• Journal of the Korea Society of Computer and Information
• /
• v.27 no.2
• /
• pp.1-8
• /
• 2022

In this paper, we design and implement a GPU-based parallel algorithm to effectively solve the traveling salesman problem through an ant color system. The repetition process of generating hundreds or thousands of tours simultaneously in TSP utilizes GPU's task-level parallelism, and the update process of pheromone trails data actively exploits data parallelism by 32x32 thread blocks. In particular, through simultaneous memory access of multiple threads, the coalesced accesses on continuous memory addresses and concurrent accesses on shared memory are supported. This experiment used 127 to 1002 city data provided by TSPLIB, and compared the performance of sequential and parallel algorithms by using Intel Core i9-9900K CPU and Nvidia Titan RTX system. Performance improvement by GPU parallelization shows speedup of about 10.13 to 11.37 times.

• Journal of Information Technology Services
• /
• v.21 no.2
• /
• pp.27-41
• /
• 2022

In a fiercely competitive environment, quality is a key factor that enables dashboard camera makers to maintain their competitive advantage. Quality affects consumer satisfaction, brand loyalty, and firm performance. Therefore, to remain competitive, it is important that manufacturers maintain product quality that meets consumer expectations. To this end, it is necessary to investigate customer preferences and product performance in terms of product quality and to properly allocate resources to improve the quality level such that the firm can maintain a competitive advantage. In this paper, we proposed the various ways in which manufacturing firms can determine which quality dimensions need improvement in order to secure competitiveness. To this end, we analyzed a case study of Urive to develop a quality improvement strategy through importance performance competitor analysis (IPCA). Urive's IPCA results showed that 14 quality dimensions, namely performance, size, price, ease of use, country of origin, manufacturer, brand, product certificate, warranty, distribution channel, market share, reliability, durability, and conformance, were not absolutely competitive compared with those of Mando, Inavi, and Finevu. In terms of color, Urive had an absolute competitive advantage over Mando, but not Inavi and Finevu. Urive's appearance was more competitive than Mando's, but not Inavi's and Finevu's. In terms of advertisement and serviceability, Urive was absolutely less competitive than Mando and Inavi, but had a competitive advantage over Finevu. Therefore, it is necessary to put resources and time as the first priority for performance, reliability, and durability, which have a large performance difference in common among the three brands. The quality dimensions in which resources and time need to be put in second place are price and ease of use, which have a large performance difference in common among the two brands.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.10
• /
• pp.1462-1468
• /
• 2022

Among the data used for the diagnosis of calf disease, feces play an important role in disease diagnosis. In the image of calf feces, the health status can be known by the shape, color, and texture. For the fecal image that can identify the health status, data of 207 normal calves and 158 calves with diarrhea were pre-processed according to fecal status and used. In this paper, images of fecal variables are detected among the collected calf data and images are trained by applying GLCM-CNN, which combines the properties of CNN and GLCM, on a dataset containing disease symptoms using convolutional network technology. There was a significant difference between CNN's 89.9% accuracy and GLCM-CNN, which showed 91.7% accuracy, and GLCM-CNN showed a high accuracy of 1.8%.

• Journal of Broadcast Engineering
• /
• v.27 no.5
• /
• pp.638-653
• /
• 2022

In this paper, we propose the new method of creating 3D virtual-human reflecting the pattern of clothes worn by the person in the high-resolution whole body front image and the body shape data about the person. To get the pattern of clothes, we proceed Instance Segmentation and clothes parsing using Cascade Mask R-CNN. After, we use Pix2Pix to blur the boundaries and estimate the background color and can get UV-Map of 3D clothes mesh proceeding UV-Map base warping. Also, we get the body shape data using SMPL-X and deform the original clothes and body mesh. With UV-Map of clothes and deformed clothes and body mesh, user finally can see the animation of 3D virtual-human reflecting user's appearance by rendering with the state-of-the game engine, i.e. Unreal Engine.

• Journal of Aerospace System Engineering
• /
• v.16 no.5
• /
• pp.70-77
• /
• 2022

This paper proposes database generation and management system for small-pixelized airborne target recognition. The proposed system has five main features: 1) image extraction from in-flight test video frames, 2) automatic image archiving, 3) image data labeling and Meta data annotation, 4) virtual image data generation based on color channel convert conversion and seamless cloning and 5) HOG/LBP-based tiny-pixelized target augmented image data. The proposed framework is Python-based PyQt5 and has an interface that includes OpenCV. Using video files collected from flight tests, an image dataset for airborne target recognition on generates by using the proposed system and system input.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.5
• /
• pp.975-986
• /
• 2022

Deep learning is attracting great attention, showing excellent performance in image processing, but is vulnerable to adversarial attacks that cause the model to misclassify through perturbation on input data. Adversarial examples generated by adversarial attacks are minimally perturbated where it is difficult to identify, so visual features of the images are not generally changed. Unlikely deep learning models, people are not fooled by adversarial examples, because they classify the images based on such visual features of images. This paper proposes adversarial attack detection method using Symbolic Representation, which is a visual and symbolic features such as color, shape of the image. We detect a adversarial examples by comparing the converted Symbolic Representation from the classification results for the input image and Symbolic Representation extracted from the input images. As a result of measuring performance on adversarial examples by various attack method, detection rates differed depending on attack targets and methods, but was up to 99.02% for specific target attack.

