• Journal of KIISE:Software and Applications
• /
• v.29 no.1_2
• /
• pp.1-9
• /
• 2002

Evolutionary artificial neural networks (EANNs) are towards the near optimal ANN using the global search of evolutionary instead of trial-and-error process. However, many real-world problems are too hard to be solved by only one ANN. Recently there has been plenty of interest on combining ANNs in the last generation to improve the performance and reliability. This paper proposes a new approach of constructing multiple ANNs which complement each other by speciation. Also, we develop a multiple ANN to combine the results in abstract, rank, and measurement levels. The experimental results on Australian credit approval data from UCI benchmark data set have shown that combining of the speciated EANNs have better recognition ability than EANNs which are not speciated, and the average error rate of 0.105 proves the superiority of the proposed EANNs.

• Imaging Science in Dentistry
• /
• v.51 no.3
• /
• pp.299-306
• /
• 2021

Purpose: This study aimed to propose a fully automatic landmark identification model based on a deep learning algorithm using real clinical data and to verify its accuracy considering inter-examiner variability. Materials and Methods: In total, 950 lateral cephalometric images from Yonsei Dental Hospital were used. Two calibrated examiners manually identified the 13 most important landmarks to set as references. The proposed deep learning model has a 2-step structure-a region of interest machine and a detection machine-each consisting of 8 convolution layers, 5 pooling layers, and 2 fully connected layers. The distance errors of detection between 2 examiners were used as a clinically acceptable range for performance evaluation. Results: The 13 landmarks were automatically detected using the proposed model. Inter-examiner agreement for all landmarks indicated excellent reliability based on the 95% confidence interval. The average clinically acceptable range for all 13 landmarks was 1.24 mm. The mean radial error between the reference values assigned by 1 expert and the proposed model was 1.84 mm, exhibiting a successful detection rate of 36.1%. The A-point, the incisal tip of the maxillary and mandibular incisors, and ANS showed lower mean radial error than the calibrated expert variability. Conclusion: This experiment demonstrated that the proposed deep learning model can perform fully automatic identification of cephalometric landmarks and achieve better results than examiners for some landmarks. It is meaningful to consider between-examiner variability for clinical applicability when evaluating the performance of deep learning methods in cephalometric landmark identification.

• Journal of Information Technology Applications and Management
• /
• v.28 no.1
• /
• pp.53-61
• /
• 2021

The year 2020 was a time when the coronavirus infections-19 (COVID-19) caused various changes in society. In particular, the fields that have been conducted face-to-face have been greatly confused by the transition to an online non-face-to-face method, and this is the case with the field of education. There are two main advantages of offline education. The first is that we can improve our understanding through communication with teachers, and the second is that we can develop social skills through interaction with friends. But as online classes progressed due to corona 19, interaction could not be achieved. As a result, the motivation for learning has been reduced due to difficulties in real-time feedback, and the participation rate has been significantly lowered, especially in lower grades, raising concerns about the learning gap that will occur after corona 19. However, there are some cases in which online classes were conducted as effectively as offline classes by utilizing various contents. What they have in common is the use of content. Teachers generally improved the quality of education by linking interesting sights and videos that enhance learning comprehension. The provided video conveys learning-related content into stories, enabling intuitive observation. Many students were already enjoying these videos through VOD (Video on Demand) such as TV and YouTube, they were able to connect their easy access to content and interest in learning. Appropriate use of video content has rather increased the learning effect and should continue after corona 19. Therefore, it is necessary to study methodologies that apply video content efficiently to education. This study looked at the steps that needed content application through the development of education programs, and observed its meaning. Students were curious about the content, motivated to learn and participated in learning on their own. Intuitive learning, conducted through appreciation, play and content production, provided an opportunity to learn on their own in everyday life.

