• Journal of Intelligence and Information Systems
• /
• v.28 no.4
• /
• pp.329-346
• /
• 2022

Recently, as the market for short form videos (Instagram, TikTok, YouTube) on social media has gradually increased, research using them is actively being conducted in the artificial intelligence field. A representative research field is Video to Shop, which detects fashion products in videos and searches for product images. In such a video-based artificial intelligence model, product features are extracted using convolution operations. However, due to the limitation of computational resources, extracting features using all the frames in the video is practically impossible. For this reason, existing studies have improved the model's performance by sampling only a part of the entire frame or developing a sampling method using the subject's characteristics. In the existing Video to Shop study, when sampling frames, some frames are randomly sampled or sampled at even intervals. However, this sampling method degrades the performance of the fashion product search model while sampling noise frames where the product does not exist. Therefore, this paper proposes a sampling method MF (Missing Fashion items on frame) sampler that removes noise frames and improves the performance of the search model. MF sampler has improved the problem of resource limitations by developing a keyframe mechanism. In addition, the performance of the search model is improved through noise frame removal using the noise detection model. As a result of the experiment, it was confirmed that the proposed method improves the model's performance and helps the model training to be effective.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.4
• /
• pp.333-342
• /
• 2022

This paper has theoretically and practically reviewed the ISO (International Standard Organization) 17123-8 standard not only to raise the appropriateness for introducing performance criteria of GNSS (Global Navigation Satellite Systems) surveying equipment based on RTK (Real-Time Kinematic) accuracy but also to derive its proper test procedure by introducing the international standard. Field experiments have been performed to appreciate the GNSS-RTK accuracy of five selected receivers via the full testing procedure of the ISO standard, which statistically compares the so-called experimental standard deviations with themselves and with the reference accuracy. A series of statistical tests have revealed that the RTK accuracy of the same class receivers is not identical, whereas that of the different classes can be equivalent. Such a result evidences the urgency of adopting an RTK accuracy-based specification of the GNSS equipment to the performance standard, currently referenced to the static observation technique only. It is believed that this transition helps the maximization of a new generation of cost-effective receivers to legal surveying applications. Finally, this study proposes the ISO full test, comparing an experimental standard deviation with its referenced value, for a potential field verification procedure of the new performance standard.

• Journal of the Korean GEO-environmental Society
• /
• v.22 no.12
• /
• pp.65-70
• /
• 2021

River-side Embankment collapse involves various causes. The embankment collapse due to internal erosion around embedded structures reaches up to more than 10% in Korea. Many studies are being attempted to prevent from the collapse of the embankment rooted from overtopping and instability as well as internal erosion. One of them is the study on the application of biopolymers. The application of biopolymers to soils are divided into enhancing strength, vegetation and erosion resistance. This study investigated the effect of biopolymer treated soil on erosion resistance. The main goal of the study is to obtain basic data for real-scale experiments to verify the effectiveness of biopolymer treated soil embankment including a review of the collapse pattern in the model embankment with various test conditions. The optimized experimental conditions were selected by examining the erosion patterns according to each induction path with three compaction degree of the model embankment. As a result of the experiment, the internal erosion rate in the embankment to which the biopolymer treated soil was applied is greatly reduced, and it could be concluded that it might be applied to the actual embankment. However, in this study, the conclusion was drawn only within the scaled-down model embankment. In order to practically apply the biopolymer treated soil to the embankment, the study considering the scale effect would be needed.

• Journal of Practical Engineering Education
• /
• v.14 no.2
• /
• pp.267-277
• /
• 2022

Achievement at university is recognized in a comprehensive sense as the level of qualitative change and development that students have embodied as a result of their experience in university education. Therefore, the academic achievement of university students will be given meaning in cooperation with the historical and social demands for diverse human resources such as creativity, leadership, and global ability, but it is practically an indicator of the outcome of university education. Measurement of academic achievement by such credits involves many problems, but in particular, standardization of academic achievement by credits based on evaluation methods, contents, and university rankings is a very difficult problem. In this study, we present a model that uses machine learning techniques to predict whether or not academic achievement is excellent for D-University graduates. The variables used were analyzed using up to 96 personal information and bachelor's information such as graduation year, department number, department name, etc., but when establishing a future education course, only the data after enrollment works effectively. Therefore, the items to be analyzed are limited to the recommended ability to improve the academic achievement of the department/student. In this research, we implemented an academic achievement prediction model through analysis of core abilities that reflect the philosophy, goals, human resources image, and utilized machine learning to affect the impact of the introduction of the prediction model on academic achievement. We plan to apply the results of future research to the establishment of curriculum and student guidance conducted in the department to establish a basis for improving academic achievement.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1263-1263
• /
• 2001

