• Journal of Information Processing Systems
• /
• v.14 no.6
• /
• pp.1464-1479
• /
• 2018

In face recognition, sometimes the number of available training samples for single category is insufficient. Therefore, the performances of models trained by convolutional neural network are not ideal. The small sample face recognition algorithm based on novel Siamese network is proposed in this paper, which doesn't need rich samples for training. The algorithm designs and realizes a new Siamese network model, SiameseFacel, which uses pairs of face images as inputs and maps them to target space so that the $L_2$ norm distance in target space can represent the semantic distance in input space. The mapping is represented by the neural network in supervised learning. Moreover, a more lightweight Siamese network model, SiameseFace2, is designed to reduce the network parameters without losing accuracy. We also present a new method to generate training data and expand the number of training samples for single category in AR and labeled faces in the wild (LFW) datasets, which improves the recognition accuracy of the models. Four loss functions are adopted to carry out experiments on AR and LFW datasets. The results show that the contrastive loss function combined with new Siamese network model in this paper can effectively improve the accuracy of face recognition.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.11
• /
• pp.1574-1580
• /
• 2021

This paper designs and presents the results of an air network simulation radio transmission link applied with a multi-beam active phase array antenna simulator in a testbed system for verifying an air network currently underway as a technology development task. Using the Ku band, the Radio transmission link was designed in consideration of the link budget to satisfy the requirements for the system being developed. Considering short-distance links and long-distance links, the required EIRP and G/T performance scales of multi-beam repeaters and mission planes were applied to confirm the minimum and maximum link margins based on Eb/No. In this Radio Transmission Link design, the application analysis results such as rainfall availability are used to effectively establish standards when selecting the operating radius of the multi-beam relay system and related system standards.

• Fisheries and Aquatic Sciences
• /
• v.11 no.1
• /
• pp.61-69
• /
• 2008

Using empirical data of fish performance and physiological limits as well as physical stimuli and environmental data, a cod-end selectivity model based on a chaotic behavior model using the psycho-hydraulic wheel and neural-network approach was established to predict fish escape or herding responses in trawl and cod-end designs. Fish responses in the cod-end were categorized as escape or herding reactions based on their relative positions and reactions to the net wall. Fish movements were regulated by three factors: escape time, a visual looming effect, and an index of body girth-mesh size. The model was applied to haddock in a North Sea bottom trawl including frequencies of movement components, swimming speed, angular velocity, distance to net wall, and the caught-fish ratio; simulation results were similar to field observations. The ratio of retained fish in the cod-end was limited to 37-95% by optomotor coefficient values of 0.3-1.0 and to 13-67% by looming coefficient values of 0.1-1.0. The selectivity curves generated by this model were sensitive to changes in mesh size, towing speed, mesh type, and mesh shape.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.05a
• /
• pp.712-714
• /
• 2018

The system of this paper designs and implements a used trading system that is commonly used on the Internet. In order to pursue local transactions based on location, we designed the system by dividing the region in consideration of the general characteristics of used transactions that are restricted by transactions through distance and location. To do this, I used MySQL as the database engine and I used PHP, an Internet server-based scripting language, for database interworking. Due to the characteristics of secondary transactions, information security and design of each product are considered.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.05a
• /
• pp.353-355
• /
• 2018

The system of this paper designs and implements a used trading system that is commonly used on the Internet. In order to pursue local transactions based on location, we designed the system by dividing the region in consideration of the general characteristics of used transactions that are restricted by transactions through distance and location. To do this, I used MySQL as the database engine and I used PHP, an Internet server-based scripting language, for database interworking. Due to the characteristics of secondary transactions, information security and design of each product are considered.

• Journal of the Korean Institute of Educational Facilities
• /
• v.22 no.2
• /
• pp.15-22
• /
• 2015

The latest educational curriculum in Korea started in 1997, and the 'Departmentalized Classroom System', which is a distinguishing modern concept compared with conventional pedagogy, has already been used in several high schools since 2009 and will be expanded to all middle school and high schools by 2014. However, most Korean schools are not yet physically ready to implement the new system. Moreover, despite many trials during the last three years, the characteristic and architectural function of a place called 'Home-base' is still vague and has caused many controversial arguments. In fact, a large number of home-base designs failed to accomplish their original task and there are counter evidences that shows reality was quite different over architect's personal and subjective radical plan. Therefore, this study seeks to introduce and objective scientific method based on an mathematical model of an optimized algorithm for planning a home-base area in a school design. The algorithm developed in this study is designed to determine the most efficient location for a home-base area to minimize the distance students typically need to walk.

