• Journal of Korea Multimedia Society
• /
• v.24 no.12
• /
• pp.1598-1605
• /
• 2021

Recently, convolutional neural networks (CNNs) have been widely used with excellent performance in various computer vision fields, including super-resolution (SR). However, CNN is computationally intensive and requires a lot of memory, making it difficult to apply to limited hardware resources such as mobile or Internet of Things devices. To solve these limitations, network lightening studies have been actively conducted to reduce the depth or size of pre-trained deep CNN models while maintaining their performance as much as possible. This paper aims to lighten the SR CNN model, SRGAN, using the knowledge distillation among network lightening technologies; thus, it proposes four techniques with different methods of transferring the knowledge of the teacher network to the student network and presents experiments to compare and analyze the performance of each technique. In our experimental results, it was confirmed through quantitative and qualitative evaluation indicators that student networks with knowledge transfer performed better than those without knowledge transfer, and among the four knowledge transfer techniques, the technique of conducting adversarial learning after transferring knowledge from the teacher generator to the student generator showed the best performance.

• 한국정보통신설비학회:학술대회논문집
• /
• 2009.08a
• /
• pp.25-27
• /
• 2009

In this paper, a sensor network system for providing intelligent home network services is suggested. It steadily collects biological data of resident people and automatically detects emergency situations LEID(Lighting Embedded Information Device) system are the most essential component of the sensor network. They embed sensor network technology into lightening devices which are indispensable most living spaces. To verify practicality of the proposed intelligent home network service system, a prototypical system is realized in the Smart Home Industrialization Support Center at Kookmin University, and is tested within many practical circumstances.

• The Journal of Korea Robotics Society
• /
• v.18 no.1
• /
• pp.88-92
• /
• 2023

Lithium-ion batteries are actively used in various industrial sites such as field robots, drones, and electric vehicles due to their high energy efficiency, light weight, long life span, and low self-discharge rate. When using a lithium-ion battery in a field, it is important to accurately estimate the SoC (State of Charge) of batteries to prevent damage. In recent years, SoC estimation using data-based artificial neural networks has been in the spotlight, but it has been difficult to deploy in the embedded board environment at the actual site because the computation is heavy and complex. To solve this problem, neural network lightening technologies such as network pruning have recently attracted attention. When pruning a neural network, the performance varies depending on which layer and how much pruning is performed. In this paper, we introduce an optimized pruning technique by improving the existing pruning method, and perform a comparative experiment to analyze the results.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.2B
• /
• pp.225-231
• /
• 2010

In this paper, a sensor network system for providing u-health services is suggested and is implemented. It steadily collects biological data of resident people and automatically detects emergency situations. LEID (Lighting Embedded Information Device) nodes are the most essential component of the sensor network. They embed sensor network technology into lightening devices which are indispensable to most living spaces. To verify practicality of the proposed u-health system, a prototypical system is realized in the Smart Home Industrialization Support Center at Kookmin University, and is tested within many practical circumstances. The proposed u-health system can be used at various places where many patients are continuously cared.

• The Proceedings of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.7 no.1
• /
• pp.21-29
• /
• 1993

In this paper, we studied the 50% flashover voltage of lightening impulse which affect the most serious damages on the insulation of the electric power network system. Also its V -t characteristics and corona process phenomena of pure $SF_6, N_2, SF_6-N_2$mixtures under the circumstances of nonuniform field gap are researched. Comparing the characteristics of pure $SF_6$ with that of $SF_6, N_2$mixtures, we discussed that breakdown processes and $SF_6, N_2$ mixture's application to economics.As a results, 50% flashover voltage of $SF_6$ 50% - $N_2$ 50% for impulse voltage is higher then that of 80% of pure SF6, measured data and calculated data by equal area law are almost equal from the points of view of V-t characteristics. Therefore, it has been known that $SF_6$ 50% - $N_2$ 50% mixtures can be used as an economic constitution gas of pure $SF_6$, it is verified that corona processes from Lichtenberg figure.

