• Title/Summary/Keyword: distributed systems

Search Result 4,025, Processing Time 0.03 seconds

A modified U-net for crack segmentation by Self-Attention-Self-Adaption neuron and random elastic deformation

  • Zhao, Jin;Hu, Fangqiao;Qiao, Weidong;Zhai, Weida;Xu, Yang;Bao, Yuequan;Li, Hui
    • Smart Structures and Systems
    • /
    • v.29 no.1
    • /
    • pp.1-16
    • /
    • 2022
  • Despite recent breakthroughs in deep learning and computer vision fields, the pixel-wise identification of tiny objects in high-resolution images with complex disturbances remains challenging. This study proposes a modified U-net for tiny crack segmentation in real-world steel-box-girder bridges. The modified U-net adopts the common U-net framework and a novel Self-Attention-Self-Adaption (SASA) neuron as the fundamental computing element. The Self-Attention module applies softmax and gate operations to obtain the attention vector. It enables the neuron to focus on the most significant receptive fields when processing large-scale feature maps. The Self-Adaption module consists of a multiplayer perceptron subnet and achieves deeper feature extraction inside a single neuron. For data augmentation, a grid-based crack random elastic deformation (CRED) algorithm is designed to enrich the diversities and irregular shapes of distributed cracks. Grid-based uniform control nodes are first set on both input images and binary labels, random offsets are then employed on these control nodes, and bilinear interpolation is performed for the rest pixels. The proposed SASA neuron and CRED algorithm are simultaneously deployed to train the modified U-net. 200 raw images with a high resolution of 4928 × 3264 are collected, 160 for training and the rest 40 for the test. 512 × 512 patches are generated from the original images by a sliding window with an overlap of 256 as inputs. Results show that the average IoU between the recognized and ground-truth cracks reaches 0.409, which is 29.8% higher than the regular U-net. A five-fold cross-validation study is performed to verify that the proposed method is robust to different training and test images. Ablation experiments further demonstrate the effectiveness of the proposed SASA neuron and CRED algorithm. Promotions of the average IoU individually utilizing the SASA and CRED module add up to the final promotion of the full model, indicating that the SASA and CRED modules contribute to the different stages of model and data in the training process.

The Effect of Rice Co-Brand Assets, Trust, and Attachment on Loyalty (쌀 공동브랜드의 자산, 신뢰, 애착이 충성도에 미치는 영향)

  • Kim, Shine
    • Journal of Digital Convergence
    • /
    • v.20 no.5
    • /
    • pp.401-410
    • /
    • 2022
  • This study deals with the relationship among trust, attachment and brand loyalty of agricultural products' rice co-brands, which are the staple food of the people. The research method established the hypothesis of the study under the foundation of prior research and developed the survey. The subjects of the study were distributed, retrieved, and analyzed the survey of 163 rice farmers in Buyeo-gun, Chungcheongnam-do. The empirical analysis results show that: First, hypothesis 1 of the brand awareness and image that "rice brand assets will be a positive relationship to trust" were statistically adopted. In particular, statistical t values showed a difference in consumer confidence over recognition>images. Second, hypothesis 2 of the trust of agricultural rice brands will be a positive influence on attachment and loyalty' statistically supported. In this regard, brand trust was higher in loyalty than attachment. Third, the attachment of agricultural products to rice brands will be a positive influence on loyalty,' was statistically supported. The strategic implications of this study are as follows. First, consumers should be given clues of trust(ex, GAP of Natioanl Approval Licesing, Fam Tour) as they distrust the perceived quality of the rice in the market. Second, the effect of the origin of rice is questionable, so the spread of the production power system should prevent the mixing of rice varieties, that is the spread of the production history systems.

Innovative Technology of Teaching Moodle in Higher Pedagogical Education: from Theory to Pactice

