• Journal of the Korea Society of Computer and Information
• /
• v.27 no.3
• /
• pp.63-69
• /
• 2022

In this paper, we propose a new GPU-based framework for quickly calculating adaptive and continuous SDF(Signed distance fields), and examine cases related to rendering/collision processing using them. The quadtree constructed from the triangle mesh is transferred to the GPU memory, and the Euclidean distance to the triangle is processed in parallel for each thread by using it to find the shortest continuous distance without discontinuity in the adaptive grid space. In this process, it is shown through experiments that the cut-off view of the adaptive distance field, the distance value inquiry at a specific location, real-time raytracing, and collision handling can be performed quickly and efficiently. Using the proposed method, the adaptive sign distance field can be calculated quickly in about 1 second even on a high polygon mesh, so it is a method that can be fully utilized not only for rigid bodies but also for deformable bodies. It shows the stability of the algorithm through various experimental results whether it can accurately sample and represent distance values in various models.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.21 no.2
• /
• pp.132-151
• /
• 2022

Vehicle localization is one of the core technologies for autonomous driving. Image-based localization provides location information efficiently, and various related studies have been conducted. However, the image-based localization methods using feature points or lane information has a limitation that positioning accuracy may be greatly affected by road and driving environments. In this study, we propose a line segment matching framework for accurate vehicle localization. The proposed framework consists of four steps: line segment extraction, merging, overlap area detection, and MSLD-based segment matching. The proposed framework stably performed line segment matching at a sufficient level for vehicle positioning regardless of vehicle speed, driving method, and surrounding environment.

• 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.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.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.4
• /
• pp.709-718
• /
• 2021

In this paper, it can be seen that the number of abandoned dogs increases every year through the statistics on the occurrence of abandoned animals. With the goal of reducing the number of stray dogs, a smart pet dog shoulder strap is implemented based on a real-time location tracking system using the ESP32 module and GPS sensor. It is an ESP32 module with a built-in Bluetooth module. It is input to the MCU using various sensors, and finally outputs to a smart-phone application, and communicates through the built-in blue-tooth module. In addition, it uses Neopixels to compensate the weaknesses at night through LED light emission, and automatically sets the warning distance to design a music playback system through the LED flashing effect and MP3 module. In addition, a smart pet dog shoulder strap was designed to help pet dog health care by measuring the moving distance according to the amount of activity through the gyro sensor.

• Journal of Forest and Environmental Science
• /
• v.38 no.4
• /
• pp.239-248
• /
• 2022

Forestry crane work in a forest harvester or forwarder is regarded as one of most hard work requiring a very high level of operation skill. The operator must handle two or more multi-axes joysticks simultaneously to control the multiple manipulators for maneuvering the crane-tip to its intended location. This study has been carried out to develop a crane-tip controller which can intuitively maneuver the crane-tip, resulting in improving the productivity by decreasing the technical difficulty of control as well as reducing the workload. The crane-tip controller consists of a single 2-axis joystick and a control algorithm run on microcontroller. Lab-scale forestry crane was constructed using electric cylinders. The crane-tip control algorithm has the crane-tip follow the waypoints generated on the given path considering the dead band region using LBO (Lateral Boundary Offset). A speed control gain to change the speed of relevant cylinders relatively is applied as well. By the P (Proportional) control within the control interval of 20 msec, the average error of crane-tip control on the predefined straight path turned out to be 14.5 mm in all directions. When the joystick is used the waypoints are generated in real time by the direction signal from the joystick. In this case, the average error of path control was 12.4 mm for straight up, straight forward and straight down movements successively at a certain constant speed setting. In the slant movement of crane-tip by controlling two axes of joystick simultaneously, the movement of crane-tip was controlled in the average error of 15.9 mm when the crane-tip is moved up and down while moving toward forward direction. It concluded that the crane-tip control was possible using the control algorithm developed in this study.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.12
• /
• pp.3836-3854
• /
• 2022

The student group division is benefit for universities to do the student management based on the group profile. With the widespread use of student smart cards on campus, especially where students living in campus residence halls, students' daily activities on campus are recorded with information such as smart card swiping time and location. Therefore, it is feasible to depict the students with the daily activity data and accordingly group students based on objective measuring from their campus behavior with some regular student attributions collected in the management system. However, it is challenge in feature representation due to diverse forms of the student data. To effectively and comprehensively represent students' behaviors for further student group division, we proposed to adopt activity data from student smart cards and student attributes as input data with taking account of activity and attribution relationship types from different perspective. Specially, we propose a novel student group division method based on a multi-view student attribute heterogeneous information network (MSA-HIN). The network nodes in our proposed MSA-HIN represent students with their multi-dimensional attribute information. Meanwhile, the edges are constructed to characterize student different relationships, such as co-major, co-occurrence, and co-borrowing books. Based on the MSA-HIN, embedded representations of students are learned and a deep graph cluster algorithm is applied to divide students into groups. Comparative experiments have been done on a real-life campus dataset collected from a university. The experimental results demonstrate that our method can effectively reveal the variability of student attributes and relationships and accordingly achieves the best clustering results for group division.

