• Journal of IKEEE
• /
• v.23 no.1
• /
• pp.112-119
• /
• 2019

In this paper, we propose a deep learning structure suitable for embedded system. The flame detection process of the proposed deep learning structure consists of four steps : flame area detection using flame color model, flame image classification using deep learning structure for flame color specialization, $N{\times}N$ cell separation in detected flame area, flame image classification using deep learning structure for flame shape specialization. First, only the color of the flame is extracted from the input image and then labeled to detect the flame area. Second, area of flame detected is the input of a deep learning structure specialized in flame color and is classified as flame image only if the probability of flame class at the output is greater than 75%. Third, divide the detected flame region of the images classified as flame images less than 75% in the preceding section into $N{\times}N$ units. Fourthly, small cells divided into $N{\times}N$ units are inserted into the input of a deep learning structure specialized to the shape of the flame and each cell is judged to be flame proof and classified as flame images if more than 50% of cells are classified as flame images. To verify the effectiveness of the proposed deep learning structure, we experimented with a flame database of ImageNet. Experimental results show that the proposed deep learning structure has an average resource occupancy rate of 29.86% and an 8 second fast flame detection time. The flame detection rate averaged 0.95% lower compared to the existing deep learning structure, but this was the result of light construction of the deep learning structure for application to embedded systems. Therefore, the deep learning structure for flame detection proposed in this paper has been proved suitable for the application of embedded system.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.97-97
• /
• 2021

Hydrometeorological variables contain highly complex system for temporal revolution and it is quite challenging to illustrate the system with a temporal linear and nonlinear models. In recent years, deep learning algorithms have been developed and a number of studies has focused to model the complex hydrometeorological system with deep learning models. In the current study, we investigated the temporal structure inside deep learning models for the hydrometeorological variables such as streamflow and climate indices. The results present a quite striking such that each hidden unit of the deep learning model presents different dependence structure and when the number of hidden units meet a proper boundary, it reaches the best model performance. This indicates that the deep dependence structure of deep learning models can be used to model selection or investigating whether the constructed model setup present efficient or not.

• Journal of Ocean Engineering and Technology
• /
• v.26 no.1
• /
• pp.78-83
• /
• 2012

The abundant nutrients contained in deep seawater are delivered by natural upwellings from the deep sea to the surface sea. However, the natural upwelling phenomenon is limited to specific areas of the sea; in other areas, the thermocline separates the surface sea from the lower layer. Thus, the surface layer is often deficient in nutritive salts, causing the deterioration of its primary productivity and ultimately leading to an imbalance in the marine ecosystem. Without a consistent supply of nitrogenous nutritive salts, they are absorbed by phytoplankton, resulting in a considerable problem in primary productivity. To solve this issue, a floating type of artificial upwelling system is suggested to artificially pump up, distribute, and diffuse deep seawater containing rich nutritive salts. The key technologies for developing such a floating artificial upwelling system are a floating offshore structure with a large diameter riser, self-supplying energy system, density current generating system, method for estimating the emission and absorption of CO2, and way to evaluate the primary production variation. Strengthening the primary production of the sea by supplying deep seawater to the sea surface will result in a sea environment with abundant fishery resources.

• Journal of IKEEE
• /
• v.22 no.4
• /
• pp.933-940
• /
• 2018

In this paper, we propose a multi-block deep learning structure for improving face recognition rate. The recognition structure of the proposed deep learning consists of three steps: multi-blocking of the input image, multi-block selection by facial feature numerical analysis, and perform deep learning of the selected multi-block. First, the input image is divided into 4 blocks by multi-block. Secondly, in the multi-block selection by feature analysis, the feature values of the quadruple multi-blocks are checked, and only the blocks with many features are selected. The third step is to perform deep learning with the selected multi-block, and the result is obtained as an efficient block with high feature value by performing recognition on the deep learning model in which the selected multi-block part is learned. To evaluate the performance of the proposed deep learning structure, we used CAS-PEAL face database. Experimental results show that the proposed multi-block deep learning structure shows 2.3% higher face recognition rate than the existing deep learning structure.

• Journal of IKEEE
• /
• v.25 no.2
• /
• pp.356-361
• /
• 2021

Super Junction structure is the proposed structure to minimize the Trade-off phenomenon of power devices. Super Junction can have On-resistance(Ron) characteristics as less as five times than conventional structure. There are process methods that Multi-Epi and Deep-Trench of Super Junction structure. The reason for this is that Deep-Trench process is known to be a relatively difficult manufacturing method because it is easy to form a P-Pillar by burying impurities on top of a silicon substrate through a Deep-Trench process. However, the structure created by the Deep-Trench process has low On-resistance and high breakdown voltage, showing better efficiency. In this paper, we suggested a novel method in the process and designed structure with Charge Balance theory.