• Applied Chemistry for Engineering
• /
• v.32 no.1
• /
• pp.102-109
• /
• 2021

Recently, an interest in risk calculation methods has been increasing in Korea due to the establishment of classification code for explosive hazardous area on gas facility (KGS CODE GC101), which is based on the international standard of classification of areas - explosive gas atmospheres (IEC 60079-10-1). However, experiments to check for leaks of combustible or toxic gases are very difficult. These experiments can lead to fire, explosion, and toxic poisoning. Therefore, even if someone tries to provide a laboratory for this experiment, it is difficult to install a gas leakage equipment. In this study we find out differences among actual experiments, CFD by using FLACS and calculation based on classification code for explosive hazardous area on gas facility (KGS CODE GC101) by comparing to each other. We develpoed KGS HAC (hazardous area classification) program which based on KGS GC101 for convenience and popularization. As a result, actual gas leak, CFD and KGS HAC are showing slightly different results. The results of dispersion of 1.8 to 2.7 m were shown in the actual experiment, and the CFD and KGS HAC showed a linear increase of about 0.4 to 1 m depending on the increase in a flow rate. In the actual experiment, the application of 3/8" tubes and orifice to take into account the momentum drop resulted in an increase in the hazardous distance of about 1.95 m. Comparing three methods was able to identify similarities between real and CFD, and also similarities and limitations of CFD and KGS HAC. We hope these results will provide a good basis for future experiments and risk calculations.

• Journal of Information Technology Services
• /
• v.21 no.2
• /
• pp.97-107
• /
• 2022

Recently, interest in smart factories is increasing. Investments to improve intelligence/automation are also being made continuously in manufacturing plants. Facility automation based on sensor data collection is now essential. In addition, we are operating our factories based on data generated in all areas of production, including production management, facility operation, and quality management, and an integrated standard information system. When producing LCD polarizer products, it is most important to link trace information between data generated by individual production processes. All systems involved in production must ensure that there is no data loss and data integrity is ensured. The large-capacity data collected from individual systems is composed of key values linked to each other. A real-time quality analysis processing system based on connected integrated system data is required. In this study, large-capacity data collection, storage, integration and loss prevention methods were presented for optimization of LCD polarizer production. The identification Risk model of inspection products can be added, and the applicable product model is designed to be continuously expanded. A quality inspection and analysis system that maximizes the yield rate was designed by using the final inspection image of the product using big data technology. In the case of products that are predefined as analysable products, it is designed to be verified with the big data knn analysis model, and individual analysis results are continuously applied to the actual production site to operate in a virtuous cycle structure. Production Optimization was performed by applying it to the currently produced LCD polarizer production line.

• Progress in Superconductivity
• /
• v.4 no.1
• /
• pp.1-9
• /
• 2002

Transferring single flux quanta across a Josephson junction at an exactly determined rate has made highly precise voltage measurements possible. Making use of self-shunted Nb-based SINIS junctions, programmable fast-switching DC voltage standards with output voltages of up to 10 V were produced. This development is now extended from fundamental DC measurements to the precise determination of AC voltages with arbitrary waveforms. Integrated RSFQ circuits will help to replace expensive semiconductor devices for frequency control and signal coding. Easy-to-handle AC and inexpensive quantum voltmeters of fundamental accuracy would be of interest to industry. In analogy to the development in the flux regime, metallic nanocircuits comprising small-area tunnel junctions and providing the coherent transport of single electrons might play an important role in quantum current metrology. By precise counting of single charges these circuits allow prototypes of quantum standards for electric current and capacitance to be realised. Replacing single electron devices by single Cooper pair circuits, the charge transfer rates and thus the quantum currents could be significantly increased. Recently, the principles of the gate-controlled transfer of individual Cooper pairs in superconducting A1 devices in different electromagnetic environments were demonstrated. The characteristics of these quantum coherent circuits can be improved by replacing the small aluminum tunnel Junctions by niobium junctions. Due to the higher value of the superconducting energy gap ($\Delta_{Nb}$ ＝ $7\Delta_{Al}$), the characteristic energy and the frequency scales for Nb devices are substantially extended as compared to A1 devices. Although the fabrication of small Nb junctions presents a real challenge, the Nb-based metrological devices will be faster and more accurate in operation. Moreover, the Nb-based Cooper pair electrometer could be coupled to an Nb single Cooper pair qubit which can be beneficial for both, the stability of the qubit and its readout with a large signal-to-noise ratio..

