• Journal of Convergence for Information Technology
• /
• v.9 no.5
• /
• pp.158-164
• /
• 2019

Recently many studies to reduce the noise of dental hand piece which generate inevitably mechanical sound to offend to the ear of a patient have been spotlighted. Generally, methods of adding a sound absorbing material inside the exhaust valve, air pump of machine or automobile are widely reported as optimal way to reduce the mechanical noise. In this paper we studied a new UV laser aided manufacturing of micro-porous structure of EPP substrate and applied dental hand piece to improve the efficiency of sound absorption. A lot of micro-sized pores were fabricated with UV laser processing on the surface of sliced EPP substrate. From fundamental experiments, more high-performance of micro-porous EPP substrate has finally demonstrated for sound-absorbing structure of the micro muffler inside dental hand piece, which actually has the excellent potential to apply a lot of potable machine.

• Journal of Coastal Disaster Prevention
• /
• v.5 no.4
• /
• pp.163-172
• /
• 2018

To propose the design technique and the execution manual of the LWFS(Lightweight Foamed Soil) method using dredged soil, the operation system for the test-bed integrated management, and to establish an amendment for the domestic quantity per unit and specifications, and a strategy for its internationalization. In order to utilize the dredged soil from the coastal area as a construction material, we constructed the embankment with LWFS on soft ground and monitored its behavior. As a result, it can be expected that the use of LWFS as an embankment material on the soft ground can improve the economic efficiency by reducing the depth and period of soil improvement as well as the uses of nearby dredged soil. To verify the utilization of the dredged soil as a material for light-weighted roadbed, soft ground and foundation ground, and surface processing, perform an experimental construction for practical structures and analyze the behavior. It is expected to be able to improve the soft ground with dredged soil and develop technique codes and manuals of the dredged soil reclamation by constructing a test-bed in the same size of the fields, and establish the criteria and manual of effective dredged soil reclamation for practical use. The application technology of the dredged soil reclamation during harbor constructions and dredged soil reclamation constructions can be reflected during the working design stage. By using the materials immediately that occur from the reclamation during harbor and background land developments, the development time will decrease and an increase of economic feasibility will happen. It is expected to be able to apply the improved soil at dredged soil reclamation, harbor and shore protection construction, dredged soil purification projects etc. Future-work for develop the design criteria and guideline for the technology of field application of dredged soil reclamation is that review the proposed test-bed sites, consult with the institutions relevant with the test-bed, establish the space planning of the test-bed, licensing from the institutions relevant with the test-bed, select a test-bed for the dredged soil disposal area.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.5
• /
• pp.195-200
• /
• 2021

Due to the development of video sharing platforms, the amount of video uploads is exploding. Such videos often include various types of music, among which cover songs are included. In order to protect the copyright of music, an algorithm to find the original song of the cover song is essential. However, it is not easy to find the original song because the cover song is a modification of the composition, speed and overall structure of the original song. So far, there is no known effective algorithm for searching the original song of the cover song. In this paper, we propose an algorithm for searching the original song of the cover song using the inflection points of the melody line. Inflection points represent the characteristic points of change in the melody sequence. The proposed algorithm compares the original song and the cover song using the sequence of inflection points for the representative phrase of the original song. Since the characteristics of the representative phrase are used, even if the cover song is a song made by modifying the overall composition of the song, the algorithm's search performance is excellent. Also, since the proposed algorithm uses only the features of the inflection point sequence, the memory usage is very low. The efficiency of the algorithm was verified through performance evaluation.