• International journal of advanced smart convergence
• /
• v.10 no.3
• /
• pp.51-58
• /
• 2021

In this paper, we propose a deep learning classifier for estimating the number of layers in the Earth's structure. When installing a grounding system, knowledge of the subsurface in the area is absolutely necessary. The subsurface structure can be modeled by the earth parameters. Knowing the exact number of layers can significantly reduce the amount of computation to estimate these parameters. The classifier consists of a feedforward neural network. Apparent resistivity curves were used to train the deep learning classifier. The apparent resistivity at 20 equally spaced log points in each curve are used as the features for the input of the deep learning classifier. Apparent resistivity curve data sets are collected either by theoretical calculations or by Wenner's measurement method. Deep learning classifiers are coded by Keras, an open source neural network library written in Python. This model has been shown to converge with close to 100% accuracy.

• Journal of Korea Multimedia Society
• /
• v.24 no.4
• /
• pp.519-527
• /
• 2021

Alzheimer's disease is one of the challenges to tackle in the coming aging era and is attempting to diagnose and predict through various biomarkers. While the application of various deep learning-based technologies as powerful imaging technologies has recently expanded across the medical industry, empirical design is not easy because there are various deep earning neural networks architecture and categorical hyperparameters that rely on problems and data to solve. In this paper, we show the possibility of optimizing a deep learning neural network structure and hyperparameters for Alzheimer's disease classification in amyloid brain images in a representative deep earning neural networks architecture using genetic algorithms. It was observed that the optimal deep learning neural network structure and hyperparameter were chosen as the values of the experiment were converging.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.12
• /
• pp.1772-1781
• /
• 2017

In this study, the design methodology for alleviating the overfitting problem of Polynomial Neural Networks(PNN) is realized with the aid of two kinds techniques such as L2 regularization and Sum of Squared Coefficients (SSC). The PNN is widely used as a kind of mathematical modeling methods such as the identification of linear system by input/output data and the regression analysis modeling method for prediction problem. PNN is an algorithm that obtains preferred network structure by generating consecutive layers as well as nodes by using a multivariate polynomial subexpression. It has much fewer nodes and more flexible adaptability than existing neural network algorithms. However, such algorithms lead to overfitting problems due to noise sensitivity as well as excessive trainning while generation of successive network layers. To alleviate such overfitting problem and also effectively design its ensuing deep network structure, two techniques are introduced. That is we use the two techniques of both SSC(Sum of Squared Coefficients) and $L_2$ regularization for consecutive generation of each layer's nodes as well as each layer in order to construct the deep PNN structure. The technique of $L_2$ regularization is used for the minimum coefficient estimation by adding penalty term to cost function. $L_2$ regularization is a kind of representative methods of reducing the influence of noise by flattening the solution space and also lessening coefficient size. The technique for the SSC is implemented for the minimization of Sum of Squared Coefficients of polynomial instead of using the square of errors. In the sequel, the overfitting problem of the deep PNN structure is stabilized by the proposed method. This study leads to the possibility of deep network structure design as well as big data processing and also the superiority of the network performance through experiments is shown.

• Journal of the Korean Society of Mechanical Technology
• /
• v.20 no.6
• /
• pp.776-782
• /
• 2018

After CNN basic structure was introduced by LeCun in 1989, there has not been a major structure change except for more deep network until recently. The deep network enhances the expression power due to improve the abstraction ability of the network, and can learn complex problems by increasing non linearity. However, the learning of a deep network means that it has vanishing gradient or longer learning time. In this study, we proposes a CNN structure with MLP layer. The proposed CNNs are superior to the general CNN in their classification performance. It is confirmed that classification accuracy is high due to include MLP layer which improves non linearity by experiment. In order to increase the performance without making a deep network, it is confirmed that the performance is improved by increasing the non linearity of the network.

• Journal of the Korea Society for Simulation
• /
• v.3 no.2
• /
• pp.1-13
• /
• 1994

Deep reasoning procedures are model-based, inferring single or multiple causes and/or timing relations from the knowledge of behavior of component models and their causal structure. The overall goal of this paper is to develop an automated deep reasoning methodology that exploits deep knowledge of structure and behavior of a system. We have proceeded by building a software environment that uses such knowledge to reason from advanced symbolic simulation techniques introduced by Chi and Zeigler. Such reasoning system has been implemented and tested on several examples in the domain of performance evaluation, and event-based control.