• The Journal of the Korea Contents Association
• /
• v.21 no.2
• /
• pp.140-147
• /
• 2021

EEG(electroencephalogram) are measured to accurately determine the level of sleep in various sleep examinations. In general, measurements are more accurate as the number of sensor channels increases. EEG can interfere with sleep by attaching electrodes to the skin when measuring. It is necessary for self sleep care to select the minimum number of EEG channels that take into account both the user's discomfort and the accuracy of the measurement data. In this paper, we proposed a sleep stage analysis model based on machine learning and conducted experiments for using from one channel to four channels. We obtained estimation accuracy for sleep stage as following 82.28% for one channel, 85.77% for two channels, 80.33% for three channels and 68.87% for four channels. Although the measurement location is limited, the results of this study compare the accuracy according to the number of channels and provide information on the selection of channel numbers in the EEG sleep analysis.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.50 no.1
• /
• pp.39-45
• /
• 2022

In this paper, we propose methods to improve image-based auto-focus that can compensate for drawbacks of traditional auto-focus control. When adjusting the focus, there is a problem that the focus window cannot be set to the same position if the camera's LOS is not directed at the same location and flow or shake. To address this issue, we applied image tracking techniques to improve optimal focus localization accuracy. And also, although the same focus value should be calculated at the same focus step, but different values can be calculated by camera's fine shaking or image disturbance due to atmospheric scattering. To tackle this problem a SAFS (Stable Adjacency Frame Selection) has been proposed. As a result of this study, our proposed methodology shows more accurate than traditional methods in terms of finding best focus position.

• Journal of Convergence for Information Technology
• /
• v.12 no.1
• /
• pp.39-44
• /
• 2022

The purpose of this study is to extract the type, location, and absolute size of an object in an image using a deep learning algorithm, predict the relative distance between objects, and use this to detect contact between objects. To analyze the size ratio of objects, YOLO, a CNN-based object detection algorithm, is used. Through the YOLO algorithm, the absolute size and position of an object are extracted in the form of coordinates. The extraction result extracts the ratio between the size in the image and the actual size from the standard object-size list having the same object name and size stored in advance, and predicts the relative distance between the camera and the object in the image. Based on the predicted value, it detects whether the objects are in contact.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.57 no.4
• /
• pp.334-341
• /
• 2021

The fishery compensation by marine spatial planning such as routeing of ships and offshore wind farms is required objective data on whether fishing vessels are engaged in a target area. There has still been no research that calculated the number of fishing operation days scientifically. This study proposes a novel method for calculating the number of fishing operation days using the fishing trajectory data when investigating fishery compensation in marine spatial planning areas. It was calculated by multiplying the average reporting interval of trajectory data, the number of collected data, the status weighting factor, and the weighting factor for fishery compensation according to the location of each fishing vessel. In particular, the number of fishing operation days for the compensation of driftnet fishery was considered the daily average number of large vessels from the port and the fishery loss hours for avoiding collisions with them. The target area for applying the proposed method is the routeing area of ships of Jeju outer port. The yearly average fishing operation days were calculated from three years of data from 2017 to 2019. As a result of the study, the yearly average fishing operation days for the compensation of each fishing village fraternity varied from 0.0 to 39.0 days. The proposed method can be used for fishery compensation as an objective indicator in various marine spatial planning areas.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.7
• /
• pp.2131-2153
• /
• 2022

To achieve accurate detection of tuberculosis (TB) areas in chest radiographs, we design a chest X-ray TB area detection algorithm. The algorithm consists of two stages: the chest X-ray TB classification network (CXTCNet) and the chest X-ray TB area detection network (CXTDNet). CXTCNet is used to judge the presence or absence of TB areas in chest X-ray images, thereby excluding the influence of other lung diseases on the detection of TB areas. It can reduce false positives in the detection network and improve the accuracy of detection results. In CXTCNet, we propose a channel attention mechanism (CAM) module and combine it with DenseNet. This module enables the network to learn more spatial and channel features information about chest X-ray images, thereby improving network performance. CXTDNet is a design based on a sparse object detection algorithm (Sparse R-CNN). A group of fixed learnable proposal boxes and learnable proposal features are using for classification and location. The predictions of the algorithm are output directly without non-maximal suppression post-processing. Furthermore, we use CLAHE to reduce image noise and improve image quality for data preprocessing. Experiments on dataset TBX11K show that the accuracy of the proposed CXTCNet is up to 99.10%, which is better than most current TB classification algorithms. Finally, our proposed chest X-ray TB detection algorithm could achieve AP of 45.35% and AP50 of 74.20%. We also establish a chest X-ray TB dataset with 304 sheets. And experiments on this dataset showed that the accuracy of the diagnosis was comparable to that of radiologists. We hope that our proposed algorithm and established dataset will advance the field of TB detection.

