• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.2
• /
• pp.187-192
• /
• 2021

This study proposes a system for take-off in a forest or similar complex environment using an object detector. In the simulator, a raspberry pi is mounted on a quadcopter with a length of 550mm between motors on a diagonal line, and the experiment is conducted based on edge computing. As for the images to be used for learning, about 150 images of 640⁎480 size were obtained by selecting three points inside Kunsan University, and then converting them to black and white, and pre-processing the binarization by placing a boundary value of 127. After that, we trained the SSD_Inception model. In the simulation, as a result of the experiment of taking off the drone through the model trained with the verification image as an input, a trajectory similar to the takeoff was drawn using the label.

• Wind and Structures
• /
• v.34 no.1
• /
• pp.15-27
• /
• 2022

This work comes within the framework of the "Benchmark on the Aerodynamics of a Rectangular Cylinder" that investigates a rectangular cylinder of length-to-depth ratio equal to 5. The present study reports and discusses velocity fields acquired using planar Particle Image Velocitmetry for several angles of attack and Reynolds numbers. In particular, for a cylinder depth-based Reynolds number of 2 × 10⁴ and zero incidence angle, the flow features along the lateral (parallel to the freestream) upper and lower surfaces of the cylinder are reported. Using first and second order statistics of the velocity field, the main flow features are discussed, especially the size and location of the time-averaged flow structures and the distribution of the Reynolds stresses. The variation of the flow features with the incidence is also studied considering angles of attack up to 6°. It is shown that the time-averaged flow is fully detached for incidence higher than 2°. For an angle of attack of 0°, the effects of the Reynolds number varying between 5 × 10³ and 2 × 10⁴ are investigated looking at flow statistics. It is shown that the time-averaged location of the reattachment point and the shape and position of the time-averaged main vortex are mostly constant with the Reynolds number. However, the size of the inner region located below the time-averaged shear layer and just downstream the leading edge corner appears to be strongly dependent on the Reynolds number.

• Journal of The Korean Astronomical Society
• /
• v.54 no.4
• /
• pp.113-119
• /
• 2021

It is difficult for observers to conduct an optical alignment at an observatory without the assistance of an optical engineer if optomechanical parts are to be replaced at night. We present a practical tilt correction method to obtain the optimal optical alignment condition using the symmetricity of optical aberrations of a wide-field on-axis telescope at night. We conducted coarse tilt correction by visually examining the symmetry of two representative star shapes obtained at two guide chips facing each other, such as east-west or north-south pairs. After coarse correction, we observed four sets of small stamp images using four guide cameras located at each cardinal position by changing the focus positions in 10-㎛ increments and passing through the optimum focus position in the range of ±200 ㎛. The standard deviation of each image, as a function of the focus position, was fitted with a second-order polynomial function to derive the optimal focus position at each cardinal edge. We derived the tilt angles from the slopes converted by the distance and the focus position difference between two paired guide chip combinations such as east-west and north-south. We used this method to collimate the on-axis wide-field telescope KMTNet in Chile after replacing two old focus actuators. The total optical alignment time was less than 30 min. Our method is practical and straightforward for maintaining the optical performance of wide-field telescopes such as KMTNet.

• Structural Engineering and Mechanics
• /
• v.80 no.5
• /
• pp.503-522
• /
• 2021

This paper investigates experimentally and numerically the influence of drilling process on the mechanical and thermomechanical behaviors of woven glass fiber reinforced polymer (GFRP) composite plate. Through the experimental analysis, a CNC machine with cemented carbide drill (point angles 𝜙=118° and 6 mm diameter) was used to drill a woven GFRP laminated squared plate with a length of 36.6 mm and different thicknesses. A produced temperature during drilling "heat affected zone (HAZ)" was measured by two different procedures using thermal IR camera and thermocouples. A thrust force and cutting torque were measured by a Kistler 9272 dynamometer. The delamination factors were evaluated by the image processing technique. Finite element model (FEM) has been developed by using LS-Dyna to simulate the drilling processing and validate the thrust force and torque with those obtained by experimental technique. It is found that, the present finite element model has the capability to predict the force and torque efficiently at various drilling conditions. Numerical parametric analysis is presented to illustrate the influences of the speeding up, coefficient of friction, element type, and mass scaling effects on the calculated thrust force, torque and calculation's cost. It is found that, the cutting time can be adjusted by drilling parameters (feed, speed, and specimen thickness) to control the induced temperature and thus, the force, torque and delamination factor in drilling GFRP composites. The delamination of woven GFRP is accompanied with edge chipping, spalling, and uncut fibers.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.353-360
• /
• 2022

Due to the dense and complicated working environment, the construction industry is susceptible to many accidents. Worker's fall is a severe problem at the construction site, including falling into holes or openings because of the inadequate coverings as per the safety rules. During the construction or demolition of a building, openings and holes are formed in the floors and roofs. Many workers neglect to cover openings for ease of work while being aware of the risks of holes, openings, and gaps at heights. However, there are safety rules for worker safety; the holes and openings must be covered to prevent falls. The safety inspector typically examines it by visiting the construction site, which is time-consuming and requires safety manager efforts. Therefore, this study presented a worker-driven approach (the worker is involved in the reporting process) to facilitate safety managers by developing integrated computer vision and inertia sensors-based mobile applications to identify openings. The TensorFlow framework is used to design Convolutional Neural Network (CNN); the designed CNN is trained on a custom dataset for binary class openings and covered and deployed on an android smartphone. When an application captures an image, the device also extracts the accelerometer values to determine the inclination in parallel with the classification task of the device to predict the final output as floor (openings/ covered), wall (openings/covered), and roof (openings / covered). The proposed worker-driven approach will be extended with other case scenarios at the construction site.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.01a
• /
• pp.143-144
• /
• 2020