• Korean Journal of Materials Research
• /
• v.29 no.11
• /
• pp.720-726
• /
• 2019

Recently, research on cost reduction and efficiency improvement of crystalline silicon(c-Si) photovoltaic(PV) module has been conducted. In order to reduce costs, the thickness of solar cell wafers is becoming thinner. If the thickness of the wafer is reduced, cracking of wafer may occur in high temperature processes during the c-Si PV module manufacturing process. To solve this problem, a low temperature process has been proposed. Conductive paste(CP) is used for low temperature processing; it contains Sn57.6Bi0.4Ag component and can be electrically combined with solar cells and ribbons at a melting point of $150^{\circ}C$. Use of CP in the PV module manufacturing process can minimize cracks of solar cells. When CP is applied to solar cells, the output varies with the amount of CP, and so the optimum amount of CP must be found. In this paper, in order to find the optimal CP application amount, we manufactured several c-Si PV modules with different CP amounts. The amount control of CP is fixed at air pressure (500 kPa) and nozzle diameter 22G(outer diameter 0.72Ø, inner 0.42Ø) of dispenser; only speed is controlled. The c-Si PV module output is measured to analyze the difference according to the amount of CP and analyzed by optical microscope and Alpha-step. As the result, the optimum amount of CP is 0.452 ~ 0.544 g on solar cells.

• Journal of Convergence for Information Technology
• /
• v.10 no.12
• /
• pp.15-21
• /
• 2020

Although IoT systems are used in a variety of heterogeneous environments as cloud environments develop, all IoT devices are not provided with reliable protocols and services. This paper proposes an IoT data management technique that can extend the IoT cloud environment to an n-layer multi-level structure so that information collected from different heterogeneous IoT devices can be efficiently sorted and processed. The proposed technique aims to classify and process IoT information by transmitting routing information and weight information through wireless data link data collected from heterogeneous IoT devices. The proposed technique not only delivers information classified from IoT devices to the corresponding routing path but also improves the efficiency of IoT data processing by assigning priority according to weight information. The IoT devices used in the proposed technique use each other's reliable protocols, and queries for other IoT devices locally through a local cloud composed of hierarchical structures have features that ensure scalability because they maintain a certain cost.y channels of IoT information in order to make the most of the multiple antenna technology.

• Journal of Nutrition and Health
• /
• v.54 no.1
• /
• pp.116-128
• /
• 2021

Purpose: The extraction conditions for bioactive components from peanut shells, which is a byproduct of peanut processing, were optimized to enhance the total phenolic content (TPC, Y1), total flavonoid content (TFC, Y2), and 2,2-diphenyl-1-picrylhydrazyl radical scavenging activity (RSA, Y3). In addition, this study evaluated the anti-obesity effect of peanut shell extract. Methods: Optimization of ultrasonic-assisted extraction (UAE) was performed using a response surface methodology. The independent variables applied for extraction were time (X1: 5.0-55.0), temperature (X2: 26.0-94.0), and ethanol concentration (X3: 0.0%-99.5%). Quadratic regression models were derived based on the results of 17 experimental sets, and an analysis of the variance was performed to verify its accuracy and precision of the regression equations. Results: When evaluating the effects of independent variables on responses using statistically-based optimization, the independent variable with the most significant effect on the TPC, TFC, and RSA was the ethanol concentration (p = 0.0008). The optimal extraction conditions to satisfy all three responses were 35.8 minutes, 82.7℃, and 96.0% ethanol. Under these conditions, the inhibitory activities of α-glucosidase and pancreatic lipase by the extract were 86.4% and 78.5%, respectively. Conclusion: In this study, UAE showed superior extraction efficiency compared to conventional hot-water extraction in the extraction of polyphenols and bioactive materials. In addition, α-glucosidase and pancreatic lipase inhibitory effects were identified, suggesting that peanut shells can be used as effective antioxidants and anti-obesity agents in functional foods and medicines.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.3
• /
• pp.141-148
• /
• 2021

This paper presents a method to detect zebra-crossing using deep learning which combines SegNet and ResNet. For the blind, a safe crossing system is important to know exactly where the zebra-crossings are. Zebra-crossing detection by deep learning can be a good solution to this problem and robotic vision-based assistive technologies sprung up over the past few years, which focused on specific scene objects using monocular detectors. These traditional methods have achieved significant results with relatively long processing times, and enhanced the zebra-crossing perception to a large extent. However, running all detectors jointly incurs a long latency and becomes computationally prohibitive on wearable embedded systems. In this paper, we propose a model for fast and stable segmentation of zebra-crossing from captured images. The model is improved based on a combination of SegNet and ResNet and consists of three steps. First, the input image is subsampled to extract image features and the convolutional neural network of ResNet is modified to make it the new encoder. Second, through the SegNet original up-sampling network, the abstract features are restored to the original image size. Finally, the method classifies all pixels and calculates the accuracy of each pixel. The experimental results prove the efficiency of the modified semantic segmentation algorithm with a relatively high computing speed.