• Journal of Industrial Convergence
• /
• v.20 no.1
• /
• pp.63-68
• /
• 2022

In schools and corporate buildings, public WiFi Access Points are installed on the ceilings of hallways. In the case of an architectural structure in which a WiFi signal enters through a steel door made of a material with high signal attenuation, Internet connection is frequently cut off or fails when the door is closed. To solve this problem, our research implements an economical and portable WiFi extender using a Raspberry Pi and an auxiliary battery. Commercially available WiFi extenders have limitations in the location where the power plug is located, and WiFi extension using the WiFi hotspot function of an Android smartphone is possible only in some high-end models. However, because the proposed device can be installed at the position where the Wi-Fi reception signal is the best inside the door, the WiFi range can be extended while minimizing the possibility of damage to the original signal. Experimental results show that it is possible to eliminate the shadows of radio waves and to provide Internet services in the office when the door is closed, to the extent that web browsing and real-time video streaming for 720p are possible.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.11
• /
• pp.4244-4274
• /
• 2021

The utilization of UAVs in various fields has led to the development of flying ad hoc network (FANET) technology. In a network environment with highly dynamic topology and frequent link changes, the traditional routing technology of FANET cannot satisfy the new communication demands. Traditional routing algorithm, based on geographic location, can "fall" into a routing hole. In view of this problem, we propose a geolocation routing protocol based on multi-agent reinforcement learning, which decreases the packet loss rate and routing cost of the routing protocol. The protocol views each node as an intelligent agent and evaluates the value of its neighbor nodes through the local information. In the value function, nodes consider information such as link quality, residual energy and queue length, which reduces the possibility of a routing hole. The protocol uses global rewards to enable individual nodes to collaborate in transmitting data. The performance of the protocol is experimentally analyzed for UAVs under extreme conditions such as topology changes and energy constraints. Simulation results show that our proposed QLGR-S protocol has advantages in performance parameters such as throughput, end-to-end delay, and energy consumption compared with the traditional GPSR protocol. QLGR-S provides more reliable connectivity for UAV networking technology, safeguards the communication requirements between UAVs, and further promotes the development of UAV technology.

• Journal of Convergence for Information Technology
• /
• v.12 no.5
• /
• pp.212-224
• /
• 2022

This study aims to provide data on attitudes towards the use of Homecare beauty devices in correlation to narcissism of women between ages 30-59. Through statistical analysis of 563 survey questions, data displayed that respondents' age, level of education, marital status, economic status, and career status showed a strong correlation with implicit narcissism, while explicit narcissism only showed a correlation with age and career status. The most popular skincare location was shown to be 'self-provided at home', and the most popular item purchased being 'galvanic devices'. Secondly, attitudes towards the use of homecare beauty devices in correlation towards implicit narcissistic respondents were only to the consideration of its use, while explicit narcissists displayed a strong correlation between the purchase of a product and the recommendation of others. While this is the first study on attitudes towards homecare beauty devices in relation to a personality-based trait like narcissism and it displayed meaningful results, a more in-depth study in the future dealing with a larger region and respondent groups of a wider age and gender group should be undertaken.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.12
• /
• pp.4439-4455
• /
• 2021

Compared with two-stage object detection algorithms, one-stage algorithms provide a better trade-off between real-time performance and accuracy. However, these methods treat the intermediate features equally, which lacks the flexibility to emphasize meaningful information for classification and location. Besides, they ignore the interaction of contextual information from different scales, which is important for medium and small objects detection. To tackle these problems, we propose an image pyramid network based on dual attention mechanism (DAIPNet), which builds an image pyramid to enrich the spatial information while emphasizing multi-scale informative features based on dual attention mechanisms for one-stage object detection. Our framework utilizes a pre-trained backbone as standard detection network, where the designed image pyramid network (IPN) is used as auxiliary network to provide complementary information. Here, the dual attention mechanism is composed of the adaptive feature fusion module (AFFM) and the progressive attention fusion module (PAFM). AFFM is designed to automatically pay attention to the feature maps with different importance from the backbone and auxiliary network, while PAFM is utilized to adaptively learn the channel attentive information in the context transfer process. Furthermore, in the IPN, we build an image pyramid to extract scale-wise features from downsampled images of different scales, where the features are further fused at different states to enrich scale-wise information and learn more comprehensive feature representations. Experimental results are shown on MS COCO dataset. Our proposed detector with a 300 × 300 input achieves superior performance of 32.6% mAP on the MS COCO test-dev compared with state-of-the-art methods.

• Journal of Convergence for Information Technology
• /
• v.11 no.5
• /
• pp.1-8
• /
• 2021

In this paper, we propose the construction of an IOT that monitors foresters' physical abnormalities in real time, performs emergency measures, and provides alarms for natural disasters or heatstroke such as a nearby forest fire or landslide. Nodes provided to foresters include 6-axis sensors, temperature sensors, GPS, and LoRa, and transmit the measured data to the network server through the gateway using LoRa communication. The network server uses 6-axis sensor data to determine whether or not a forester has any signs of abnormal body, and performs emergency measures by tracking GPS location. After analyzing the temperature data, it provides an alarm when there is a possibility of heat stroke or when a forest fire or landslide occurs in the vicinity. In this paper, it was confirmed that the real-time detection of physical abnormalities of foresters and the establishment of disaster early warning IOT is possible by analyzing the data obtained by constructing a node and a gateway and constructing a network server.