• Fisheries and Aquatic Sciences
• /
• 제25권3호
• /
• pp.151-157
• /
• 2022

Various approaches have been applied to transform aquaculture from a manual, labour-intensive industry to one dependent on automation technologies in the era of the fourth industrial revolution. Technologies associated with the monitoring of physical condition have successfully been applied in most aquafarm facilities; however, real-time biological monitoring systems that can observe fish condition and behaviour are still required. In this study, we used a video recorder placed on top of a fish tank to observe the swimming patterns of rock bream (Oplegnathus fasciatus), first one fish alone and then a group of five fish. Rock bream in the video samples were successfully identified using the you-only-look-once v3 algorithm, which is based on the Darknet-53 convolutional neural network. In addition to recordings of swimming behaviour under normal conditions, the swimming patterns of fish under abnormal conditions were recorded on adding an anaesthetic or lowering the salinity. The abnormal conditions led to changes in the velocity of movement (3.8 ± 0.6 cm/s) involving an initial rapid increase in speed (up to 16.5 ± 3.0 cm/s, upon 2-phenoxyethanol treatment) before the fish stopped moving, as well as changing from swimming upright to dying lying on their sides. Machine learning was applied to datasets consisting of normal or abnormal behaviour patterns, to evaluate the fish behaviour. The proposed algorithm showed a high accuracy (98.1%) in discriminating normal and abnormal rock bream behaviour. We conclude that artificial intelligence-based detection of abnormal behaviour can be applied to develop an automatic bio-management system for use in the aquaculture industry.

• 전기전자학회논문지
• /
• 제26권1호
• /
• pp.62-66
• /
• 2022

최근 데이터 기반의 딥러닝 기술을 적용하여 비면허 대역의 다양한 통신 신호를 분류하는 연구가 활발히 수행되고 있다. 하지만, 복잡한 신경망 모델 사용을 기반으로 이뤄진 이러한 접근법은 높은 연산 능력을 필요로 하게 되어, 자원 제약적인 무선 인터페이스 및 사물인터넷(Internet of Things) 장비에서는 사용이 제약된다. 본 연구에서는 비면허 대역의 무선 이기종 기술을 인지하기 위한 데이터 기반의 접근 방법을 살펴보고, 신호의 특징 추출 및 데이터화의 효율화 문제를 다룬다. 구체적으로, 비면허 대역의 다른 종류의 무선 통신 기술을 구분하기 위해 수신 신호 강도 측정을 기반으로 한 시계열 데이터를 이용해 합성곱 신경망(Convolutional Neural Network, CNN) 모델을 학습시켜 신호를 분류하는 방법을 살펴본다. 이 과정에서 동일한 구조의 신경망 모델의 경량화를 위한 효율적 신호의 시계열 데이터 정보 수집시 주파수 대역의 특징을 함께 특징화하는 방법을 제안하고, 그 효과를 평가한다. Bluetooth 호환의 Ubertooth 장비를 이용한 실측 기반의 실험 결과는 제안된 샘플링 기법이 동일한 신경망에 대해서 10% 수준의 샘플링 데이터 이용만으로도 동일한 정확도를 유지함을 보여준다.

• International Journal of Computer Science & Network Security
• /
• 제22권9호
• /
• pp.123-130
• /
• 2022

Modernization of the educational sector requires globalization, democratization, and the transition to an information technology society. The main goal of education at the present stage is to solve the problem of ensuring the priority of the development of education and science. In modern conditions, the quality of training of qualified specialists is becoming particularly relevant. The great role of teacher education is emphasized by its main goal, which is to train specialists who can ensure the versatile and innovative development of a person as a person and the highest value of society, its mental, physical and aesthetic abilities, high moral qualities, and, consequently, the enrichment on this basis of the intellectual, creative and cultural potential of the people. Among the strategic tasks of modernizing higher education is to ensure informatization of the educational process and access to International Information Systems. The essence of the concept of multimedia is clarified. In the context of media education, multimedia lists a number of functions: informational, interpretive, cultural, entertainment, and educational. The need to meet the needs outlined in the article in the conditions of informatization of the educational process requires the teacher to have knowledge and skills in the field of multimedia pedagogical technologies, knowledge of advanced methods and means of modern science. It is considered what relevant concepts of media education have been developed and are being developed in Ukraine and form an important basis for the modernization of education, which will contribute to the construction of an information society in the country and the formation of civil society. Distance learning is considered - the most democratic form of education that allows broad segments of society to get an education. Distance learning methods are used in higher education institutions, in school education, in the system of advanced training of teachers, in the system of training managerial personnel.