• Structural Monitoring and Maintenance
• /
• v.10 no.3
• /
• pp.243-260
• /
• 2023

Indonesia has had seismic codes for earthquake-resistant structures designs since 1970 and has been updated five times to the latest in 2019. In updating the Indonesian seismic codes, seismic hazard maps for design also update, and there are changes to the Peak Ground Acceleration (PGA). Indonesian seismic design uses the concept of building performance levels consisting of Immediate occupancy (IO), Life Safety (LS), and Collapse Prevention (CP). Related to this performance level, cases still found that buildings were damaged more than their performance targets after the earthquake. Based on the above issues, this study aims to analyze the performance of base isolation design on existing target buildings and analyze the seismic fragility for a case study in Indonesia. The target building is a prototype design 8-story medium-rise residential building using the reinforced concrete moment frame structure. Seismic fragility analysis uses Incremental Dynamic Analysis (IDA) with Nonlinear Time History Analysis (NLTHA) and eleven selected ground motions based on soil classification, magnitude, fault distance, and earthquake source mechanism. The comparison result of IDA shows a trend of significant performance improvement, with the same performance level target and risk category, the base isolation structure can be used at 1.46-3.20 times higher PGA than the fixed base structure. Then the fragility analysis results show that the fixed base structure has a safety margin of 30% and a base isolation structure of 62.5% from the PGA design. This result is useful for assessing existing buildings or considering a new building's performance.

• Structural Engineering and Mechanics
• /
• v.37 no.6
• /
• pp.575-592
• /
• 2011

For seismic resistant design of critical structures, a dynamic analysis, either response spectrum or time history is frequently required. Owing to the lack of recorded data and the randomness of earthquake ground motion that may be experienced by structure in the future, usually it is difficult to obtain recorded data which fit the requirements (site type, epicenteral distance, etc.) well. Therefore, the artificial seismic records are widely used in seismic designs, verification of seismic capacity and seismic assessment of structures. The purpose of this paper is to develop a numerical method using Artificial Neural Network (ANN) and wavelet packet transform in best basis method which is presented for the decomposition of artificial earthquake records consistent with any arbitrarily specified target response spectra requirements. The ground motion has been modeled as a non-stationary process using wavelet packet. This study shows that the procedure using ANN-based models and wavelet packets in best-basis method are applicable to generate artificial earthquakes compatible with any response spectra. Several numerical examples are given to verify the developed model.

• The KIPS Transactions:PartA
• /
• v.13A no.6 s.103
• /
• pp.555-560
• /
• 2006

In contrast with the earlier remote education that simply recorded off-line classes, modern remote education emphasizes on offering additional functions that could maximize learning efficiency. Usage of such multimedia information as the texts, graphics, sounds, animations is considered fundamental element in offering the additional functions. This paper designs and implements an encoder/decoder that could accommodate the multimedia information with emphasis on demonstrating execution files. Instructors can demonstrate my type of execution files or application data files, and the remote learners can freely try running the corresponding execution files by themselves. Consequently, a high-level of learning efficiency can be achieved by the proposed encoder/decoder.

• Journal of IKEEE
• /
• v.24 no.1
• /
• pp.40-45
• /
• 2020

In this paper, we consider an efficient design of quantizers at sensor nodes for sequence-based localization (SBL) systems which recently show a competitive performance for in-door positioning, Since SBL systems locate targets by partitioning the sensor field into subregions, each with an unique sequence number, we use the distance samples between sensors and the sequences for quantizer design in order to propose a low weight design process. Furthermore, we present a new cost function devised to assign the number of samples and the number of unique sequences uniformly into each of quantization partitions and design quantizers by searching the quantization partitions and codewords that minimize the cost function. We finally conduct experiments to demonstrate that the proposed algorithm offers an outstanding localization performance over typical designs while maintaining a substantial reduction of design complexity.