• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.4A
• /
• pp.809-816
• /
• 2006

More than effectively judging the existence of reinforcing bars under multi boundary layers and void shapes in concrete, this study aims to develop the analysis algorithm of radar response on multi boundary layers in reinforced concrete and radar capable of estimation of the shape of specific voids in plain concrete. To detect or estimate reinforcing bars and void shapes in these conditions, the simulation analysis model of transmission and reflection wave of electromagnetic radar is used. This radar simulation model is carried out with reinforced or non reinforced concrete of various boundary conditions and void shapes. And, the output signals (images) of radar simulation results are calculated and represented by convolution method. As the results, it is clarified that this simulation analysis technique can be used to analyze radar response on multi boundary layers in reinforced concrete and void shapes in concrete.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.3
• /
• pp.459-466
• /
• 2006

Numerical values of thermodynamic properties such as temperature, pressure, dryness, volume, enthalpy and entropy are required in numerical analysis on evaluating the thermal performance. But the steam table itself cannot be used without modelling. From this point of view the neural network with function approximation characteristics can be an alternative. the multi-layer neural networks were made for saturated vapor region and superheated vapor region separately. For saturated vapor region the neural network consists of one input layer with 1 node, two hidden layers with 10 and 20 nodes each and one output layer with 7 nodes. For superheated vapor region it consists of one input layer with 2 nodes, two hidden layers with 15 and 25 nodes each and one output layer with 3 nodes. The proposed model gives very successful results with ${\pm}0.005%$ of percentage error for temperature, enthalpy and entropy and ${\pm}0.025%$ for pressure and specific volume. From these successful results, it is confirmed that the neural networks could be powerful method in function approximation of the steam table.

• Journal of the Optical Society of Korea
• /
• v.18 no.3
• /
• pp.256-260
• /
• 2014

Columnar-structured cesium iodide (CsI) scintillators doped with thallium (Tl) are frequently used as x-ray converters in medical and industrial imaging. In this study we investigated the imaging characteristics of CsI:Tl films with various reflective layers-aluminum (Al), chromium (Cr), and titanium dioxide ($TiO_2$) powder-coated on glass substrates. We used two effusion-cell sources in a thermal evaporator system to fabricate CsI:Tl films on substrates. The scintillators were observed via scanning electron microscopy (SEM), and scintillation characteristics were evaluated on the basis of the emission spectrum, light output, light response to x-ray dose, modulation transfer function (MTF), and x-ray images. Compared to control films without a reflective layer, CsI:Tl films with reflective layers showed better sensitivity and light collection efficiency, and the film with a $TiO_2$ reflective layer showed the best properties.

• Structural Engineering and Mechanics
• /
• v.77 no.1
• /
• pp.47-56
• /
• 2021

This paper deals with damage detection using a Gapped Smoothing Method (GSM) combined with deep learning. Convolutional Neural Network (CNN) is a model of deep learning. CNN has an input layer, an output layer, and a number of hidden layers that consist of convolutional layers. The input layer is a tensor with shape (number of images) × (image width) × (image height) × (image depth). An activation function is applied each time to this tensor passing through a hidden layer and the last layer is the fully connected layer. After the fully connected layer, the output layer, which is the final layer, is predicted by CNN. In this paper, a complete machine learning system is introduced. The training data was taken from a Finite Element (FE) model. The input images are the contour plots of curvature gapped smooth damage index. A free-free beam is used as a case study. In the first step, the FE model of the beam was used to generate data. The collected data were then divided into two parts, i.e. 70% for training and 30% for validation. In the second step, the proposed CNN was trained using training data and then validated using available data. Furthermore, a vibration experiment on steel damaged beam in free-free support condition was carried out in the laboratory to test the method. A total number of 15 accelerometers were set up to measure the mode shapes and calculate the curvature gapped smooth of the damaged beam. Two scenarios were introduced with different severities of the damage. The results showed that the trained CNN was successful in detecting the location as well as the severity of the damage in the experimental damaged beam.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.7
• /
• pp.37-44
• /
• 2021