• International Journal of Computer Science & Network Security
• /
• 제22권9호
• /
• pp.251-257
• /
• 2022

Nowadays, the issues of comprehensive formation of a person capable of self-education, self-development and creative self-realization in the conditions of distance education are relevant. There is a need to solve this problem, which is due to social, cultural, and pedagogical factors. This makes it necessary to find effective means of personality formation. In this matter, great importance is attached to the modern method of forming a creative personality - art therapy. Various approaches to the definition of art therapy have been clarified. They consider various forms of art therapy when working with children, adolescents and adults in the context of distance education. The most relevant are the two main forms of work - individual and group art therapy. Art therapy develops the individual's creativity. Therefore, during art therapy, attention is focused on the inner world, experiences, and feelings. Therefore, we believe that in the context of distance education, art therapy has everything for the powerful potential of personality formation. Scientists consider this therapy as therapy by means of art, which is based on experiences, conflicts that can be expressed in the visual arts and music. Art therapy helps to get rid of conflicts and experiences. This happens in the context of distance education through the development of attention to feelings, strengthening one's own personal value and increasing artistic competence. The article describes the signs that characterize art therapy. Art-therapeutic technologies in the context of distance education, which are now actively used by psychologists, teachers and art therapists themselves, are highlighted. The advantages of distance learning are considered. The characteristic features of distance learning and features of the use of art therapy by means of distance education in the process of professional training of specialists are determined.

• International Journal of Computer Science & Network Security
• /
• 제22권9호
• /
• pp.95-102
• /
• 2022

The article considers the fundamentalization of education in improving the future specialists professional training with usage of multimedia technologies by various scientists. Various points of view and approaches to defining the concepts of fundamentalization of education and multimedia technologies are identified. The concept of fundamentalization of professional training of a future specialist is based on the goals and functions of fundamentalization and - on the ways and means of achieving it, etc. Most authors agree only in their views that the fundamentalization of education is aimed at improving the quality of education and the education of the individual. Others involve the formation of a culture and worldview, increasing the creative and intellectual potential, forming the professional competence of a specialist and the potential for further education, and so on. The term multimedia refers to interactive systems that provide processing of moving and still video images, animated graphics, high-quality audio and speech. It is found out that professional training of a specialist by means of multimedia technologies includes not only the activities of the teacher and student, which form the learning process, but also the independent activity of the subject, self-development, assimilation of experience by the subject through analysis, comprehension and transformation of the field of activity in which he is included. It is revealed through the implementation of which approaches to the fundamentalization of higher professional education, it becomes possible to fully present theoretical training courses and effectively pass practical training by students, which contributes to improving the quality of training of future specialists in higher education institutions. Theoretical analysis of scientific views indicates a fairly serious attention of scientists to the problem of professional readiness of specialists and the possibility of higher educational institutions in preparing for it. At the same time, professional readiness is considered from different positions: as an active state of a person, which manifests itself in activity; as a result of activity; as goals of activity; as a quality that characterizes the attitude to solving professional problems and social situations; as a prerequisite for purposeful activity; as a form of activity of the subject; as an integral formation of personality; as a component of socio-professional culture; as a complex professionally significant neoplasm of the individual.

• 전자통신동향분석
• /
• 제39권4호
• /
• pp.54-62
• /
• 2024

The emergence and resurgence of novel respiratory infectious diseases since the turn of the millennium, including SARS, H1N1 flu, MERS, and COVID-19, have posed a significant global health threat. Efforts to combat these threats have involved various approaches, however, continued research and development are crucial to prepare for the possibility of emerging viruses and viral variants. Direct detection methods for viral pathogens include molecular diagnostic techniques and immunodiagnostic methods, while indirect diagnostic methods involve detecting changes in the condition of infected patients through imaging diagnostics, gas analysis, and biosignal measurement. Molecular diagnostic techniques, utilizing advanced technologies such as gene editing, are being developed to enable faster detection than traditional PCR methods, and research is underway to improve the efficiency of diagnostic devices. Diagnostic technologies for infectious diseases continue to evolve, and several key trends are expected to emerge in the future. Automation will facilitate widespread adoption of rapid and accurate diagnostics, portable diagnostic devices will enable immediate on-site diagnosis by healthcare professionals, and advancements in AI-based deep learning diagnostic models will enhance diagnostic accuracy.

• 한국멀티미디어학회논문지
• /
• 제19권9호
• /
• pp.1647-1658
• /
• 2016

RGB+D database has been widely used in object recognition, object tracking, robot control, to name a few. While rapid advance of active depth sensing technologies allows for the widespread of indoor RGB+D databases, there are only few outdoor RGB+D databases largely due to an inherent limitation of active depth cameras. In this paper, we propose a novel method used to build outdoor RGB+D databases. Instead of using active depth cameras such as Kinect or LIDAR, we acquire a pair of stereo image using high-resolution stereo camera and then obtain a depth map by applying stereo matching algorithm. To deal with estimation errors that inevitably exist in the depth map obtained from stereo matching methods, we develop an approach that estimates confidence of depth maps based on unsupervised learning. Unlike existing confidence estimation approaches, we explicitly consider a spatial correlation that may exist in the confidence map. Specifically, we focus on refining confidence feature with the assumption that the confidence feature and resultant confidence map are smoothly-varying in spatial domain and are highly correlated to each other. Experimental result shows that the proposed method outperforms existing confidence measure based approaches in various benchmark dataset.

