• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.1
• /
• pp.27-32
• /
• 2019

Research that implements the classification process through Deep Learning algorithm, one of the outstanding human abilities, includes a unimodal model, a multi-modal model, and a multi-modal method using music videos. In this study, the results were better by suggesting a system to analyze each song's spectrum into short samples and vote for the results. Among Deep Learning algorithms, CNN showed superior performance in the category of music genre compared to RNN, and improved performance when CNN and RNN were applied together. The system of voting for each CNN result by Deep Learning a short sample of music showed better results than the previous model and the model with Softmax layer added to the model performed best. The need for the explosive growth of digital media and the automatic classification of music genres in numerous streaming services is increasing. Future research will need to reduce the proportion of undifferentiated songs and develop algorithms for the last category classification of undivided songs.

• The Research Journal of the Costume Culture
• /
• v.27 no.5
• /
• pp.433-448
• /
• 2019

Oversized fashion is again in the spotlight due to the influence of retro fashion. This has created new fashion trends with methods different from those of the past. This analysis examines recent trends by sorting these looks according to new and different methods of judging their appearance. A new categorization of the oversized look and its configurations has been created, one which separates "big" looks, partial changes, and layered looks. This research was based on historical review and previous studies. Three thousand one hundred thirty-six photos of oversized looks that have appeared in collections over the past five years were gathered, and their appearance was categorized according to type. The categorization results showed that big looks (55.1%) were most prevalent, followed by partial alterations (36.35%), and layered looks (8.45%). In comparison to prior oversized clothing production, new permutations of the "Big" look expanded the silhouettes of torso, shoulders, neckline and collar. Partial changes have expanded from the broadened shoulders of the 1980s. Today these styles expand the shoulders and armholes vertically or horizontally, which dramatically exaggerates the sleeves and collar. The layered look no longer simply features overlapping layers but takes the form of over-layering through cuts and insertions. Through such analysis it is clear that modern oversized looks break away from the simple expanded forms and production methods of the past. They now attempt to realize an exaggerated beauty of form regarding each clothing component and also maximize decorative effects through innovative drafting or sewing methods.

• Journal of Digital Convergence
• /
• v.19 no.5
• /
• pp.307-313
• /
• 2021

In this paper, a lightweight network with fewer parameters compared to the existing object detection method is proposed. In the case of the currently used detection model, the network complexity has been greatly increased to improve accuracy. Therefore, the proposed network uses EfficientNet as a feature extraction network, and the subsequent layers are formed in a pyramid structure to utilize low-level detailed features and high-level semantic features. An attention process was applied between pyramid structures to suppress unnecessary noise for prediction. All computational processes of the network are replaced by depth-wise and point-wise convolutions to minimize the amount of computation. The proposed network was trained and evaluated using the PASCAL VOC dataset. The features fused through the experiment showed robust properties for various objects through a refinement process. Compared with the CNN-based detection model, detection accuracy is improved with a small amount of computation. It is considered necessary to adjust the anchor ratio according to the size of the object as a future study.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.32 no.5
• /
• pp.323-331
• /
• 2019

In this study, a computational model approach for the modeling of hydrodynamic pressures acting on radial gates during strong earthquakes is proposed. The use of the dynamic layering method with the Arbitrary Lagrangian Eulerian (ALE) algorithm and the SIMPLE method for simulating free reservoir surface flow in addition to moving boundary interfaces between the fluid domain and a structure due to earthquake excitation are suggested. The verification and validation of the proposed approach are realized by comparisons performed using the renowned formulation derived by the experimental results for vertical and inclined dam surfaces subjected to earthquake excitation. A parameter study for the truncated lengths of the two-dimensional fluid domain demonstrates that twice the water level leads to efficient and converged computational results. Finally, numerical simulations for large radial gates with different curvatures subjected to two strong earthquakes are successfully performed using the suggested computational model.

• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.1141-1147
• /
• 2020

The hornet species are so similar in shape that they are difficult for non-experts to classify, and because the size of the objects is small and move fast, it is more difficult to detect and classify the species in real time. In this paper, we developed a system that classifies hornets species in real time based on a deep learning algorithm using a boundary box. In order to minimize the background area included in the bounding box when labeling the training image, we propose a method of selecting only the head and body of the hornet. It also experimentally compares existing boundary box-based object recognition algorithms to find the best algorithms that can detect wasps in real time and classify their species. As a result of the experiment, when the mish function was applied as the activation function of the convolution layer and the hornet images were tested using the YOLOv4 model with the Spatial Attention Module (SAM) applied before the object detection block, the average precision was 97.89% and the average recall was 98.69%.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.6
• /
• pp.714-721
• /
• 2020