Automatic classification of brain MRI images play an important role in early diagnosis of brain tumors. In this work, we present a deep learning-based brain tumor classification model in MRI images using ensemble of deep features. In our proposed framework, three different deep features from brain MR image are extracted using three different pre-trained models. After that, the extracted deep features are fed to the classification module. In the classification module, the three different deep features are first fed into the fully-connected layers individually to reduce the dimension of the features. After that, the output features from the fully-connected layers are concatenated and fed into the fully-connected layer to predict the final output. To evaluate our proposed model, we use openly accessible brain MRI dataset from web. Experimental results show that our proposed model outperforms other machine learning-based models.

• International Journal of Computer Science & Network Security
• /
• v.22 no.7
• /
• pp.91-102
• /
• 2022

With the evolution of wireless sensor network (WSN) technology, the routing policy has foremost importance in the Internet of Things (IoT). A systematic routing policy is one of the primary mechanics to make certain the precise and robust transmission of wireless sensor networks in an energy-efficient manner. In an IoT environment, WSN is utilized for controlling services concerning data like, data gathering, sensing and transmission. With the advantages of IoT potentialities, the traditional routing in a WSN are augmented with decision-making in an energy efficient manner to concur finer optimization. In this paper, we study how to combine IoT-based deep learning classifier with routing called, Kriging Regressive Deep Belief Neural Learning (KR-DBNL) to propose an efficient data packet routing to cope with scalability issues and therefore ensure robust data packet transmission. The KR-DBNL method includes four layers, namely input layer, two hidden layers and one output layer for performing data transmission between source and destination sensor node. Initially, the KR-DBNL method acquires the patient data from different location. Followed by which, the input layer transmits sensor nodes to first hidden layer where analysis of energy consumption, bandwidth consumption and light intensity are made using kriging regression function to perform classification. According to classified results, sensor nodes are classified into higher performance and lower performance sensor nodes. The higher performance sensor nodes are then transmitted to second hidden layer. Here high performance sensor nodes neighbouring sensor with higher signal strength and frequency are selected and sent to the output layer where the actual data packet transmission is performed. Experimental evaluation is carried out on factors such as energy consumption, packet delivery ratio, packet loss rate and end-to-end delay with respect to number of patient data packets and sensor nodes.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.2
• /
• pp.465-470
• /
• 2024

The advancement of AI-based image generation technology has resulted in the creation of various images, emphasizing the need for technology capable of accurately discerning them. The amount of generated image data is limited, and to achieve high performance with a limited dataset, this study proposes a model for discriminating generated images using transfer learning. Applying pre-trained models from the ImageNet dataset directly to the CIFAKE input dataset, we reduce training time cost followed by adding three hidden layers and one output layer to fine-tune the model. The modeling results revealed an improvement in the performance of the model when adjusting the final layer. Using transfer learning and then adjusting layers close to the output layer, small image data-related accuracy issues can be reduced and generated images can be classified.