• Journal of Industrial Technology
• /
• v.28 no.B
• /
• pp.143-148
• /
• 2008

A real-time face detection is to find human faces robustly under the cluttered background free from the effect of occlusion by other objects or various lightening conditions. We propose a face detection system for real-time applications using wavelet decomposition method based on Gabor features. Firstly, skin candidate regions are extracted from the given image by skin color filtering and projection method. Then Gabor-feature based template matching is performed to choose face cadidate from the skin candidate regions. The chosen face candidate region is transformed into 2-level wavelet decomposition images, from which feature vectors are extracted for classification. Based on the extracted feature vectors, the face candidate region is finally classified into either face or nonface class by the Levenberg-Marguardt back-propagation neural network.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.14 no.3
• /
• pp.69-75
• /
• 2018

Recently, various convergence technologies are attracting attention due to the block chain innovation technology in the M2M environment. Although the block-chain-based technology is known to be secure in its own right, there are various problems such as security and weight reduction in various M2M environments connected with this. In this paper, we propose a new lightweight method for the hash tree generation of block chains to solve the lightweight problem. It is designed considering extensibility without affecting the existing block chain. Performance analysis shows that the computation performance increases with decreasing the existing hash length.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.221-223
• /
• 2022

최근 CNN(Convolutional Neural Network)은 초해상화(super-resolution)를 포함한 다양한 컴퓨터 비전 분야에서 우수한 성능을 보이며 널리 사용되고 있다. 그러나 CNN은 계산 집약적이고 많은 메모리가 요구되어 한정적인 하드웨어 자원인 모바일이나 IoT(Internet of Things) 기기에 적용하기 어렵다는 문제가 있다. 이런 한계를 해결하기 위해, 기 학습된 깊은 CNN 모델의 성능을 최대한 유지하며 네트워크의 깊이나 크기를 줄이는 경량화 연구가 활발히 진행되고 있다. 본 논문은 네트워크 경량화 기술인 지식증류(knowledge distillation) 중 자가증류(self-distillation)를 초해상화 CNN 모델에 적용하여 성능을 평가, 분석한다. 실험 결과, 정량적 평가지표를 통하여 자가증류를 통해서도 성능이 우수한 경량화된 초해상화 모델을 얻을 수 있음을 확인하였다.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.5
• /
• pp.91-98
• /
• 2020

The mobile agent, which handles a given task while migrating between the sensor nodes, moves including the execution commands and task processing results. This increases the size of the mobile agent, causing the network to load, leading to the migration time delay and the loss of migration reliability. This paper presents the method of lightening the mobile agent using distributed object technology and the algorithm for exploring and providing the optimal migration path that is actively performed in the event of network traffic, and it proposes a method to ensure the reliability of the mobile agent migration by applying them. In addition, through the comparative analysis experiments based on agent size and network traffic for the migration time of mobile agent equipped with active rules in sensor network-based mobile agent middleware environment, applying the proposed methods proves to ensure the autonomy and migration reliability of the mobile agent.

• Transactions on Electrical and Electronic Materials
• /
• v.18 no.3
• /
• pp.151-154
• /
• 2017

In this article, we investigated the various influencing factors that degraded the lifetime of suspension insulators in 154 kV transmission lines, and showed the possible solutions to avoid such breakdowns. With respect to achieve safety, reliability and aesthetical considerations, the characteristics of transmission and distribution network power cables should be improved. Suspension insulators are particularly important to study, as they have developed to be the main component of transmission lines due to their ability to withstand the electrical conductivity of high-voltage power transmission. Suspension insulators are mostly made from glass, rubber and ceramic material due to their high resistivity. In Korea, porcelain suspension insulators are typically used in the transmission line system, as they are cheaper and more flexible compared to other types of insulators. This is effective from preventing very high and steep lightening impulse voltages from causing the breakdown of suspension insulators used in power lines. Other influential factors affect the lifetime of suspension insulators that we studied include temperature, water moisture, contamination, mechanical vibration and electrical stress.