• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.46 no.1
• /
• pp.55-67
• /
• 2023

Lightweight face recognition models, as one of the most popular and long-standing topics in the field of computer vision, has achieved vigorous development and has been widely used in many real-world applications due to fewer number of parameters, lower floating-point operations, and smaller model size. However, few surveys reviewed lightweight models and reimplemented these lightweight models by using the same calculating resource and training dataset. In this survey article, we present a comprehensive review about the recent research advances on the end-to-end efficient lightweight face recognition models and reimplement several of the most popular models. To start with, we introduce the overview of face recognition with lightweight models. Then, based on the construction of models, we categorize the lightweight models into: (1) artificially designing lightweight FR models, (2) pruned models to face recognition, (3) efficient automatic neural network architecture design based on neural architecture searching, (4) Knowledge distillation and (5) low-rank decomposition. As an example, we also introduce the SqueezeFaceNet and EfficientFaceNet by pruning SqueezeNet and EfficientNet. Additionally, we reimplement and present a detailed performance comparison of different lightweight models on the nine different test benchmarks. At last, the challenges and future works are provided. There are three main contributions in our survey: firstly, the categorized lightweight models can be conveniently identified so that we can explore new lightweight models for face recognition; secondly, the comprehensive performance comparisons are carried out so that ones can choose models when a state-of-the-art end-to-end face recognition system is deployed on mobile devices; thirdly, the challenges and future trends are stated to inspire our future works.

• Journal of IKEEE
• /
• v.28 no.2
• /
• pp.123-130
• /
• 2024

As technology advances swiftly and the lifespan of products becomes increasingly short, there is a demand to fasten the pace of research outcomes, product development, and market introduction. As a result, the researchers and developers need a computational experiment environment that enables rapid verification of the experiment and application of research findings. Such an environment must efficiently harness all available computational resources, manage simulations across diverse test scenarios, and support the experimental data collection. This research introduces the design and implementation of an experimental frame based on a microservice architecture. The experimental frame leverages scripts to utilize computing resources optimally, making it more straightforward for users to conduct simulations. It features an experimental frame capable of automatically deploying scenarios to the computing components. This setup allows for the automatic configuration of both the computing environment and experiments based on user-provided scenarios and experimental software, facilitating effortless execution of simulations.

• International Journal of Highway Engineering
• /
• v.15 no.2
• /
• pp.19-27
• /
• 2013

PURPOSES : This study aims to investigate the snow-melt effects of an underground electric heater's snow-melt system via a field performance test, for evaluating the suitability of the system for use on a concrete pavement. The study also investigates the effectiveness of dynamic measures for clearing snow after snowfall events. METHODS : In order to check the field applicability, in November 2010, specimens were prepared from materials used for constructing concrete pavements, and underground electric heating meshes (HOT-mesh) were buried at depths of 50 mm and 100 mm at the site of the Incheon International Airport Construction Research Institute. Further, an automatic heating control system, including a motion sensor and pavement-temperature-controlled sensor, were installed at the site; the former sensor was intended for determining snow-melt effects of the heating control system for different snowfall intensities. Pavement snow-melt effects on snowy days from December 2010 to January 2011 were examined by managing the electric heating meshes and the heating control system. In addition, data on pavement temperature changes resulting from the use of the heating meshes and heating control system and on the dependence of the correlation between the outdoor air temperature and the time taken for the required temperature rise on the depth of the heating meshes were collected and analyzed. RESULTS : The effects of the heating control system's preheat temperature and the hot meshes buried at depths of 50 mm and 100 mm on the melting of snow for snowfalls of different intensities have been verified. From the study of the time taken for the specimen's surface temperature to increase from the preheat temperature ($0^{\circ}C$) to the reference temperature ($5{\sim}8^{\circ}C$) for different snowfall intensities, the correlation between the burial depth and outdoor air temperature has been determined to be as follows: Time=15.10+1.141Depth-6.465Temp CONCLUSIONS : The following measures are suggested. For the effective use of the electric heating mesh, it should be located under a slab it may be put to practical use by positioning it under a slab. From the management aspect, the heating control system should be adjusted according to weather conditions, that is, the snowfall intensity.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.8
• /
• pp.1103-1109
• /
• 2021

ECS is a control device so that the warship can perform the mission stably by controlling and monitoring the entire propulsion system. As the recent provisions of the warship, it's propelling system is complicated than past, as the demand performance and mission of the warships are diverse. In accordance with the complicated propulsion system configuration, the demand for automatic control function of the ECS is increasing for convenient and stable propulsion system control for convenient and stable. As a result, verification of ECS stability and reliability is required. In this paper, we develop an interlocking signal simulator for verifying ECS control logic and communication protocol for warship with CODLOG propulsion systems. The simulator developed was implemented to simulate a signal of gas turbine, propulsion motors, diesel generator and 11 kinds of auxiliary equipment. The reliability of ECS was verified through the ECS communication program and the I/O signal static test with the simulator.