• Atmosphere
• /
• v.24 no.1
• /
• pp.49-68
• /
• 2014

This study analyzed atmospheric conditions for the convergent cloud band (Cu-Cb line) in developing stage and its neighbouring convections formed over the East Sea on 1 February 2012, by using synoptic, satellites data, and WRF numerical simulation output of high resolution. In both satellite images and the WRF numerical simulation outputs, the Cu-Cb line that stretched out toward northwest-southeast was shown in the East Sea, and cloud lines of the L mode were aligned in accordance with the prevailing surface wind direction. However, those of the T mode were aligned in the direction of NE-SW, which was nearly perpendicular direction to the surface winds. The directions of the wind shear vectors connecting top winds and bottom winds of the moist layers of the L mode and the T mode were identical with those of the cloud lines of L mode and T mode, respectively. From the WRF simulation convection circulations with a convergence in the lower layer of atmosphere and a divergence above 1.5 km ASL (Above Sea Level) were identified in the Cu-Cb line. A series of small sized vortexes (maximum vortex: $320{\times}10^{-5}s^{-1}$) of meso-${\gamma}$-scale formed by convergences was found along the Cu-Cb lines, suggesting that Cu-Cb lines, consisting of numerous convective clouds, were closely associated with a series of the small vortexes. There was an absolute unstable layer (${\partial}{\theta}/{\partial}z$ < 0) between sfc and ~0.3 km ASL, and a stable layer (${\partial}{\theta}/{\partial}z$ > 0) above ~2 km ASL over the Cu-Cb line and cloud zones. Not only convectively unstable layers (${\partial}{\theta}_e/{\partial}z$ < 0) but also neutral layers (${\partial}{\theta}_e/{\partial}z{\approx}=0$) in the lower atmosphere (sfc~1.5 km ASL) were scattered around over the cloud zones. Particularly, for the Cu-Cb line there were convectively unstable layers in the surface layer, and neutral layers (${\partial}{\theta}_e/{\partial}z{\approx}=0$) between 0.2 and ~1.5 km ASL over near the center of the Cu-Cb line, and the neutralization of unstable layers came from the release of convective instability.

• Proceedings of the KIEE Conference
• /
• 2008.07a
• /
• pp.1246-1247
• /
• 2008

With recent advanced in portable electric devices, wireless sensor, MEMS and bio-Mechanics device, the new typed power supply, not conventional battery but self-powered energy source is needed. Particularly, the system that harvests from their environments are interests for use in self powered devices. For very low powered devices, environmental energy may be enough to use power source. Therefore, in other to made piezoelectric energy harvesting device, PMN-PZT thick film was formed by the screen printing method on the Ag/Pd coated alumina substrate. The layer was 8 layers and slurry where a-terpineol, ethycellulose, ferro B-75001 as Vehicle, PMN-PZT powder used are fabricated by ball mill. The output power quality was be also investigated by changing the load resistance, weight and frequency. The made piezoelectric energy harvesting device was resulted from the conditions of 33$k{\Omega}$, 0.25g, 197Hz respectively. The thick film was prepared at the condition of 2.75Vrms, and its power was 230${\mu} W$ and its thickness was 56${mu}m$. The piezoelectric energy harvesting device output voltage was increased, when the load weight, load resistance was increasing and resonance frequency was diminishing. The other side, resonance frequency was diminished, when the weight was increasing. And output power was continuously it changed by load resistance, output voltage, weight and resonance frequency.

• Korean Journal of Materials Research
• /
• v.22 no.11
• /
• pp.626-630
• /
• 2012

The properties of SOFC unit cells manufactured using the decalcomania method were investigated. SOFC unit cell manufacturing using the decalcomania method is a very simple process. In order to minimize the ohmic loss of flattened tube type anode supports of solid oxide fuel cells(SOFC), the cells were fabricated by producing an anode function layer, YSZ electrolyte, LSM electrode, etc., on the supports and laminating them. The influence of these materials on the power output characteristics was studied when laminating the components and laminating the anode function layer between the anode and the electrolyte to improve the output characteristics. Regarding the performance of the SOFC unit cell, the output was 246 $mW/cm^2$ at a temperature of $800^{\circ}C$ in the case of not laminating the anode function layer; however, this value was improved by a factor of two to 574 $mW/cm^2$ due to the decrease of the ohmic resistance and polarization resistance of the cell in the case of laminating the anode function layer. The outputs appeared to be as high as 574 and 246 $mW/cm^2$ at a temperature of $800^{\circ}C$ in the case of using decalcomania paper when laminating the electrolyte layer using the in dip-coating method; however, the reason for this is that interfacial adhesion was improved due to the dense structure, which leads to a thin thickness of the electrolyte layer.