• International journal of advanced smart convergence
• /
• 제8권1호
• /
• pp.24-34
• /
• 2019

In this research, a practical deep learning framework to differentiate the lesions and nodules in breast acquired with ultrasound imaging has been proposed. 7408 ultrasound breast images of 5151 patient cases were collected. All cases were biopsy proven and lesions were semi-automatically segmented. To compensate for the shift caused in the segmentation, the boundaries of each lesion were drawn using Fully Convolutional Networks(FCN) segmentation method based on the radiologist's specified point. The data set consists of 4254 benign and 3154 malignant lesions. In 7408 ultrasound breast images, the number of training images is 6579, and the number of test images is 829. The margin between the boundary of each lesion and the boundary of the image itself varied for training image augmentation. The training images were augmented by varying the margin between the boundary of each lesion and the boundary of the image itself. The images were processed through histogram equalization, image cropping, and margin augmentation. The networks trained on the data with augmentation and the data without augmentation all had AUC over 0.95. The network exhibited about 90% accuracy, 0.86 sensitivity and 0.95 specificity. Although the proposed framework still requires to point to the location of the target ROI with the help of radiologists, the result of the suggested framework showed promising results. It supports human radiologist to give successful performance and helps to create a fluent diagnostic workflow that meets the fundamental purpose of CADx.

• International journal of advanced smart convergence
• /
• 제8권4호
• /
• pp.68-74
• /
• 2019

Along with the deeper architecture in the deep learning approaches, the need for the data becomes very big. In the real problem, to get huge data in some disciplines is very costly. Therefore, learning on limited data in the recent years turns to be a very appealing area. Meta-learning offers a new perspective to learn a model with this limitation. A state-of-the-art model that is made using a meta-learning framework, Meta-SGD, is proposed with a key idea of learning a hyperparameter or a learning rate of the fast adaptation stage in the outer update. However, this learning rate usually is set to be very small. In consequence, the objective function of SGD will give a little improvement to our weight parameters. In other words, the prior is being a key value of getting a good adaptation. As a goal of meta-learning approaches, learning using a single gradient step in the inner update may lead to a bad performance. Especially if the prior that we use is far from the expected one, or it works in the opposite way that it is very effective to adapt the model. By this reason, we propose to add a weight term to decrease, or increase in some conditions, the effect of this prior. The experiment on few-shot learning shows that emphasizing or weakening the prior can give better performance than using its original value.

• 융합정보논문지
• /
• 제11권10호
• /
• pp.302-313
• /
• 2021

국내 컨설팅산업이 성숙해지면서 컨설턴트에게는 고객의 비즈니스와 컨설팅 비즈니스에 대한 통찰력이 필수적으로 요구된다. 이러한 통찰력은 해당 비즈니스 도메인에 대한 깊은 이해와 고도의 역량을 필요로 하며, 날로 치열해지는 컨설팅 시장에서 지속적인 경쟁우위를 유지하기 위해 반드시 갖추어야 할 역량 요소로 평가된다. 본 연구는 소규모 컨설팅사에 소속되거나 개인적으로 활동하는 컨설턴트가 갖추어야 할 모든 요소를 담고 있는 ICMCI 역량모델을 국내 컨설턴트에게 적용해 보고자 하였으며, 설문조사로 수집된 데이터를 실증적으로 분석하였다. 연구결과, 비즈니스역량과 기술역량은 프로젝트성과에 긍정적인 영향을 미치는 것으로 나타났으나 가치 및 행동역량은 통계적으로 유의하지 않았다. 반면에, 사회관계역량에는 기술역량과 가치 및 행동역량만이 긍정적인 영향을 미치는 것으로 나타났다. 프로젝트성과와 사회관계역량에 미치는 역량의 요소와 영향의 크기가 정반대로 나타나 프로젝트성과가 우수한 컨설턴트는 사회관계를 소홀히 하고 프로젝트성과가 약한 컨설턴트는 사회관계에 더 집중한다는 의미로 해석된다. 본 연구를 통해 전문적인 컨설턴트로 성장하기 위해서는 컨설팅 비즈니스에 대한 깊은 이해와 지각이 필요하다는 것을 확인하였으나, 적은 표본으로 모집단의 특성을 충분히 반영하지 못해 연구결과를 일반화 하는데는 한계가 있다.

