• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.5
• /
• pp.135-140
• /
• 2021

In this paper, we propose an on-line mechanism that simultaneously detects cracks and profiling pavement blocks to detect the displacement of ground surface adjacent to the excavation in the urban area. The proposed method utilizes a 2D laser to profile the information about pavement blocks including the depth and distance among them. In particular, it is designed to enable the detection of cracks and portholes at runtime. For the experiment, real data was collected through Gocator, and trainng was carried out using Faster R-CNN. The performance evaluation shows that our detection precision and recall are more than 90% and the pavement blocks are profiled at the same time. Our proposed mechanism can be used for monitoring management to quantitatively detect the level of excavation risk before a large-scale ground collapse occurs.

• Journal of the Korea Convergence Society
• /
• v.12 no.2
• /
• pp.1-6
• /
• 2021

In the modern society, traffic problems are occurring as vehicle ownership increases. In particular, the incidence of highway traffic accidents is low, but the fatality rate is high. Therefore, a technology for detecting an abnormality in a vehicle is being studied. Among them, there is a vehicle anomaly detection technology using deep learning. This detects vehicle abnormalities such as a stopped vehicle due to an accident or engine failure. However, if an abnormality occurs on the road, it is possible to quickly respond to the driver's location. In this study, we propose a deep learning-based vehicle anomaly detection using road CCTV data. The proposed method preprocesses the road CCTV data. The pre-processing uses the background extraction algorithm MOG2 to separate the background and the foreground. The foreground refers to a vehicle with displacement, and a vehicle with an abnormality on the road is judged as a background because there is no displacement. The image that the background is extracted detects an object using YOLOv4. It is determined that the vehicle is abnormal.

• Journal of the Korean Recycled Construction Resources Institute
• /
• v.10 no.3
• /
• pp.235-242
• /
• 2022

The percolation threshold of the CNT-reinforced cement paste is closely related to the optimal CNT amount to maximize the sensing ability of self-sensing concrete. However, the percolation threshold has various values depending on the cement, CNT, and water-to-cement ratio used. In this study, a percolation simulation model was proposed to predict the percolation threshold of the CNT-reinforced cement paste. The proposed model can simulate the percolation according to the amount of CNT using only the properties of CNT and cement, and for this, the concept of the number of aggregated CNT particles was used. The percolation simulation consists of forming a pre-hydrated cement paste model, random dispersion of CNTs, and percolation investigation. The simulation used CNT-reinforced cement paste with a water-cement ratio of 0.4 to 0.6, and the simulated percolation threshold point showed high accuracy with a simulation residual ratio of up to 7.5 % compared to the literature results.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.9
• /
• pp.1259-1265
• /
• 2022

Recently, AI, big data, and IoT technologies are being used in various solutions such as fire detection and gas or dangerous substance detection for safety accident prevention. According to the status of occupational accidents published by the Ministry of Employment and Labor in 2021, the accident rate, the number of injured, and the number of deaths have increased compared to 2020. In this paper, referring to the dataset construction guidelines provided by the National Intelligence Service Agency(NIA), the dataset is directly collected from the field and learned with YOLOv4 to propose a collision risk object detection system through object detection. The accuracy of the dangerous situation rule violation was 88% indoors and 92% outdoors. Through this system, it is thought that it will be possible to analyze safety accidents that occur in industrial sites in advance and use them to intelligent platforms research.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.12
• /
• pp.67-77
• /
• 2023

This study represents an innovative research conducted in the smart farm environment, developing a deep learning-based disease and pest detection model and applying it to the Intelligent Internet of Things (IoT) platform to explore new possibilities in the implementation of digital agricultural environments. The core of the research was the integration of the latest ImageNet models such as Pseudo-Labeling, RegNet, EfficientNet, and preprocessing methods to detect various diseases and pests in complex agricultural environments with high accuracy. To this end, ensemble learning techniques were applied to maximize the accuracy and stability of the model, and the model was evaluated using various performance indicators such as mean Average Precision (mAP), precision, recall, accuracy, and box loss. Additionally, the SHAP framework was utilized to gain a deeper understanding of the model's prediction criteria, making the decision-making process more transparent. This analysis provided significant insights into how the model considers various variables to detect diseases and pests.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.26 no.5
• /
• pp.506-513
• /
• 2015

