• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권9호
• /
• pp.4665-4683
• /
• 2019

Previous methods build image annotation model by leveraging three basic dependencies: relations between image and label (image/label), between images (image/image) and between labels (label/label). Even though plenty of researches show that multiple dependencies can work jointly to improve annotation performance, different dependencies actually do not "work jointly" in their diagram, whose performance is largely depending on the result predicted by image/label section. To address this problem, we propose the adaptive attention annotation model (AAAM) to associate these dependencies with the prediction path, which is composed of a series of labels (tags) in the order they are detected. In particular, we optimize the prediction path by detecting the relevant labels from the easy-to-detect to the hard-to-detect, which are found using Binary Cross-Entropy (BCE) and Triplet Margin (TM) losses, respectively. Besides, in order to capture the inforamtion of each label, instead of explicitly extracting regional featutres, we propose the self-attention machanism to implicitly enhance the relevant region and restrain those irrelevant. To validate the effective of the model, we conduct experiments on three well-known public datasets, COCO 2014, IAPR TC-12 and NUSWIDE, and achieve better performance than the state-of-the-art methods.

• International Journal of Computer Science & Network Security
• /
• 제22권9호
• /
• pp.327-333
• /
• 2022

Wireless Sensor Networks (WSNs) plays an important role in our everyday life. WSN is distributed in all the places. Nowadays WSN devices are developing our world as smart and easy to access and user-friendly. The sensor is connected to all the resources based on the uses of devices and the environment [1]. In WSN, Quality of Service is based on time synchronization and scheduling. Scheduling is important in WSN. The schedule is based on time synchronization. Min-Max data size scheduling is used in this proposed work. It is used to reduce the Delay & Energy. In this proposed work, Two-hop neighboring node is used to reduce energy consumption. Data Scheduling is used to identify the shortest path and transmit the data based on weightage. The data size is identified by three size of measurement Min, Max and Medium. The data transmission is based on time, energy, delivery, etc., the data are sent through the first level shortest path, then the data size medium, the second level shortest path is used to send the data, then the data size is small, it should be sent through the third level shortest path.

• 한국정밀공학회지
• /
• 제11권6호
• /
• pp.28-41
• /
• 1994

An efficient algorithm is proposed to select the proper tools and generate their paths for NC rough cutting of dies and molds with sculptured surfaces. Even though a milling process consists of roughing, semi-finishing, and finishing, most material is removed by a rough cutting process. Therfore it can be said that the rough cutting process occupy an important portion of the NC milling process, and accordingly, an efficient rough cutting method contributes to an efficient milling process. In order work, the following basic assumption is accepted for the efficient machining. That is, to machine a region bounded by a profile, larger tools should be used in the far inside and the region adjacent to relatively simple portion of the boundary while smaller tools are used in the regions adjacent to the relatively complex protion. Thus the tools are selected based on the complexity of the boundary profile adjacent to the region to be machined. An index called cutting path ratio is proposed in this work as a measure of the relative complexity of the profile with respect to a tool diameter. Once the tools are selected, their tool paths are calculated starting from the largest to the smallest tool.

• 대한안전경영과학회지
• /
• 제11권4호
• /
• pp.57-67
• /
• 2009

The psychosocial stress and musculoskeletal disorders(MSDs) have been one of major health problems for hospital workers. This study tried to understand the relationship between symptoms associated with MSDs and risk factors such as working posture, job stress, psychosocial stress and fatigue. A total number of 655 hospital workers participated in this study. Specifically, REBA was applied for evaluating working posture and a checklist prepared by KOSHA(Korean Occupational Safety and Health Agency) was used for symptom survey. A questionnaire from KOSHA was also used for collecting data associated with job stress, psychosocial stress and fatigue. All these data were formulated and modeled by path analysis which was one of major statistical tools in this study. Specifically, path analysis for the data we collected came up with several major findings. As a result, as for body part(neck), (waist) and (arms) the degree of risk of work posture measured with the use of job stress(KOSS), psycho-social stress(PWI-SF) and REBA is significantly more affected by fatigue than muscular skeletal disease related consciousness symptom. However, regarding bod(wrist), the degree of risk of work posture measured with the use of job stress(KOSS) and REBA is directly affected by muscular skeletal disease related consciousness symptom. This study is meaningful in that the study clarified the causal relations of the degree of risk of work posture, degree of fatigue, and muscular skeletal disease related consciousness symptoms by each body part measured with the use of work stress(KOSS), psycho-social stress(PWI-SF) and REBA.