The amounts of organic matter present in soil and the rate of soil organic matter (SOM) turnover are influenced by agricultural management practice, such as rotation, tillage, forage plow down direct seeding and manure application. The amount of nutrients released from SOM is highly dependent upon the state of the organic matter. If it contains a large proportion of light fractions (low-density) more nutrients will be available to the glowing crops. However, if it contains mostly heavy fractions (high-density) that are difficult to breakdown, then lesser amounts of nutrients will be available. The state of the SOM and subsequent release of nutrients into the soil can be predicted by NIRS as long as a robust regression equation is developed. The NIRS method is known for its rapidity, convenience, simplicity, accuracy and ability to analyze many constituents at the same time. Our hypothesis is that the NIRS technique allows researchers to investigate fully and in more detail each field for the status of SOM, available moisture and other soil properties in Alberta soils for precision farming in the near future. One hundred thirty one (131) Alberta soils with various levels (low 2-6%, medium 6-10%, and high >10%) of organic matter content and most of dry land soils, including some irrigated soils from Southern Alberta, under various management practices were collected throughout Northern, Central and Southern Alberta. Two depths (0- 15 cm and 15-30 cm) of soils from Northern Alberta were also collected. These air-dried soil samples were ground through 2 mm sieve and scanned using Foss NIR System 6500 with transport module and natural product cell. With particle size above 150 microns only, the “Ludox” method (Meijboom, Hassink and van Noorwijk, Soil Biol. Biochem.27: 1109-1111, 1995) which uses stable silica, was used to fractionate SOM into light, medium and heavy fractions with densities of <1.13, 1.13-1.37 and >1.37 respectively, The SOM fraction with the particle size below 150 microns was discarded because practically, this fraction with very fine particles can't be further separated by wet sieving based on density. Total organic matter content, mechanical texture, ash after 375$^{\circ}C$, and dry matter (DM) were also determined by “standard” soil analysis methods. The NIRS regression equations were developed using Infra-Soft-International (ISI) software, version 3.11.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1251-1251
• /
• 2001

The routine analysis of milk chemical components is of major importance both for the management of animals in dairy farms and for quality control in dairy industries. NIRS technology is an analytical technique which greatly simplifies this routine. One of the most critical aspects in NIRS analysis of milk is sample preparation and analysis modes which should be fast and straightforward. An important difficulty when obtaining NIR spectra of milk is the high water content (80 to 90%) of this product, since water absorbs most of the infrared radiation, and, therefore, limits the accuracy of calibrating for other constituents. To avoid this problem, the DESIR system was set up. Other ways of radiation-sample interaction adapted for liquids or semi-liquids exist, which are practically instantaneous and with limited or null necessity of sample preparation: Transmission and Folded Transmission or Transflectance. The objective of the present work is to compare the precision and accuracy of milk calibration equations in two analysis modes: Reflectance (dry milk) and Folded Transmission (liquid milk). A FOSS-NIR Systems 6500 I spectrophotometer (400-2500 nm) provided with a spinning module was used. Two NIR spectroscopic methods for milk analysis were compared: a) folded transmission: liquid milk samples in a 0.1 pathlength sample cell (ref. IH-0345) and b) reflectance: dried milk samples in glass fibre filters placed in a standard ring cell. A set of 101 milk samples was used to develop the calibration equations, for the two NIR analysis modes, to predict casein, protein, fat and dry matter contents, and 48 milk samples to predict Somatic Cell Count (SCC). The calibrations obtained for protein, fat and dry matter have an excellent quantitative prediction power, since they present $r^2$ values higher than 0.9. The $r^2$ values are slightly lower for casein and SCC (0.88 and 0.89 respectively), but they still are sufficiently high. The accuracy of casein, protein and SCC equations is not affected by the analysis modes, since their ETVC values are very similar in reflectance and folded transmission (0.19% vs 0.21%; 0.16% vs 0.19% and 55.57% vs 53.11% respectively), Lower SECV values were obtained for the prediction of fat and dry matter with the folded transmission equations (0.14% and 0.25% respectively) compared to the results with the reflectance ones (0.43% and 0.34% respectively). In terms of accuracy and speed of analytical response, NIRS analysis of liquid milk is recommended (folded transmission), since the drying procedure takes 24 hours. However, both analysis modes offer satisfactory results.