  • Iryna, Rodionova;Serhii, Petrenko;Nataliia, Hoha;Kushevska, Natalia;Tetiana, Siroshtan
    • International Journal of Computer Science & Network Security
    • /
    • v.22 no.8
    • /
    • pp.153-162
    • /
    • 2022
  • Relevance. Innovative activities in education should be aimed at ensuring the comprehensive development of the individual and professional development of students. The main idea of modular technology is that the student should learn by himself, and the teacher manages his learning activities. The advantage of modular technology is the ability of the teacher to design the study of the material in the most interesting and accessible forms for this part of the study group and at the same time achieve the best learning results. Innovative Moodle technology. it is gaining popularity every day, significantly expanding the space of teaching and learning, allowing students to study inter-faculty university programs in depth. The purpose of this study is to assess the quality of implementation of the e-learning system Moodle. The study was conducted at the South Ukrainian National Pedagogical University named after K. D. Ushinsky in order to identify barriers to the effective implementation of innovative distance learning technologies Moodle and introduce a new model that will have a positive impact on the development of e-learning. Methodology. The paper used a combination of theoretical and empirical research methods. These include: scientific analysis of sources on this issue, which allowed us to formulate the initial provisions of the study; analysis of the results of students 'educational activities; pedagogical experiment; questionnaires; monitoring of students' activities in practical classes. Results. This article evaluates the implementation of the principles of distance learning in the process of teaching and learning at the University in terms of quality. The experiment involved 1,250 students studying at the South Ukrainian National Pedagogical University named after K. D. Ushinsky. The survey helped to identify the main barriers to the effective implementation of modern distance learning technologies in the educational process of the University: the lack of readiness of teachers and parents, the lack of necessary skills in applying computer systems of online learning, the inability to interact with the teaching staff and teachers, the lack of a sufficient number of academic consultants online. In addition, internal problems are investigated: limited resources, unevenly distributed marketing advantages, inappropriate administrative structure, and lack of innovative physical capabilities. The article allows us to solve these problems by gradually implementing a distance learning model that is suitable for any university, regardless of its specialization. The Moodle-based e-learning system proposed in this paper was designed to eliminate the identified barriers. Models for implementing distance learning in the learning process were built according to the CAPDM methodology, which helps universities and other educational service providers develop and manage world-class online distance learning programs. Prospects for further research focus on evaluating students' knowledge and abilities over the next six months after the introduction of the proposed Moodle-based program.

Study on Effective Operational Plan of Maritime Safety Enforcement Task : Focus on Vessel Safety Management (해양안전 집행업무의 효율적 운영방안 연구 -선박안전 관리를 중심으로-)

  • Byungdoo Yun
    • Journal of the Korean Society of Marine Environment & Safety
    • /
    • v.28 no.7
    • /
    • pp.1169-1178
    • /
    • 2022
  • This study examines an effective operational plan of a maritime safety enforcement task that is distributed in the Ministry of Oceans and Fisheries and the Korea Coast Guard for maritime safety. Considering the sinking of the MV Sewol and Seohae vessels and the MV Hebei Spirit oil spill, normally marine safety accidents tend to not only damage human life, property, and marine environment but can also expand to disasters; therefore, precautionary measures are required. However, the Korean government takes superficial efforts in case of large-scale accidents. In case of the MV Sewol ferry, the government only focused on punishing the officials involved as a follow up and never mentioned any effective plan, such as "unification of maritime safety enforcement," in contrast to an advanced country. As a result, there are endless major and minor backward accidents. The probability of large-scale maritime accidents and backwardness accidents is increasing owing to vessels becoming large sized, high speed, and aged; sailors being unqualified; port traffic increase, development of marine leisure; and inefficiency dual marine safety systems. Therefore, based on the review of previous studies related to maritime safety, major advanced country's cases, and unified case of the vessel traffic service, this study suggests effective methods such as coastal vessel safety management, port state control and aid to navigation management, which are directly connected with maritime safety.

Fake News Detection on YouTube Using Related Video Information (관련 동영상 정보를 활용한 YouTube 가짜뉴스 탐지 기법)

  • Junho Kim;Yongjun Shin;Hyunchul Ahn
    • Journal of Intelligence and Information Systems
    • /
    • v.29 no.3
    • /
    • pp.19-36
    • /
    • 2023
  • As advances in information and communication technology have made it easier for anyone to produce and disseminate information, a new problem has emerged: fake news, which is false information intentionally shared to mislead people. Initially spread mainly through text, fake news has gradually evolved and is now distributed in multimedia formats. Since its founding in 2005, YouTube has become the world's leading video platform and is used by most people worldwide. However, it has also become a primary source of fake news, causing social problems. Various researchers have been working on detecting fake news on YouTube. There are content-based and background information-based approaches to fake news detection. Still, content-based approaches are dominant when looking at conventional fake news research and YouTube fake news detection research. This study proposes a fake news detection method based on background information rather than content-based fake news detection. In detail, we suggest detecting fake news by utilizing related video information from YouTube. Specifically, the method detects fake news through CNN, a deep learning network, from the vectorized information obtained from related videos and the original video using Doc2vec, an embedding technique. The empirical analysis shows that the proposed method has better prediction performance than the existing content-based approach to detecting fake news on YouTube. The proposed method in this study contributes to making our society safer and more reliable by preventing the spread of fake news on YouTube, which is highly contagious.