• Journal of Fashion Business
• /
• v.26 no.2
• /
• pp.15-27
• /
• 2022

The purpose of this study is to perform research of ecology concepts expressed in fashion and to propose a new ecology fashion design based on the results. As a specific research method, first, to determine the concept of ecology, the contents related to ecology were extracted and organized through literature research, and then a fashion collection research was conducted to acquire basic design data, such as silhouettes, materials, details, and colors. From WGSN, and a total of 57 images were selected and used as basic data for the design suggestions. As a result of collecting the collection images, it was found that cotton or denim fabrics were mainly used, and in particular, the frequency of use was high mainly for bright tones. The use of chambray, lace, and wrinkles, was also frequent, and the use of expression techniques using burn-out, bleach, and bleaching effects or parts, or entire dyeing was often noticeable. The colors showed neutral and pastel-toned characteristics, and the silhouette was mainly composed of long silhouettes centered on maxi, such as a top or calf based on the knee, rather than a mini. Fabrics for the final designs were linen and sappan wood, and they were used for the dye and red was used as the overall color. The silhouettes were simplified, and care was taken to prevent unnecessary waste, such as paper or materials, from being generated during the production stage. We tried to achieve the purpose of eco-logy by refraining from excessive trimming, except for essential subsidiary materials, such as zippers.

• Journal of the Korean Recycled Construction Resources Institute
• /
• v.10 no.1
• /
• pp.56-65
• /
• 2022

This paper aims at estimating self-cleaning performance of mortar coated with photocatalytic suspension under various conditions. Experimental variables included the concentration (1.5 % and 3.0 %) of photocatalytic suspension for coating mortar specimen, the presence of hydrophilic agent in photocatalytic suspension, and applying the primer on the surface of mortar. The color change of methylene blue solution increased and accordingly self-cleaning performance increased as photocatalytic concentration increased. The presence of hydrophilic agent in photocatalytic suspension slightly decreased the self-cleaning performance compared to the conventional photocatalytic suspension. Test results also showed that mortar specimen including primer and specimen not including primer did not show significantly different self-cleaning performance. In addition, cracks on the surface of mortar specimens decreased as the photocatalytic concentration increased. Therefore, increase in cracks on the surface of mortar at different photocatalytic concentration might adversely affect the self-cleaning performance of mortar specimens.

• International Journal of Internet, Broadcasting and Communication
• /
• v.15 no.1
• /
• pp.126-139
• /
• 2023

At present, naked-eye 3D appears more and more commonly on the facades of urban buildings. It brings an incredible visual experience to the audience by simulating the natural 3D 3D space effect. At the same time, it also creates enormous commercial value for city publicity and commercial advertisements. There is much research on naked-eye 3D visual effects, but for right-angle LED screens. Right-angle LED screen's brand-new expression method that has only become popular in recent years, how to convey a realistic naked-eye 3D effect through two LED screens combined at right angles has become a problem worth exploring. To explore the whole design ideas and production process of the naked-eye 3D impact of the right-angle LED screen, this paper is a preliminary study aimed at understanding the performance principle and expression features. Before the case analysis, first, understand the standard virtual 3D space construction techniques. Combining it with the optical illusion phenomenon, according to the expression principle of the naked-eye 3D effect of the right-angle LED screen, it can be summarized into seven expressions: Shadow, Color contrast, Background structure line, Magnify object, Object out of bounds, Object floating, Fusion of picture and background. By analyzing the optical illusion phenomenon used in the case, we summarized the main performance characteristics of the naked eye 3D effect. The emergence of right-angle LED screens breaks the limitation of a single plane of optical illusion art, perfectly combines building facades with naked-eye 3D visual effects, and provides designers with a brand-new creative platform. Understanding its production principles and main expressive features can help designers enter this innovative platform better.

• The Journal of Korea Robotics Society
• /
• v.16 no.4
• /
• pp.336-344
• /
• 2021

This paper presents a robust deep learning-based human tracking framework in crowded environments. For practical human tracking applications, a target must be robustly tracked even in undetected or overcrowded situations. The proposed framework consists of two parts: robust deep learning-based human detection and tracking while recognizing the aforementioned situations. In the former part, target candidates are detected using Detectron2, which is one of the powerful deep learning tools, and their weights are computed and assigned. Subsequently, a candidate with the highest weight is extracted and is utilized to track the target human using a Kalman filter. If the bounding boxes of the extracted candidate and another candidate are overlapped, it is regarded as a crowded situation. In this situation, the center information of the extracted candidate is compensated using the state estimated prior to the crowded situation. When candidates are not detected from Detectron2, it means that the target is completely occluded and the next state of the target is estimated using the Kalman prediction step only. In two experiments, people wearing the same color clothes and having a similar height roam around the given place by overlapping one another. The average error of the proposed framework was measured and compared with one of the conventional approaches. In the error result, the proposed framework showed its robustness in the crowded environments.