• 스마트미디어저널
• /
• 제10권1호
• /
• pp.99-107
• /
• 2021

4차 산업혁명, Industry 4.0 과 더불어 최근 ICT 분야의 메가트렌드는 빅데이터, IoT, 클라우드 컴퓨팅, 그리고 인공지능이라고 할 수 있다. 따라서, 4차 산업혁명 시대에 알맞은 AI 서비스들의 기술 개발과 다양한 산업 영역에서 ICT 분야의 융합에 따른 BI (Business Intelligence), IA (Intelligent Analytics, BI + AI), AIoT (Artificial Intelligence of Things), AIOPS (Artificial Intelligence for IT Operations), RPA 2.0 (Robotic Process Automation + AI) 등의 세분화된 기술 발전으로 급속한 디지털 전환 (Digital Transformation)이 진행되고 있는 추세이다. 본 연구에서는 이러한 기술적 상황에 따른 대용량 분산 Abyss 스토리지 기반으로 인프라 측면의 GPU, CDA (Connected Data Architecture) 프레임워크, 그리고 AI의 다양한 머신러닝 서비스들을 통합 및 고도화를 목표로 하며, AI 비즈니스의 수익 모델을 다양한 산업 영역에 활용하고자 한다.

• 한국컴퓨터정보학회논문지
• /
• 제28권3호
• /
• pp.35-43
• /
• 2023

운전 중 감정 인식은 사고를 예방하기 위해 꼭 필요한 과제이다. 더 나아가 자율 주행 시대에서 자동차는 모빌리티의 주체로 운전자와의 감정적인 소통이 더욱 요구되고 있으며 감정 인식 시장은 점점 확산되고 있다. 이에 따라 본 연구 방안에서는 수집하기 비교적 용이한 데이터인 심리데이터와 행동 데이터를 이용해 운전자의 감정을 분류하는 인공지능 모델을 개발하고자 한다. 오토인코더 모델을 통해 잠재 변수를 추출하고, 이를 본 분류 모델의 변수로 사용하였으며, 이는 성능 향상에 영향을 미침을 확인하였다. 또한 기존 뇌파 데이터를 포함했을 때 보다 본 논문이 제시하는 프레임워크를 사용하였을 때 성능이 향상됨도 확인하였다. 최종적으로 심리 및 개인정보데이터, 행동 데이터만을 통해 운전자의 감정 분류 정확도 81%와 F1-Score 80%를 달성하였다.

• 한국환경복원기술학회지
• /
• 제26권5호
• /
• pp.1-18
• /
• 2023

The Ecosystem Conservation Levy (formerly known as the Ecosystem Conservation Cooperation Fund) system has been in place for 20 years, and it can be said that it has now entered the settlement stage. Based on an analysis of publicly available project implementation data from 2014 to 2020, we found that: 1) As the number of return projects increases, the targets of restoration technologies are also strengthening, and it is necessary to frame a series of processes from application, creation, and monitoring for some detailed projects to improve the effectiveness and efficiency of utilizing the levy. 2) Most of the implemented projects are applied as micro-ecosystem creation, but there are many cases where the contents of the project can be seen as other project categories. This shows that the purpose of the return project needs to be approached more clearly and suggests that institutional complementation is needed from the project application stage. 3) The detailed technologies applied tend to be gradually expanding, but most of them are technologies that are not differentiated from general development projects. It is urgent to secure a more technical identity, such as a range and list of utilized technologies suitable for the characteristics and purposes of return projects. 4) It is necessary to establish a relevant evaluation system or framework to utilize the monitoring results of restoration projects. 5) There have been few cases of application of single restoration technologies since the beginning, but the content and scope of the complexity tend to expand in recent years. Even if the objectives are not comprehensive and diverse, it can be seen that many parts of the projects are oriented toward convergence, so it is necessary to conduct separate research on this. 6) As for the direction of improvement of the return project, it is possible to consider expanding the restoration and conservation focus to partially accommodate the complexity of the natural environment and human ecology. It seems that the expansion of restoration technologies that consider the role and function of humans in the natural environment should be explored.

• International journal of advanced smart convergence
• /
• 제9권4호
• /
• pp.16-25
• /
• 2020

The operation of a wastewater treatment plant (WWTP) is a complex task which requires to consider several aspects: adapting to always changing influent composition and volume, ensuring treated effluents quality complies with local regulations, ensuring dissolved oxygen levels in biological reaction tanks are sufficient to avoid anoxic conditions etc. all of it while minimizing usage of chemicals and power consumption. The traditional way of managing WWTPs consists in having employees on the field measure various parameters and make decisions based on their judgment and experience which holds various concerns such as the low frequency of data, errors in measurement and difficulty to analyze historical data to propose optimal solutions. In the case of activated sludge WWTPs, parts of the treatment process can be automated and controlled in order to satisfy various control objectives. The models developed by the International Water Association (IWA) have been extensively used worldwide in order to design and assess the performance of various control strategies. In this work, we propose to review most recent WWTP automation initiatives around the world and identify most currently used control parameters and control architectures. We then suggest a framework to select WWTP model, control parameters and control scheme in order to develop and benchmark control strategies for WWTP automation.

