT and network technology and user interface technology. This study also designed PCB module configuration and situation judgment algorithm as basic research for proposed monitoring system development. Conclusion: The ICT-based agricultural safety monitoring research proposed in this study is expected to become the basis to build a future real time monitoring system, and also is expected to contribute to social safety and welfare service improvement for farmers. Application: The ICT convergence farmer accident prevention system will make contributions to the prevention of serious farm work accidents.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.15 no.3
• /
• pp.52-59
• /
• 2016

COSMOS system which was developed in Seoul for real-time signal control was designed to judge traffic condition for practicing signal operation. However, it occurs efficiency problem that stop line detection and queue length detection could not judge overflow saturation of street. For that reason, following research process GPS data of Seoul city's corporationowned taxi to calculate travel speed that excluded existing system of stop line detection and queue length detection. Also, "Research of calculating queue length by GPS data" which was progressed with following research expressed queue length. It is based on establishing algorithm of judging congestion situation. The algorithm was applied to a few areas where appeared congestion situation consistently to confirm real time traffic condition with established network. [Entrance of the National Sport Institute ${\rightarrow}$ Gangnam station Intersection, Yuksam station intersection ${\rightarrow}$ National Sport Institute.

• IMMUNE NETWORK
• /
• v.9 no.1
• /
• pp.27-33
• /
• 2009

Macrophages generated in vitro using macrophage-colony stimulating factor (M-CSF) and interleukin (IL)-6 from bone marrow cells (BM-Mp) are defective in antigen presenting cell (APC) function as shown by their ability to induce the proliferation of anti-CD3 mAb-primed syngeneic T cells. However, they do express major histocompatibility (MHC) class I and II molecules. accessory molecules and intracellular adhesion molecules. Here we demonstrate that the defective APC function of macrophages is mainly due to production of $TGF-{\beta}1$ by BM-Mp. Methods: Microarray analysis showed that $TGF-{\beta}1$ was highly expressed in BM-Mp, compared to a macrophage cell line, B6D. which exerted efficient APC function. Production of $TGF-{\beta}1$ by BM-Mp was confirmed by neutralization experiments of $TGF-{\beta}1$ as well as by real time-polymerase chain reaction (PCR). Results: Addition of $anti-TGF-{\beta}1$ monoclonal antibody to cultures of BM-Mp and anti-CD3 mAb-primed syngeneic T cells efficiently induced the proliferation of syngeneic T cells. Conversely, the APC function of B6D cells was almost completely suppressed by addition of $TGF-{\beta}1$. Quantitative real time-PCR analysis also confirmed the enhanced expression of $TGF-{\beta}1$ in BM-Mp. Conclusion: The defective APC function of macrophages generated in vitro with M-CSF and IL-6 was mainly due to the production of $TGF-{\beta}1$ by macrophages.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.4
• /
• pp.403-411
• /
• 2014

In this paper, an algorithm of extracting the features of home appliances and automatically building a database to identify home appliances is designed and presented. For the verification, a software library supporting this algorithm is implemented and added to an power management system server, which was already implemented to support real-time monitoring of home appliances' power consumption status and controlling their power. The implemented system consists of a system server and clients, each of which measures the power consumed by a home appliance plugged in it and transmits the information to the server in real-time over a wireless network. Through experiments, it is verified that it is possible to identify any home appliance connected to a specific client.

• Journal of Digital Contents Society
• /
• v.7 no.4
• /
• pp.235-243
• /
• 2006

This paper presents the development of certain efficient method for expressing the emotion of an avatar based on the facial expression recognition. This new method is not changing a facial expression of the avatar manually. It can be changing a real time facial expression of the avatar based on recognition of a facial pattern which can be captured by a web cam. It provides a tool for recognizing some part of images captured by the web cam. Because of using the model-based approach, this tool recognizes the images faster than other approaches such as the template-based or the network-based. It is extracting the shape of user's lip after detecting the information of eyes by using the model-based approach. By using changes of lip's patterns, we define 6 patterns of avatar's facial expression by using 13 standard lip's patterns. Avatar changes a facial expression fast by using the pre-defined avatar with corresponding expression.

• Journal of Korea Multimedia Society
• /
• v.13 no.4
• /
• pp.504-511
• /
• 2010

In this paper, we propose the face region recognition system for City-Security CCTV(Closed Circuit Television) using hippocampal neural network which is modelling of human brain's hippocampus. This system is composed of feature extraction, learning and recognition part. The feature extraction part is constructed using PCA(Principal Component Analysis) and LDA(Linear Discriminants Analysis). In the learning part, it can label the features of the image-data which are inputted according to the order of hippocampal neuron structure to reaction-pattern according to the adjustment of a good impression in a dentate gyrus and remove the noise through the auto-associative memory in the CA3 region. In the CA1 region receiving the information of the CA3, it can make long-term memory learned by neuron. Experiments confirm the each recognition rate, that are shape change and light change. The experimental results show that we can compare a feature extraction and learning method proposed in this paper of any other methods, and we can confirm that the proposed method is superior to existing methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.12
• /
• pp.2349-2354
• /
• 2006

Due to technological developments and the joint effect of both new social and economic needs and constraints, telemedicine is expanding rapidly through a variety of applications. Especially, owing to the rapid aging of society and increasing the wish for well being life, we take interest in health care services for people with special needs who wish to remain independent and living in their own home. We have focused on tole-monitoring to real-time medical signal and environment factor which is an influence on medical signal. We monitor the six signal(medical signal and environment factor), and transmit that signal to computer on bluetooth network. We get the information after using the some digital signal processing system, and display that information on the real-time monitoring system. We developed the measurer as portable type in older to non-restrained monitor.