Computer Vision-based Continuous Large-scale Site Monitoring System through Edge Computing and Small-Object Detection

  • Kim, Yeonjoo;Kim, Siyeon;Hwang, Sungjoo;Hong, Seok Hwan
    • International conference on construction engineering and project management
    • /
    • 2022.06a
    • /
    • pp.1243-1244
    • /
    • 2022
  • In recent years, the growing interest in off-site construction has led to factories scaling up their manufacturing and production processes in the construction sector. Consequently, continuous large-scale site monitoring in low-variability environments, such as prefabricated components production plants (precast concrete production), has gained increasing importance. Although many studies on computer vision-based site monitoring have been conducted, challenges for deploying this technology for large-scale field applications still remain. One of the issues is collecting and transmitting vast amounts of video data. Continuous site monitoring systems are based on real-time video data collection and analysis, which requires excessive computational resources and network traffic. In addition, it is difficult to integrate various object information with different sizes and scales into a single scene. Various sizes and types of objects (e.g., workers, heavy equipment, and materials) exist in a plant production environment, and these objects should be detected simultaneously for effective site monitoring. However, with the existing object detection algorithms, it is difficult to simultaneously detect objects with significant differences in size because collecting and training massive amounts of object image data with various scales is necessary. This study thus developed a large-scale site monitoring system using edge computing and a small-object detection system to solve these problems. Edge computing is a distributed information technology architecture wherein the image or video data is processed near the originating source, not on a centralized server or cloud. By inferring information from the AI computing module equipped with CCTVs and communicating only the processed information with the server, it is possible to reduce excessive network traffic. Small-object detection is an innovative method to detect different-sized objects by cropping the raw image and setting the appropriate number of rows and columns for image splitting based on the target object size. This enables the detection of small objects from cropped and magnified images. The detected small objects can then be expressed in the original image. In the inference process, this study used the YOLO-v5 algorithm, known for its fast processing speed and widely used for real-time object detection. This method could effectively detect large and even small objects that were difficult to detect with the existing object detection algorithms. When the large-scale site monitoring system was tested, it performed well in detecting small objects, such as workers in a large-scale view of construction sites, which were inaccurately detected by the existing algorithms. Our next goal is to incorporate various safety monitoring and risk analysis algorithms into this system, such as collision risk estimation, based on the time-to-collision concept, enabling the optimization of safety routes by accumulating workers' paths and inferring the risky areas based on workers' trajectory patterns. Through such developments, this continuous large-scale site monitoring system can guide a construction plant's safety management system more effectively.

  • PDF

Photosynthetic and respiratory responses of the surfgrass, Phyllospadix japonicus, to the rising water temperature (수온 상승에 따른 게바다말의 광합성 및 호흡률 변화)

  • Hyegwang Kim;Jong-Hyeob Kim;Seung Hyeon Kim;Zhaxi Suonan;Kun-Seop Lee
    • Korean Journal of Environmental Biology
    • /
    • v.40 no.3
    • /
    • pp.352-362
    • /
    • 2022
  • Photosynthesis and respiration of seagrasses are mainly controlled by water temperature. In this study, the photosynthetic physiology and respiratory changes of the Asian surfgrass Phyllospadix japonicus, which is mainly distributed on the eastern and southern coasts of Korea, were investigated in response to changing water temperature (5, 10, 15, 20, 25, and 30℃) by conducting mesocosm experiments. Photosynthetic parameters (maximum photosynthetic rate, Pmax; compensation irradiance, Ic; and saturation irradiance, Ik) and respiration rate of surfgrass increased with rising water temperature, whereas photosynthetic efficiency (α) was fairly constant among the water temperature conditions. The Pmax and Ik dramatically decreased under the highest water temperature condition (30℃), whereas the Ic and respiration rate increased continuously with the increasing water temperature. Ratios of maximum photosynthetic rates to respiration rates (Pmax : R) were highest at 5℃ and declined markedly at higher temperatures with the lowest ratio at 30℃. The minimum requirement of Hsat (the daily period of irradiance-saturated photosynthesis) of P. japonicus was 2.5 hours at 5℃ and 10.6 hours at 30℃ for the positive carbon balance. Because longer Hsat was required for the positive carbon balance of P. japonicus under the increased water temperature, the rising water temperature should have negatively affected the growth, distribution, and survival of P. japonicus on the coast of Korea. Since the temperature in the temperate coastal waters is rising gradually due to global warming, the results of this study could provide insights into surfgrass responses to future severe sea warming and light attenuation.