• 융합경영연구
• /
• 제8권2호
• /
• pp.1-12
• /
• 2020

Purpose: The rise of the phrase Work Life Balance was bought up in 1986 when amid many Americans there was prevalence of detrimental work place practices like neglecting families, leisure activities and friends in order to achieve their study place goals. The significance of work-life balance has been gaining ground in recent years to grasp a wider range of groups, including students. Searching and finding a balance can be complex and challenging for many individuals and students. Research design, data and methodology: Through this paper we will explore how students balance the competing demands of work, study, and social activities. Several factors have increased imbalances within Educational organizations, and technology specifically has been influential. However, technology also provides a novel solution to this organizational performance management issue. A Study-Life Optimization model (SLO) is suggested, which incorporates information systems, analytics, and decision support into a Smart Service System. A general framework for this model, detailing data collection, measurement, and ethical issues is explained briefly. Results: Outcomes include improved WLB, greater perceived quality of life, and increased Educational organizational performance. Conclusions: This paper contributes to the relevant literature as it pays attention to the various students' of varying lifestyles school-work-personal lives. Findings of this study will provide a meaningful of the Work/school-life balance issues faced by students. The research could be helpful to the various stakeholders of a University, the curriculum designers, program coordinators etc.

• International journal of advanced smart convergence
• /
• 제12권4호
• /
• pp.182-189
• /
• 2023

A recently research, object detection and segmentation have emerged as crucial technologies widely utilized in various fields such as autonomous driving systems, surveillance and image editing. This paper proposes a program that utilizes the QT framework to perform real-time object detection and precise instance segmentation by integrating YOLO(You Only Look Once) and Mask R CNN. This system provides users with a diverse image editing environment, offering features such as selecting specific modes, drawing masks, inspecting detailed image information and employing various image processing techniques, including those based on deep learning. The program advantage the efficiency of YOLO to enable fast and accurate object detection, providing information about bounding boxes. Additionally, it performs precise segmentation using the functionalities of Mask R CNN, allowing users to accurately distinguish and edit objects within images. The QT interface ensures an intuitive and user-friendly environment for program control and enhancing accessibility. Through experiments and evaluations, our proposed system has been demonstrated to be effective in various scenarios. This program provides convenience and powerful image processing and editing capabilities to both beginners and experts, smoothly integrating computer vision technology. This paper contributes to the growth of the computer vision application field and showing the potential to integrate various image processing algorithms on a user-friendly platform

• International journal of advanced smart convergence
• /
• 제4권2호
• /
• pp.109-119
• /
• 2015

The interaction between wireless sensors such as Internet of Things (IoT) and Cloud is a new paradigm of communication virtualization to overcome resource and efficiency restriction. Cloud computing provides unlimited platform, resources, services and also covers almost every area of computing. On the other hand, Wireless Sensor Networks (WSN) has gained attention for their potential supports and attractive solutions such as IoT, environment monitoring, healthcare, military, critical infrastructure monitoring, home and industrial automation, transportation, business, etc. Besides, our virtual groups and social networks are in main role of information sharing. However, this sensor network lacks resource, storage capacity and computational power along with extensibility, fault-tolerance, reliability and openness. These data are not available to community groups or cloud environment for general purpose research or utilization yet. If we reduce the gap between real and virtual world by adding this WSN driven data to cloud environment and virtual communities, then it can gain a remarkable attention from all over, along with giving us the benefit in various sectors. We have proposed a Pub/Sub-based sensor virtualization framework Cloud environment. This integration provides resource, service, and storage with sensor driven data to the community. We have virtualized physical sensors as virtual sensors on cloud computing, while this middleware and virtual sensors are provisioned automatically to end users whenever they required. Our architecture provides service to end users without being concerned about its implementation details. Furthermore, we have proposed an efficient content-based event matching algorithm to analyze subscriptions and to publish proper contents in a cost-effective manner. We have evaluated our algorithm which shows better performance while comparing to that of previously proposed algorithms.