• Journal of Microbiology and Biotechnology
• /
• v.32 no.10
• /
• pp.1344-1354
• /
• 2022

Laryngeal cancer is one of the highest incidence, most prevalently diagnosed head and neck cancers, making it critically necessary to probe effective targets for laryngeal cancer treatment. Here, real-time quantitative reverse transcription PCR (qRT-PCR) and western blot analysis were used to detect gene expression levels in laryngeal cancer cell lines. Fluorescence in situ hybridization (FISH) and subcellular fractionation assays were used to detect the subcellular location. Functional assays encompassing Cell Counting Kit-8 (CCK-8), 5-ethynyl-2'-deoxyuridine (EdU), transwell and wound healing assays were performed to examine the effects of target genes on cell proliferation and migration in laryngeal cancer. The in vivo effects were proved by animal experiments. RNA-binding protein immunoprecipitation (RIP), RNA pulldown and luciferase reporter assays were used to investigate the underlying regulatory mechanisms. The results showed that KIF26B antisense RNA 1 (KIF26B-AS1) propels cell proliferation and migration in laryngeal cancer and regulates the toll-like receptor 4 (TLR4) signaling pathway. KIF26B-AS1 also recruits FUS to stabilize TLR4 mRNA, consequently activating the TLR4 signaling pathway. Furthermore, KIF26B-AS1 plays an oncogenic role in laryngeal cancer via upregulating TLR4 expression as well as the FUS/TLR4 pathway axis, findings which offer novel insight for targeted therapies in the treatment of laryngeal cancer patients.

• Corrosion Science and Technology
• /
• v.22 no.1
• /
• pp.44-54
• /
• 2023

The rupture disk corrosion test (RDCT) method was recently developed to evaluate stress corrosion cracking (SCC) and was found to have great potential for the real-time detection of SCC initiation in a high temperature and pressure environment, simulating the primary water coolant of pressurized water reactors. However, it is difficult to directly measure the stress applied to a disk specimen, which is an essential factor in SCC initiation. In this work, finite element analysis (FEA) was performed using ABAQUS^TM to calculate the stress and deformation of a disk specimen. To determine the best mesh design for a thin disk specimen, hexahedron, hex-dominated, and tetrahedron models were used in FEA. All models revealed similar dome-shaped deformation behavior of the disk specimen. However, there was a considerable difference in stress distribution in the disk specimens. In the hex-dominated model, the applied stress was calculated to be the maximum at the dome center, whereas the stress was calculated to be the maximum at the dome edge in the hexahedron and tetrahedron models. From a comparison of the FEA results with deformation behavior and SCC location on the disk specimen after RDCT, the most proper FE model was found to be the tetrahedron model.

• International conference on construction engineering and project management
• /
• 2009.05a
• /
• pp.250-256
• /
• 2009

This paper presents a robotic tower-crane system using encoder and gyroscope sensors as path tracking devices. Tower crane work is often associated with falling accidents and industrial disasters. Such problems often incur a loss of time and money for the contractor. For this reason, many studies have been done on an automatic tower crane. As a part of 5-year 23-million-dollar research project in Korea, we are developing a robotic tower crane which aims to improve the safety level and productivity. We selected a luffing tower crane, which is commonly used in urban construction projects today, as a platform for the robotic tower crane system. This system comprises two modules: the automated path planning module and the path tracking module. The automated path planning system uses the 3D Cartesian coordinates. When the robotic tower crane lifts construction material, the algorithm creates a line, which represents a lifting path, in virtual space. This algorithm seeks and generates the best route to lift construction material while avoiding known obstacles from real construction site. The path tracking system detects the location of a lifted material in terms of the 3D coordinate values using various types of sensors including adopts encoder and gyroscope sensors. We are testing various sensors as a candidate for the path tracking device. This specific study focuses on how to employ encoder and gyroscope sensors in the robotic crane These sensors measure a movement and rotary motion of the robotic tower crane. Finally, the movement of the robotic tower crane is displayed in a virtual space that synthesizes the data from two modules: the automatically planned path and the tracked paths. We are currently field-testing the feasibility of the proposed system using an actual tower crane. In the next step, the robotic tower crane will be applied to actual construction sites with a following analysis of the crane's productivity in order to ascertain its economic efficiency.