• International Journal of Computer Science & Network Security
• /
• 제21권6호
• /
• pp.258-268
• /
• 2021

Online learning systems are becoming an effective educational medium for many universities. The accessibility of online learning system in universities means that every student, including the visually impaired, is able use all the site's services. This research focuses on investigating the accessibility of online learning systems for visually impaired users. The paper purpose is to understand the perception of visually impaired undergraduate students towards Blackboard's accessibility and to make recommendations for a new Blackboard design with accessible features that support their needs. Impact of a new Blackboard design with accessible features on visually impaired students, using Taif University students as a case study is evaluated in this paper, as it is similar to most learning systems used by Saudi universities. A study on Taif University's utilization of Blackboard was conducted using mixed method approaches (an automatic tool and a user study). In the first phase, Taif's use of Blackboard was evaluated by the web accessibility tool called AChecker. In the second phase, we conducted a user study to verify previously discovered accessibility challenges to fully assess them according to the accessibility and usability guidelines. In this study, the accessibility of Taif University's Blackboard was evaluated by thirteen visually impaired undergraduate students. The results of the study show that Blackboard has accessibility issues, which are confusing navigation, incompatibility with assistive technologies, untitled pages or parts, unclear identification for visual elements, and inaccessible PDF files. This paper also introduces a set of recommendations that aim to improve the accessibility of Blackboard and other educational websites developed for this population. It also highlights the serious need for universities to enhance web accessibility for online learning systems for students with disabilities.

• 대한원격탐사학회지
• /
• 제38권6_2호
• /
• pp.1737-1741
• /
• 2022

고해상도 드론영상과 딥러닝 기술을 결합한 활성산불 감시는 이제 초기단계로 다방면의 연구개발을 필요로 한다. 이 단보에서는 드론영상 산불탐지에 아직 사용되지 않았던 state-of-the-art (SOTA) 모델인 You Only Look Once Version 7 (YOLOv7) 기반의 활성산불 객체탐지를 수행하였으며, 동일한 데이터셋을 사용한 선행연구에 비해 F1점수가 약 0.05 향상된 성과를 거두었다. 향후 우리나라에서도 광역적인 산불감시에 적용될 수 있도록 추가적인 기술 개발이 계속 필요할 것이다.

• 지능정보연구
• /
• 제18권3호
• /
• pp.79-96
• /
• 2012

기술의 발전과 융합이 빠르게 이루어지고 있는 오늘날 유망기술을 어떻게 파악하여, 다양한 후보군들 중에서 최적의 R&D 대상을 어떻게 선정할 것인가에 대한 문제는 주요한 경영의사결정문제 중 하나로 부상하고 있다. 본 연구에서는 이러한 R&D 기술 선정 의사결정을 지원할 수 있는 새로운 지능형 의사결정지원시스템을 제안한다. 본 연구의 의사결정지원시스템은 크게 3가지 모듈로 구성되는데, 우선 첫 번째 모듈인 '기술가치 평가' 모듈에서는 기업이 관심을 갖고 있는 분야의 특허들을 분석하여 유망기술 파악에 요구되는 다양한 차원의 기술가치 평가지수 값들을 산출하는 작업이 이루어진다. 이를 통해, 현재 시점에서의 각 기술의 가치가 다양한 차원에서 평가가 이루어지고 나면, 두 번째 모듈인 '미래기술가치 예측' 모듈에서 이들의 시간 흐름에 따른 변화를 학습한 인공지능 모형을 토대로 각 후보기술들이 미래 시점에 어떤 가치지수값을 갖게 될 것인지 예측값을 산출하게 된다. 마지막 세 번째 모듈인 '최적 R&D 대상기술 선정 지원' 모듈에서는 앞서 두 번째 모듈에서 산출된 각 차원별 예상 가치지수값들을 적절히 가중합하여 기술의 종합적인 미래가치 예측값을 산출하여 의사결정자에게 제공하는 기능을 수행한다. 이를 통해 의사결정자가 자사에 적합한 최적의 R&D 대상기술을 선정할 수 있도록 하였다. 본 연구에서는 제안된 시스템의 적용 가능성을 검증하기 위해, 10년치 특허데이터에 인공신경망 기법을 적용하여 실제 기술가치 예측모형을 구축해 보고, 그 효과를 살펴본다.