• 한국산업보건학회지
• /
• 제21권4호
• /
• pp.209-214
• /
• 2011

Objectives: To assess the risk of pushing or pulling the yarn-carrying cart, the survey was performed in some fiber manufacturing factories. Methods: We selected 6 fiber-twisting factories which agreed to in-site survey of their workplace. To measure both initial and sustained forces of the push-pull tasks, Chatillon CSD500 dynamometer(2004, Ametek, USA) was used. The mean of 3 tests for the same cart was adopted as the measured forces. Height and width of cart, weight of spooled yarns, and distance of movement were also measured. Inspection of cart wheel, moving path, and the actual hand position while moving was done. Results: More than one pushing or pulling task exceeded the push-pull force limits of design goal in 5 factories. Though the cart was not loaded the heaviest weight in the factory, the measured push or pull force exceeded the limits several times. A few cart wheels were worn out and tangled with pieces of yarn. It was also observed some holes in the moving path. Conclusions: While the push-pull task is not included in the 11 scopes of over-burdened work notified by Korean government, it should be recognized as risk factor of work-related musculoskeletal disorders. The maintenance work such as regular change and frequent cleaning of cart wheel, the use of fitting wheel, and flattening of bumpy floor through the moving path should be advised importantly in the worksite management of work-related musculoskeletal disorders.

• 한국멀티미디어학회논문지
• /
• 제7권10호
• /
• pp.1421-1435
• /
• 2004

기론 이동 에이전트 모델은 사용자로부터 다양한 작업을 부여 받았을 때, 수동적 라우팅 스케줄 지정에 따라 많은 트래픽이 발생한 시점에서 노드 간을 이주하며 작업을 처리한 후 작업 결과를 서비스하기까지 분산 노드들에 대한 순회 작업 처리 시간 비용이 증가하게 된다. 따라서 기존 이동 에이전트 시스템들은 에이전트 시스템들 간의 에이전트의 크기 증가나 상호 호환성 결여를 해결하기 위한 이주 기법이 요구된다. 이에 본 논문에서는 이동 에이전트의 노드 이주시 네트워크 트래픽 감지를 통한 최적 경로 탐색에 의해 분산된 노드들로의 호율적 작업 처리를 위한 CORBA 기반의 이동 에이전트 모델(CORBA-based Mobile Agent Model : CMAM)을 설계 및 구현한다.

• 치위생과학회지
• /
• 제18권1호
• /
• pp.32-41
• /
• 2018

This study investigated the relationship between the organizational culture, organizational support, organizational health, personal health, and quality of work life of dental hygienists and analyzed the factors affecting the quality of work life in order to identify ways to improve their quality of work life. A total of 320 dental hygienists completed a self-administered survey; after excluding data from 21 respondents, 299 responses were included in the analysis. Frequency analyses, t-tests, one-way analysis of variation (ANOVA), and correlation analyses were conducted. A path analysis was also conducted to confirm the causal relationships. The findings are as follows. First, there was a significant difference in several general characteristics of the organizational culture including years in the current job and the number of dental hygienists; organizational support including age and the number of dental hygienists; organizational health including years in the current job and annual salary; and personal health including annual salary. Second, the quality of work life showed a positive correlation with organizational culture, organizational support, personal health, and organizational health in that order. Third, the results of path analysis revealed that organizational culture had a positive effect on organizational support; organizational support and personal health on organizational health; organizational support on personal health; and organizational support and organizational health on quality of work life. In addition, organizational support and organizational health had a direct effect on the quality of work life, while organizational culture, organizational support, and personal health had an indirect effect. These results indicated existence of a relationship among organizational culture, organizational support, organizational health, personal health, and quality of work life. It is necessary to identify ways to improve the quality of work life of dental hygienists.

• Journal of Electrical Engineering and Technology
• /
• 제12권2호
• /
• pp.918-925
• /
• 2017

This paper presents a path planning problem for multi-robot system in the environment with dynamic obstacles. In order to guide the robots move along a collision-free path efficiently and reach the goal position quickly, a navigation method based on fuzzy logic controllers has been developed by using proximity sensors. There are two kinds of fuzzy controllers developed in this work, one is used for obstacle avoidance and the other is used for orientation to the target. Both static and dynamic obstacles are included in the environment and the dynamic obstacles are defined with no type of restriction of direction and velocity. Here, the environment is unknown for all the robots and the robots should detect the surrounding information only by the sensors installed on their bodies. The simulation results show that the proposed method has a positive effectiveness for the path planning problem.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 2002년도 춘계학술대회 논문집
• /
• pp.184-189
• /
• 2002

Collision free task planning for dual-arm robot which perform many subtasks in a common work space can be achieved in two steps : path planning and trajectory planning. Path planning finds the order of tasks for each robot to minimize path lengths as well as to avoid collision with static obstacles. A trajectory planning strategy is to let each robot move along its path as fast as possible and delay one robot at its initial position or reduce speed at the middle of its path to avoid collision with the other robot.

• ETRI Journal
• /
• 제31권1호
• /
• pp.31-41
• /
• 2009

Optical burst switching is a promising switching paradigm for the next IP-over-optical network backbones. However, its burst loss performance is greatly affected by burst contention. Several methods have been proposed to address this problem, some of them requiring the network to be flooded by frequent state dissemination signaling messages. In this work, we present a traffic engineering approach for path selection with the objective of minimizing contention using only topological information. The main idea is to balance the traffic across the network to reduce congestion without incurring link state dissemination protocol penalties. We propose and evaluate two path selection strategies that clearly outperform shortest path routing. The proposed path selection strategies can be used in combination with other contention resolution methods to achieve higher levels of performance and support the network reaching stability when it is pushed under stringent working conditions. Results show that the network connectivity is an important parameter to consider.