XQuery Query Rewriting for Query Optimization in Distributed Environments (분산 환경에 질의 최적화를 위한 XQuery 질의 재작성)

  • Park, Jong-Hyun;Kang, Ji-Hoon
    • Journal of the Korea Society of Computer and Information
    • /
    • v.14 no.3
    • /
    • pp.1-11
    • /
    • 2009
  • XQuery query proposed by W3C is one of the standard query languages for XML data and is widely accepted by many applications. Therefore the studies for efficient Processing of XQuery query have become a topic of critical importance recently and the optimization of XQuery query is one of new issues in these studies. However, previous researches just focus on the optimization techniques for a specific XML data management system and these optimization techniques can not be used under the any XML data management systems. Also, some previous researches use predefined XML data structure information such as XML schema or DTD for the optimization. In the real situation, however applications do not all refer to the structure information for XML data. Therefore, this paper analyzes only a XQuery query and optimize by using itself of the XQuery query. In this paper, we propose 3 kinds of optimization method that considers the characteristic of XQuery query. First method removes the redundant expressions described in XQuery query second method replaces the processing order of operation and clause in XQuery query and third method rewrites the XQuery query based on FOR clause. In case of third method, we consider FOR clause because generally FOR clause generates a loop in XQuery query and the loop often rises to execution frequency of redundant operation. Through a performance evaluation, we show that the processing time for rewritten queries is less than for original queries. also each method in our XQuery query optimizer can be used separately because the each method is independent.

Design and Implementation of an Ethereum-Based Deliverables Management System for Public Information Software Project (이더리움 기반 공공정보 소프트웨어 사업산출물 관리 시스템 설계 및 구현)

  • Lee, Eun Ju;Kim, Jin Wook
    • KIPS Transactions on Computer and Communication Systems
    • /
    • v.11 no.6
    • /
    • pp.175-184
    • /
    • 2022
  • Blockchain is being studied in various fields such as logistics, fintech, medical care, and the public sector. In the public information software project, some deliverables are omitted because the developed deliverables and the deliverables requested by the project management methodology do not match, and an additional process is required for payment. In this paper, we propose the deliverables management system for public information software project which is configured a distributed environment using the Ethereum blockchain and which has an automatic payment system only when all deliverables are approved. This system can keep the service available in case of system failure, provide transparency and traceability of deliverables management, and can reduce conflicts between the ordering company and the contractor through automatic payment. In this system, the information of deliverables is stored in the blockchain, and the deliverables that their file name is the hash value calculated by using the version information and the hash value of the previous version deliverable, are stored in the SFTP server. Experimental results show that the hash value of the deliverables registered by the contractor is correct, the file name of the deliverables stored in the SFTP server is the same as the hash value registered in the Ethereum blockchain, and the payment is made automatically to the Ethereum address of the contractor when all deliverables are approved.

Research Trends in Record Management Using Unstructured Text Data Analysis (비정형 텍스트 데이터 분석을 활용한 기록관리 분야 연구동향)

  • Deokyong Hong;Junseok Heo
    • Journal of Korean Society of Archives and Records Management
    • /
    • v.23 no.4
    • /
    • pp.73-89
    • /
    • 2023
  • This study aims to analyze the frequency of keywords used in Korean abstracts, which are unstructured text data in the domestic record management research field, using text mining techniques to identify domestic record management research trends through distance analysis between keywords. To this end, 1,157 keywords of 77,578 journals were visualized by extracting 1,157 articles from 7 journal types (28 types) searched by major category (complex study) and middle category (literature informatics) from the institutional statistics (registered site, candidate site) of the Korean Citation Index (KCI). Analysis of t-Distributed Stochastic Neighbor Embedding (t-SNE) and Scattertext using Word2vec was performed. As a result of the analysis, first, it was confirmed that keywords such as "record management" (889 times), "analysis" (888 times), "archive" (742 times), "record" (562 times), and "utilization" (449 times) were treated as significant topics by researchers. Second, Word2vec analysis generated vector representations between keywords, and similarity distances were investigated and visualized using t-SNE and Scattertext. In the visualization results, the research area for record management was divided into two groups, with keywords such as "archiving," "national record management," "standardization," "official documents," and "record management systems" occurring frequently in the first group (past). On the other hand, keywords such as "community," "data," "record information service," "online," and "digital archives" in the second group (current) were garnering substantial focus.