• Journal of Intelligence and Information Systems
• /
• v.20 no.2
• /
• pp.59-71
• /
• 2014

System trading is becoming more popular among Korean traders recently. System traders use automatic order systems based on the system generated buy and sell signals. These signals are generated from the predetermined entry and exit rules that were coded by system traders. Most researches on system trading have focused on designing profitable entry and exit rules using technical indicators. However, market conditions, strategy characteristics, and money management also have influences on the profitability of the system trading. Unexpected price deviations from the predetermined trading rules can incur large losses to system traders. Therefore, most professional traders use strategy portfolios rather than only one strategy. Building a good strategy portfolio is important because trading performance depends on strategy portfolios. Despite of the importance of designing strategy portfolio, rule of thumb methods have been used to select trading strategies. In this study, we propose a SVM-based strategy portfolio management system. SVM were introduced by Vapnik and is known to be effective for data mining area. It can build good portfolios within a very short period of time. Since SVM minimizes structural risks, it is best suitable for the futures trading market in which prices do not move exactly the same as the past. Our system trading strategies include moving-average cross system, MACD cross system, trend-following system, buy dips and sell rallies system, DMI system, Keltner channel system, Bollinger Bands system, and Fibonacci system. These strategies are well known and frequently being used by many professional traders. We program these strategies for generating automated system signals for entry and exit. We propose SVM-based strategies selection system and portfolio construction and order routing system. Strategies selection system is a portfolio training system. It generates training data and makes SVM model using optimal portfolio. We make $m{\times}n$ data matrix by dividing KOSPI 200 index futures data with a same period. Optimal strategy portfolio is derived from analyzing each strategy performance. SVM model is generated based on this data and optimal strategy portfolio. We use 80% of the data for training and the remaining 20% is used for testing the strategy. For training, we select two strategies which show the highest profit in the next day. Selection method 1 selects two strategies and method 2 selects maximum two strategies which show profit more than 0.1 point. We use one-against-all method which has fast processing time. We analyse the daily data of KOSPI 200 index futures contracts from January 1990 to November 2011. Price change rates for 50 days are used as SVM input data. The training period is from January 1990 to March 2007 and the test period is from March 2007 to November 2011. We suggest three benchmark strategies portfolio. BM1 holds two contracts of KOSPI 200 index futures for testing period. BM2 is constructed as two strategies which show the largest cumulative profit during 30 days before testing starts. BM3 has two strategies which show best profits during testing period. Trading cost include brokerage commission cost and slippage cost. The proposed strategy portfolio management system shows profit more than double of the benchmark portfolios. BM1 shows 103.44 point profit, BM2 shows 488.61 point profit, and BM3 shows 502.41 point profit after deducting trading cost. The best benchmark is the portfolio of the two best profit strategies during the test period. The proposed system 1 shows 706.22 point profit and proposed system 2 shows 768.95 point profit after deducting trading cost. The equity curves for the entire period show stable pattern. With higher profit, this suggests a good trading direction for system traders. We can make more stable and more profitable portfolios if we add money management module to the system.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2006.06b
• /
• pp.221-226
• /
• 2006

This study is about backwash condition and membrane fouling at continuous filtration process in ballast water treatment. Displayed result that handle particle contaminant and hydrospace organism included a number of ballast that is happened in ship using automatic back washing filter. Reason that removes first contaminant that is included in number of ballast is that heighten processing effect of after processing process of the filter. Another advantage is to drop off the solids with controlling revolution of drum screen in pretreatment filtration process. The capacity of pilot plant was $10m^3/h$. The result of the test, Backwash cycle time and duration time and a signification effect on the efficiency of system and backwash Backwash duration time was determined to be fixed in 6 seconds of the system with more than 95% removal rate, It needed 1hour backwash frequency. Filtration system removal aquatic organism over $70{\mu}m$ in ballast water. This study shows that the filtration treatment system has a potential for the treatment of ballast water.