In this paper, we proposed a novel ethanol concentration sensor using microfluidic metamaterial absorber. The metamaterial absorber comprises a split-ring-cross resonator(SRCR) and a microfluidic channel. The SRCR can generate LC resonance that is very sensitive to changes in the effective dielectric constant around the capacitive gap. In addition, microfluidic channels can change the effective dielectric constant of the dielectric substrate by using an infinitesimal quantity of a liquid on the order of microliters. The proposed absorber can detect the electrical properties of different concentration of ethanol. The performance of the proposed absorber is demonstrated using the absorption measurements of a fabricated prototype sample with waveguides. In addition, the simulated results and measurement results show good agreement.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.1B
• /
• pp.8-19
• /
• 2010

In this paper, we propose the NCP (Network Control Platform)-based defense mechanism against DDoS (Distributed Denial of Service) attacks in order to guarantee the transmission of normal traffic and prevent the flood of abnormal traffic. We also define defense modules, the threshold and packet drop-rate used for the response against DDoS attacks. NCP analyzes whether DDoS attacks are occurred or not based on the flow and queue information collected from SR (Source Router) and VR (Victim Router). Attack packets are dopped according to drop rate decided from NCP. The performance is simulated using OPNET and evaluated in terms of the queue size of both SR and VR, the transmitted volumes of legitimate and attack packets at SR.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.11 no.6
• /
• pp.561-566
• /
• 2016

A subway platform is equipped with screen doors in oder to avoid accidents of passengers, where Area sensors are installed for detecting obstacles in the screen doors. However, there exist frequent operating errors in screen doors due to dusts, sunlight, snow, and bugs. It is required to develope a detection device which reduces errors and elaborates detection function. In this paper, we compared the detection rates of the Area sensor the 3D sensor using CCTV-based image data with installing sensors at the screen door in Munyang station Daegu, where 3D sensor is applied with the space division multiple detection algorithms. It is measured that the detection rate of 3D sensor and Area sensor is approximately 89.61% and 78.88%, respectively. The results confirmed that 3D senor has higher detection rate compared with Area sensor with the rate of 6.87~9.79%, and 3D sensor has benefit in the aspect of installation fee.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.262-265
• /
• 2021

The purpose of this development is to prevent accidents in the telephone poles caused by bird nests in advance. It is a sensor node installed on a telephone pole to recognize a bird's nest. This is to remove the bird before it builds a nest and lays eggs. It is in the system that recognizes the bird nest by the change of the distance when the sensor is first installed and the distance value measured thereafter. In this paper, we have designed and tested infrared rays with concrete, iron plate, wood, and plastic bag are targeted. This is an object that can be detected within a telephone pole was tested. The value of the spectrum detected by the spectral reflectance was obtained through a photodiode. Through the standard deviation graph of these values, it became possible to predict the target of the object and measure the distance. As a result of this experiment, target information (concrete, iron plate, wood, plastic bag) about dangerous substances in the telephone pole was acquired through the infrared sensor. Through this, it is expected that it will contribute to the establishment of a safe power grid and a coexistence environment with nature through power grid monitoring.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.7
• /
• pp.67-77
• /
• 2016

In this paper, we propose an algorithm for open/closed eye detection based on modified Hausdorff distance. The proposed algorithm consists of two parts, face detection and open/closed eye detection parts. To detect faces in an image, MCT (Modified Census Transform) is employed based on characteristics of the local structure which uses relative pixel values in the area with fixed size. Then, the coordinates of eyes are found and open/closed eyes are detected using MHD (Modified Hausdorff Distance) in the detected face region. Firstly, face detection process creates an MCT image in terms of various face images and extract criteria features by PCA(Principle Component Analysis) on offline. After extraction of criteria features, it detects a face region via the process which compares features newly extracted from the input face image and criteria features by using Euclidean distance. Afterward, the process finds out the coordinates of eyes and detects open/closed eye using template matching based on MHD in each eye region. In performance evaluation, the proposed algorithm achieved 94.04% accuracy in average for open/closed eye detection in terms of test video sequences of gray scale with 30FPS/$320{\times}180$ resolution.