• Journal of Korean Library and Information Science Society
• /
• v.54 no.1
• /
• pp.91-114
• /
• 2023

This study increases awareness of the survey and demand for libraries in correctional facilities, which have not received much attention so far, and is necessary for improving the operation of existing research and libraries in Korea, which is conducting a current status survey of most correctional facility libraries. The contents of the service operation of the external correctional facility library were explored and analyzed. As a result of the analysis, the United States, the United Kingdom, and Germany, where correctional facility libraries are relatively well operated, are practically operating social rehabilitation programs while reducing recidivism of inmates, and library associations or local governments in each country are promoting the vitalization of correctional facility libraries. It helps and organizes committees or organizations involving related experts to discuss policy support, issue checks, research, report publication, drafting and revising guide guidelines, cooperation between libraries, and reflect them in actual operation. In addition, it maintains steady support and interest through public-private cooperation, and makes good use of the linkage system with local public libraries. As a result, it contributes to reducing the recidivism of prisoners. Through this study, the following three areas were suggested for the administrative improvement direction for correctional facility library services. First, the production and distribution of guidelines or guidebooks for correctional facility libraries, second, the establishment of a specialized correctional facility library organization within the correctional headquarters or library association, and third, the strengthening of activities and the operation of law libraries in correctional facilities.

• Maritime Security
• /
• v.5 no.1
• /
• pp.55-83
• /
• 2022

For humankind, the ocean is a treasure trove of natural resources and an important area that provides major transportation routes. However, marine ecosystems are under threat amidst the global climate change crisis. This is partly due to various sources of pollution emitted from ships, shore facilities, and other sources. In response, the navies of advanced countries such as the United States, the United Kingdom, and Europe have assessed such a climate change crisis as a new maritime security threat. These countries have made early efforts to prevent marine pollution from warships. These efforts have been legally embodied. This was legislation for technical and organizational structures to be practically applied in the field. Among these navies of developed countries, France has established parliamentary and intergovernmental plans to become a leader in environmental protection at the defense level, not only in the European region but throughout the world. Within this framework, the French Navy, through its internal instruction, has inclined its legal efforts to prevent marine pollution from warships. Therefore, this study examines the legislation for the prevention of marine pollution from warships within the French Navy and compares it with the marine environmental legislation applicable to the ROK Navy. It then deduces the implications for the ROK navy, which is advancing toward a Blue-Water navy.

• Journal of the Korean Society of Safety
• /
• v.38 no.3
• /
• pp.77-84
• /
• 2023

Welding and cutting, which are representative tasks in handling firearms at industrial sites, are the basis for production and maintenance processes across all industries. They are also essential in the root industry. Specifically, they are widely used in the manufacturing industry, including equipment industries such as shipbuilding, automobiles, and chemicals, and subsequent maintenance work and general facility repair. However, such hot work carries a high fire risk owing to sparks scattering and inadequate management, resulting in a high occurrence of accidents. In response, the government and relevant organizations have recently revised the Occupational Safety and Health Act to prevent accidents during hot work. These revisions impose more stringent regulations than before, which are expected to help prevent actual fire accidents. However, whether the fire observer system, which is the core element of the revision, would be practically applied and maintained is unclear. Therefore, this study compared the fire observer system in the revised Occupational Safety and Health Act with those in the laws and systems of developed countries, conducted interviews with safety and health experts to assess the suitability of the new system for fire observer operations, and improvement plans were derived accordingly. Therefore, the laws and systems of developed countries grant more authority to fire observers compared with those of Korea. Moreover, professional training in handling emergency is required. Interviews with safety and health experts revealed that regardless of company size, the same operating standards were applied, and standards for deploying fire observers in various locations were unclear. Furthermore, there was a lack of professional education and training, and the role and authority of fire observers were limited. These findings revealed a problem in this sector. The results of this study are expected to serve as basic data for establishing a practical system for placing fire observers and supplementing laws, guidelines, and systems for preventing fire accidents.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.5
• /
• pp.403-411
• /
• 2023

Functional ultrasound imaging, such as elasticity imaging and micro-blood flow Doppler imaging, enhances diagnostic capability by providing useful mechanical and functional information about tissues. However, the implementation of functional ultrasound imaging poses limitations such as the storage of vast amounts of data in Radio Frequency (RF) data acquisition and processing. In this paper, we propose a sub-Nyquist approach that reduces the amount of acquired axial samples for efficient shear-wave elasticity imaging. The proposed method acquires data at a sampling rate one-third lower than the conventional Nyquist sampling rate and tracks shear-wave signals through RF signals reconstructed using band-pass filtering-based interpolation. In this approach, the RF signal is assumed to have a fractional bandwidth of 67 %. To validate the approach, we reconstruct the shear-wave velocity images using shear-wave tracking data obtained by conventional and proposed approaches, and compare the group velocity, contrast-to-noise ratio, and structural similarity index measurement. We qualitatively and quantitatively demonstrate the potential of sub-Nyquist sampling-based shear-wave elasticity imaging, indicating that our approach could be practically useful in three-dimensional shear-wave elasticity imaging, where a massive amount of ultrasound data is required.