• Journal of Satellite, Information and Communications
• /
• v.12 no.3
• /
• pp.22-34
• /
• 2017

The K-DRUM(K-water hydrologic & hydraulic Distributed RUnoff Model), a distributed rainfall-runoff model of K-water, calculates predicted runoff and water surface level of a dam using precipitation data. In order to obtain long-term hydrometeorological information, K-DRUM requires long-term weather forecast. In this study, we built a system providing long-term hydrometeorological information using predicted rainfall ensemble of GloSea5(Global Seasonal Forecast System version 5), which is the seasonal meteorological forecasting system of KMA introduced in 2014. This system produces K-DRUM input data by automatic pre-processing and bias-correcting GloSea5 data, then derives long-term inflow predictions via K-DRUM. Web-based UI was developed for users to monitor the hydrometeorological information such as rainfall, runoff, and water surface level of dams. Through this UI, users can also test various dam management scenarios by adjusting discharge amount for decision-making.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.21 no.1
• /
• pp.167-175
• /
• 2019

To effectively deal with the fire in tunnel which is mostly the vehicle fire, it's more important to suppress the fire at early stage. In urban tunnel, however, accessibility to the scene of fire by the fire fighter is very limited due to severe traffic congestion which causes the difficulty with firefighting activity in timely manner and such a problem would be further worsened in underground road (double-deck tunnel) which has been increasingly extended and deepened. In preparation for the disaster in Korea, the range of life safety facilities for installation is defined based on category of the extension and fire protection referring to risk hazard index which is determined depending on tunnel length and conditions, and particularly to directly deal with the tunnel fire, fire extinguisher, indoor hydrant and sprinkler are designated as the mandatory facilities depending on category. But such fire extinguishing installations are found inappropriate functionally and technically and thus the measure to improve the system needs to be taken. Particularly in a double-deck tunnel which accommodates the traffic in both directions within a single tunnel of which section is divided by intermediate slab, the facility or the system which functions more rapidly and effectively is more than important. This study, thus, is intended to supplement the problems with existing tunnel life safety system (fire extinguishing) and develop the remote-controlled automatic fire extinguishing system which is optimized for a double-deck tunnel. Consequently, the system considering low floor height and extended length as well as indoor hydrant for a wide range of use have been developed together with the performance verification and the process for commercialization before applying to the tunnel is underway now.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2000.11b
• /
• pp.472-479
• /
• 2000

A persimmon harvesting vehicle that can be operated in hilly orchards as well as a manipulator that can be used to harvest persimmons located in remote positions in the trees were designed and developed. The vehicle could be operated with keeping balanced position in an inclined field and its working platform could be moved up and down easy to approach fruits in a remote region with the aids of a hydraulic and a electrical and electronics systems. The weight of the vehicle was 927 kg and the center of gravity was located at 427 mm to the inner side from the center of a right driving caterpillar, 607 mm to a rear axle from the center of a front axle, and 562 mm to upward from ground. The automatic level control sensor for leveling the working platform was activated within 14.5 ∼ 16.5 degrees of slope variation. The total length of the manipulator was 1.39 m and weight is 975 g. It was powered by a 12 V geared motor to detach persimmon fruits with a rotational force. The gripper was made of plastic and rubber to increase a frictional force. In a performance evaluation test, static tipping angle, dynamic tipping angle toward front side when the vehicle was moving downward, climbing angle, driving speed of the vehicle were measured or calculated. In persimmon harvesting tests 24.9% of yield was increased by hand picking with the aid of the vehicle and additional 7% of yield were increased when the manipulator was used. Therefore, 99010 of total possible yield was achievable when both of the vehicle and the manipulator were used for the manual persimmon harvesting. Increase in 22.5% of total yield was achieved with the manipulator only.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.2D
• /
• pp.245-253
• /
• 2009

Road facility is the most fundamental infrastructure for traffic and various information for smooth traffic is included in road surfaces. Various information included in road surfaces should be often removed and reinstalled by partial damage and aging. In addition, the existing road surface information should be removed in order to mark new information as traffic information changes. The existing road surface removal method suing grinders and torches had problems such as dangerous working equipment and workers' direct exposure to cars. In addition, although water-jet system using super high-pressure water was used to remove road surface in order of improvement of traditional method, there are another problems such as limitations according to water-tanks and water frost on the road surfaces after work. Therefore, this study analyzes and suggests systems to develop prototype after analyzing dry ice blaster in order to improve the current road surface removal methods. In addition, the study analyzes the possibility of introduction of dry ice blaster through a test for proposing an automatized equipment for new road surface mark removal considering environment and work efficiency, and compare traditional method with introduced dry ice blaster for operating cost.