DNN is expanding its use in embedded systems such as robots and autonomous vehicles. For high recognition accuracy, computational complexity is greatly increased, and multiple DNNs are running aperiodically. Therefore, the ability processing multiple DNNs in embedded environments is a crucial issue. Accordingly, multicore based platforms are being released. However, most DNN models are operated in a batch process, and when multiple DNNs are operated in multicore together, the execution time deviation between each DNN may be large and the end-to-end execution time of the whole DNNs could be long depending on how they are allocated to the cores. In this paper, we solve these problems by providing a framework that decompose each DNN into individual layers and then distribute to multicores through a global queue. As a result of the experiment, the total DNN execution time was reduced by 31%, and when operating multiple identical DNNs, the deviation in execution time was reduced by up to 95.1%.

• New & Renewable Energy
• /
• v.18 no.2
• /
• pp.40-49
• /
• 2022

In this study, the phase and electrochemical properties of Co and Ti substituted layered perovskites SmBaCo_2-xTi_xO_5+d (x=0.5, 0.7, 1.0, 1.1, 1.3, and 1.5) were analyzed, and their application as electrodes in solid oxide fuel cells (SOFCs) were evaluated. After calcination at 1300℃ for 6 h, a single phase was observed for two compositions of the SmBaCo_2-xTi_xO_5+d oxide system, SmBaCoTiO_5+d (x=1.0) and SmBaCo_0.9Ti_1.1O_5+d (x=1.1). However, the phases of SmBaCoTiO_5+d (SBCTO) and SmTiO₃ coexisted for compositions with x≥1.3 (Ti content). In contrast, for compositions of x≤0.7, the SmBaCo₂O_5+d phase was observed instead of the SmTiO₃ phase. To evaluate the applicability of these materials as SOFC electrodes, the electrical conductivities were measured under various atmospheres (air, N₂, and H₂). SBCTO exhibited stable semi-conductor electrical conductivity behavior in an air and N₂ atmosphere. However, SBCTO showed insulator behavior at temperatures above 600℃ in a H₂ atmosphere. Therefore, SBCTO may only be used as cathode materials. Moreover, SBCTO had an area specific resistance (ASR) value of 0.140 Ω·cm² at 750℃.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.12
• /
• pp.1835-1845
• /
• 2021

To accurately identify black ice and warn the drivers of information in advance so they can control speed and take preventive measures. In this paper, we propose a lightweight black ice detection network based on infrared road images. A black ice recognition network model based on CNN transfer learning has been developed. Additionally, to further improve the accuracy of black ice recognition, an enhanced lightweight network based on MobileNetV2 has been developed. To reduce the amount of calculation, linear bottlenecks and inverse residuals was used, and four bottleneck groups were used. At the same time, to improve the recognition rate of the model, each bottleneck group was connected to a 3×3 convolutional layer to enhance regional feature extraction and increase the number of feature maps. Finally, a black ice recognition experiment was performed on the constructed infrared road black ice dataset. The network model proposed in this paper had an accurate recognition rate of 99.07% for black ice.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.5
• /
• pp.682-687
• /
• 2022

In this paper, the research results on monolithic three-dimensional integrated-circuit (M3DICs) stacked with tunneling field effect transistors (TFETs) are introduced. Unlike metal-oxide-semiconductor field-effect transistors (MOSFETs), TFETs are designed differently from the layout of symmetrical MOSFETs because the source and drain of TFET are asymmetrical. Various monolithic 3D inverter (M3D-INV) structures and layouts are possible due to the asymmetric structure, and among them, a simple inverter structure with the minimum metal layer is proposed. Using the proposed M3D-INV, this M3D logic gates such as NAND and NOR gates by sequentially stacking TFETs are proposed, respectively. The simulation results of voltage transfer characteristics of the proposed M3D logic gates are investigated using mixed-mode simulator of technology computer aided design (TCAD), and the operation of each logic circuit is verified. The cell area for each M3D logic gate is reduced by about 50% compared to one for the two-dimensional planar logic gates.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.9
• /
• pp.1330-1339
• /
• 2022

3D object detection generally aims to detect relatively large data such as automobiles, buses, persons, furniture, etc, so it is vulnerable to small object detection. In addition, in an environment with limited resources such as embedded devices, it is difficult to apply the model because of the huge amount of computation. In this paper, the accuracy of small object detection was improved by focusing on local features using only one layer, and the inference speed was improved through the proposed knowledge distillation method from large pre-trained network to small network and adaptive quantization method according to the parameter size. The proposed model was evaluated using SUN RGB-D Val and self-made apple tree data set. Finally, it achieved the accuracy performance of 62.04% at mAP@0.25 and 47.1% at mAP@0.5, and the inference speed was 120.5 scenes per sec, showing a fast real-time processing speed.