4차 산업 시대가 도래하여 빅데이터와 딥러닝 기술은 다양한 분야에서 아주 중요한 기술로 자리 잡고 있으며, 현재 세계 여러 분야에서 이 기술들을 이용하여 일상, 산업 분야에 적용을 시키고자 한다. 국내에서는 스마트 팩토리, 스마트 시티와 같은 분야에 적용하고 있다. 본 논문에서는 스마트 시티에 적용할 수 있는 횡단보도 상황을 인지하여 교통제어에 활용할 수 있는 빅데이터를 생산하거나 효율적인 교통제어에 활용할 수 있도록 Nvidia Jetson TX2와 실시간 객체 감지 기술인 YOLO v3를 이용하여 횡단보도용 상황 인식을 위한 영상인식 장치를 개발하였다. 제안하는 기술들을 이용하여 스마트시티 구축에 활용할 수 있고, 실시간으로 추가적으로 필요한 객체를 감지하여 확장이 용이한 장점이 있다. 또한 구현에서 효율성을 높이기 위하여 에지 컴퓨팅, 스페이스 디텍션과 같은 기술들을 활용하였다.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.3D
• /
• pp.535-540
• /
• 2006

This paper presented an effective strategy to extract the buildings and to reconstruct 3-D buildings using high-resolution multispectral stereo satellite images. Proposed scheme contained three major steps: building enhancement and segmentation using both BDT (Background Discriminant Transformation) and ISODATA algorithm, conjugate building identification using the object matching with Hausdorff distance and color indexing, and 3-D building reconstruction using photogrammetric techniques. IKONOS multispectral stereo images were used to evaluate the scheme. As a result, the BDT technique was verified as an effective tool for enhancing building areas since BDT suppressed the dominance of background to enhance the building as a non-background. In building recognition, color information itself was not enough to identify the conjugate building pairs since most buildings are composed of similar materials such as concrete. When both Hausdorff distance for edge information and color indexing for color information were combined, most segmented buildings in the stereo images were correctly identified. Finally, 3-D building models were successfully generated using the space intersection by the forward RFM (Rational Function Model).

• Journal of the Korean Applied Science and Technology
• /
• v.40 no.5
• /
• pp.1043-1054
• /
• 2023

Amidst the rapid growth of the vegan cosmetics market, consumer orientation towards environmental and ethical values has been intensifying. However, research on this subject remains limited. This study delves into the relationship between consumer value perception, environmental motivations, and perceived barriers influencing the purchase intentions of vegan cosmetics. Conducting a PLS-SEM analysis on a sample of 300 women with experience using vegan cosmetics, it was discerned that monetary value, social value, brand value, emotional value, quality value, and environmental knowledge play significant roles in influencing purchase intentions. The moderating effect analysis highlighted image barriers and value barriers as crucial factors. Through Importance-Performance Map Analysis, emotional value emerged as a pivotal element in strategizing to strengthen the purchasing intentions for vegan cosmetics. This research contributes both theoretically and practically to enhancing the competitive edge of the vegan cosmetics market and promoting sustainable consumption behavior.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.9
• /
• pp.2605-2625
• /
• 2024

Siamese-based segmentation and tracking algorithms improve accuracy and stability for video object segmentation and tracking tasks simultaneously. Although effective, variability in target appearance and background clutter can still affect segmentation accuracy and further influence the performance of tracking. In this paper, we present a memory propagation-based target-aware and mask-attention decision network for robust object segmentation and tracking. Firstly, a mask propagation-based attention module (MPAM) is constructed to explore the inherent correlation among image frames, which can mine mask information of the historical frames. By retrieving a memory bank (MB) that stores features and binary masks of historical frames, target attention maps are generated to highlight the target region on backbone features, thus suppressing the adverse effects of background clutter. Secondly, an attention refinement pathway (ARP) is designed to further refine the segmentation profile in the process of mask generation. A lightweight attention mechanism is introduced to calculate the weight of low-level features, paying more attention to low-level features sensitive to edge detail so as to obtain segmentation results. Finally, a mask fusion mechanism (MFM) is proposed to enhance the accuracy of the mask. By utilizing a mask quality assessment decision network, the corresponding quality scores of the "initial mask" and the "previous mask" can be obtained adaptively, thus achieving the assignment of weights and the fusion of masks. Therefore, the final mask enjoys higher accuracy and stability. Experimental results on multiple benchmarks demonstrate that our algorithm performs outstanding performance in a variety of challenging tracking tasks.

• Journal of Korean Society of Archives and Records Management
• /
• v.24 no.3
• /
• pp.145-156
• /
• 2024

This study explores the integration of cutting-edge technology with records management, aiming to create new value not only in work processes but also in record information services. The research focuses on the case of constructing an "AI-based cultural heritage research record learning data and search system," carried out by the National Research Institute of Cultural Heritage (NRICH) Archives, and analyzes user satisfaction results. "Discovering Cultural Heritage with ZOOM" is a system designed to proactively predict research data demand by constructing big data (learning data) from images (675,338 items) contained in 1,421 volumes of publications in the cultural heritage field, spanning from 1973 to the present, and simultaneously presenting 50 similar images. This initiative aims to foster change and development in the field of records management and cultural heritage in response to the Fourth Industrial Revolution's advanced technologies. It is expected to provide valuable information to researchers, practitioners, and the general public alike.