• The Journal of Bigdata
• /
• v.5 no.2
• /
• pp.53-67
• /
• 2020

This study employees a supervised learning prediction model to detect nonconformity in advance of processed food manufacturing and processing businesses. The study was conducted according to the standard procedure of machine learning, such as definition of objective function, data preprocessing and feature engineering and model selection and evaluation. The dependent variable was set as the number of supervised inspection detections over the past five years from 2014 to 2018, and the objective function was to maximize the probability of detecting the nonconforming companies. The data was preprocessed by reflecting not only basic attributes such as revenues, operating duration, number of employees, but also the inspections track records and extraneous climate data. After applying the feature variable extraction method, the machine learning algorithm was applied to the data by deriving the company's risk, item risk, environmental risk, and past violation history as feature variables that affect the determination of nonconformity. The f1-score of the decision tree, one of ensemble models, was much higher than those of other models. Based on the results of this study, it is expected that the official food control for food safety management will be enhanced and geared into the data-evidence based management as well as scientific administrative system.

• Journal of the Society of Cosmetic Scientists of Korea
• /
• v.47 no.2
• /
• pp.163-170
• /
• 2021

This study was conducted to create a technology to remove acne bacteria with human-friendly materials. First, the Cutibacterium acnes (C. acnes) were adsorbed to the mica disc to grow, and then the biofilm was checked through an atomic microscope to see if the biofilm had grown. Based on the topographic image, the shape changed round, the size was 17% longer on average, and the phase value of the resonance frequency separating materials was observed as a single value, the biofilm grown by covering the extracellular polymeric substrate (EPS). As a result of processing 50 mM of amino acids in the matured biofilm, the concentration of C. acnes decreased when valine, serine, arginine and leucine were treated. Scanning with nanoindentation and AFM contact modes confirmed that the hardness of biofilms treated with Valine (Val) increased. This indicates that an AFM tip measured cell which may have more solidity than that of EPS. The experiment of fluorescent tagged to EPS displays an existence of EPS at the condition of 10 mM Val, but an inhibition of growth of EPS at the 50 mM Val. Number of C. acnes was also reduced above 10 mM of Val. Weak adhesion of biofilm generated from an inhibition of EPS formation seems to induce decrease of C. acnes. Accordingly, we elucidated that Val has an efficiency which eliminates C. acnes by approach of an inhibition of EPS.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.141-148
• /
• 2022

Recently, reinforcement learning combined with deep neural network technology has achieved remarkable success in various fields such as board games such as Go and chess, computer games such as Atari and StartCraft, and robot object manipulation tasks. However, such deep reinforcement learning describes states, actions, and policies in vector representation. Therefore, the existing deep reinforcement learning has some limitations in generality and interpretability of the learned policy, and it is difficult to effectively incorporate domain knowledge into policy learning. On the other hand, dNL-RRL, a new relational reinforcement learning framework proposed to solve these problems, uses a kind of vector representation for sensor input data and lower-level motion control as in the existing deep reinforcement learning. However, for states, actions, and learned policies, It uses a relational representation with logic predicates and rules. In this paper, we present dNL-RRL-based policy learning for transportation mobile robots in a manufacturing environment. In particular, this study proposes a effective method to utilize the prior domain knowledge of human experts to improve the efficiency of relational reinforcement learning. Through various experiments, we demonstrate the performance improvement of the relational reinforcement learning by using domain knowledge as proposed in this paper.