• Journal of the Microelectronics and Packaging Society
• /
• v.31 no.1
• /
• pp.23-28
• /
• 2024

Electronic products utilizing flexible devices experience harsh mechanical deformations in real-use environments. As a result, researches on the mechanical reliability of these flexible devices have attracted considerable interest among researchers. This study employed previous bending strain models and finite element analysis to predict the maximum bending strain of metal films deposited on flexible substrates. Bending experiments were simulated using finite element analysis with variations in the material and thickness of the thin films, and the substrate thickness. The results were compared with the strains predicted by existing models. The distribution of strain on the surface of film was observed, and the error rate of the existing model was analyzed during bending. Additionally, a modified model was proposed, providing mathematical constants for each case.

• Journal of Biomedical Engineering Research
• /
• v.25 no.5
• /
• pp.391-401
• /
• 2004

3D imaging systems using 2D phased arrays have a large number of active channels, compelling to use a very expensive and bulky beamforming hardware, and suffer from low volume rate because, in principle, at least one ultrasound transmit-receive event is necessary to construct each scanline. A high speed 3D imaging method using a cross array proposed previously to solve the above limitations can implement fast scanning and dynamic focusing in the lateral direction but suffer from low resolution except at the fixed transmit focusing along the elevational direction. To overcome these limitations, we propose a new real-time volumetric imaging method using a cross array based on the synthetic aperture technique. In the proposed method, ultrasound wave is transmitted successively using each elements of an 1D transmit array transducer, one at a time, which is placed along the elevational direction and for each firing, the returning pulse echoes are received using all elements of an 1D receive array transducer placed along the lateral direction. On receive, by employing the conventional dynamic focusing and synthetic aperture method along lateral and elevational directions, respectively, ultrasound waves can be focused effectively at all imaging points. In addition, in the proposed method, a volume of interest consisting of any required number of slice images, can be constructed with the same number of transmit-receive steps as the total number of transmit array elements. Computer simulation results show that the proposed method can provide the same and greatly improved resolutions in the lateral and elevational directions, respectively, compared with the 3D imaging method using a cross array based on the conventional fixed focusing. In the accompanying paper, we will also propose a new real-time 3D imaging method using a cross array for improving transmit power and elevational spatial resolution, which uses linear wave fronts on transmit.

• Journal of Internet Computing and Services
• /
• v.20 no.6
• /
• pp.55-63
• /
• 2019

Recently, the use of real-world image data has been increasing to express realistic virtual environments in various application fields such as education, manufacturing, and construction. In particular, with increasing interest in digital twins like smart cities, realistic 3D urban models are being built using real-world images, such as aerial images. However, the captured aerial image includes shadows from the sun, and the 3D city model including the shadows has a problem of distorting and expressing information to the user. Many studies have been conducted to remove the shadow, but it is recognized as a challenging problem that is still difficult to solve. In this paper, we construct a virtual environment dataset including the height map of buildings using 3D spatial information provided by VWorld, and We propose a new shadow detection method using height map and deep learning. According to the experimental results, We can observed that the shadow detection error rate is reduced when using the height map.

• Journal of Radiation Protection and Research
• /
• v.11 no.1
• /
• pp.3-14
• /
• 1986

A stud has been carried out for figuring out real photon spectrum from an observed gamma-ray spectrum by means of response matrix method, which is known one of the relatively convenient method for the estimation of exposure rate of a complex gamma ray field in comparison with graphical analysis and least square fitting of the measured spectrum. A 3'${\times}$3' cylindrical Nal(T1) scintillation detector in association with multichannel pulse height analyzer and six reference gamma ray sources covering the photon energy range of 0.05 to 2.0 MeV were used. In dividing the energy region for the construction of response matrix, two different approaches were attempted. One is dividing the entire energy region of interest into 20 bins, one of which corresponds to a width of 0.1 MeV to form $20{\times}20$ matrix, and another is dividing the 2 MeV region into 14 bins to form $14{\times}14$ matrix consists of $0.1(MeV)^{1/2}$ intervals assuming the resolution of the detector is dependent on square root of the incident photon energy. Inversion of thus constructed matrices was performed by a computor(P-E8/32) using the program attached to the end of this paper. The resultant exposure rates obtained by this method were in good agreement, within 10% with those calculated by ordinary formula widely used for a gamma-ray field of known energy and flux. It is concluded that the photen flux obtained by the response matrix constructed under the assumption of $E^{1/2}$ dependence is more realistic than that obtained by the matrix consist of identical energy bins in dosimetrical point of view.