• 제목/요약/키워드: Visual Intelligence

검색결과 245건 처리시간 0.022초

Bioimage Analyses Using Artificial Intelligence and Future Ecological Research and Education Prospects: A Case Study of the Cichlid Fishes from Lake Malawi Using Deep Learning

  • Joo, Deokjin;You, Jungmin;Won, Yong-Jin
    • Proceedings of the National Institute of Ecology of the Republic of Korea
    • /
    • 제3권2호
    • /
    • pp.67-72
    • /
    • 2022
  • Ecological research relies on the interpretation of large amounts of visual data obtained from extensive wildlife surveys, but such large-scale image interpretation is costly and time-consuming. Using an artificial intelligence (AI) machine learning model, especially convolution neural networks (CNN), it is possible to streamline these manual tasks on image information and to protect wildlife and record and predict behavior. Ecological research using deep-learning-based object recognition technology includes various research purposes such as identifying, detecting, and identifying species of wild animals, and identification of the location of poachers in real-time. These advances in the application of AI technology can enable efficient management of endangered wildlife, animal detection in various environments, and real-time analysis of image information collected by unmanned aerial vehicles. Furthermore, the need for school education and social use on biodiversity and environmental issues using AI is raised. School education and citizen science related to ecological activities using AI technology can enhance environmental awareness, and strengthen more knowledge and problem-solving skills in science and research processes. Under these prospects, in this paper, we compare the results of our early 2013 study, which automatically identified African cichlid fish species using photographic data of them, with the results of reanalysis by CNN deep learning method. By using PyTorch and PyTorch Lightning frameworks, we achieve an accuracy of 82.54% and an F1-score of 0.77 with minimal programming and data preprocessing effort. This is a significant improvement over the previous our machine learning methods, which required heavy feature engineering costs and had 78% accuracy.

CNN 딥러닝을 활용한 경관 이미지 분석 방법 평가 - 힐링장소를 대상으로 - (Assessment of Visual Landscape Image Analysis Method Using CNN Deep Learning - Focused on Healing Place -)

  • 성정한;이경진
    • 한국조경학회지
    • /
    • 제51권3호
    • /
    • pp.166-178
    • /
    • 2023
  • 본 연구는 이용자들의 인식과 경험이 내재된 소셜미디어 사진에서 경관 이미지를 분석하기 위한 방법으로 CNN 딥러닝 방법을 소개하고 평가하는 데 그 목적이 있다. 본 연구에서는 힐링장소를 연구의 대상으로 설정하여 경관 이미지를 분석하였다. 연구를 위해 텍스트마이닝과 선행연구 고찰을 통해 힐링과 관련되는 7가지의 경관 형용사를 선정하였다. 이후 CNN 딥러닝 학습 사진 구축을 위해 50명의 평가자를 모집하였으며, 평가자들에게 포털사이트에서 '힐링', '힐링풍경', '힐링장소'로 검색되는 사진 중 7가지 형용사마다 가장 적합한 사진을 3장씩 수집하도록 하였다. 수집된 사진을 정제 및 데이터 증강 과정을 거쳐 CNN 모델을 제작하였다. 이후 힐링장소 경관 분석을 위해 포털사이트에서 '힐링'과 '힐링풍경'으로 검색되는 15,097장의 사진을 수집하여 이를 분류하였다. 연구결과 '기타'와 '실내'를 제외한 범주에서 '조용한'이 2,093장(22%)으로 가장 높게 나타났으며, '개방적인', '즐거운', '안락한', '깨끗한', '자연적인', '아름다운' 순으로 나타났다. CNN 딥러닝은 경관 이미지 분석에서도 결과를 도출 가능한 분석 방법임을 연구를 통해 알 수 있었다. 또한, 기존 경관 분석 방법을 보완할 수 있는 하나의 방법임을 시사하였고, 경관 이미지 학습 데이터 셋 구축을 통한 향후 심층적이고 다양한 경관 분석을 제안한다.

Recent update on reading disability (dyslexia) focused on neurobiology

  • Kim, Sung Koo
    • Clinical and Experimental Pediatrics
    • /
    • 제64권10호
    • /
    • pp.497-503
    • /
    • 2021
  • Reading disability (dyslexia) refers to an unexpected difficulty with reading for an individual who has the intelligence to be a much better reader. Dyslexia is most commonly caused by a difficulty in phonological processing (the appreciation of the individual sounds of spoken language), which affects the ability of an individual to speak, read, and spell. In this paper, I describe reading disabilities by focusing on their underlying neurobiological mechanisms. Neurobiological studies using functional brain imaging have uncovered the reading pathways, brain regions involved in reading, and neurobiological abnormalities of dyslexia. The reading pathway is in the order of visual analysis, letter recognition, word recognition, meaning (semantics), phonological processing, and speech production. According to functional neuroimaging studies, the important areas of the brain related to reading include the inferior frontal cortex (Broca's area), the midtemporal lobe region, the inferior parieto-temporal area, and the left occipitotemporal region (visual word form area). Interventions for dyslexia can affect reading ability by causing changes in brain function and structure. An accurate diagnosis and timely specialized intervention are important in children with dyslexia. In cases in which national infant development screening tests have been conducted, as in Korea, if language developmental delay and early predictors of dyslexia are detected, careful observation of the progression to dyslexia and early intervention should be made.

VR 야구 게임의 현실감 강화 방법 연구 (A Study on Reality Enhancement Method of VR Baseball Game)

  • 유왕윤
    • 한국게임학회 논문지
    • /
    • 제19권2호
    • /
    • pp.23-32
    • /
    • 2019
  • VR 콘텐츠의 대중화가 더딘 것은 시각적인 새로운 경험 즉, '흥미' 이상의 '효용'을 만들어내지 못했기 때문이다. 가상현실 콘텐츠의 효용은 기능적 현실감에서 출발하며 그것을 증진시키기 위해서 사실적인 인터랙션이 요구된다. 본 연구는 구체적으로 네트워크 플레이, 캐릭터 인공지능, 햅틱 구현의 3가지 방법을 제시하고 있다. 가설을 확인하기 위하여 기획에서부터 콘텐츠 제작, 플레이 테스트, 기술 검증까지 야구를 소재로 한 VR 콘텐츠 제작의 전 단계를 수행하였다. 최종 결과물에 대한 사용자 및 평가 기관의 테스트를 통하여 사실적인 시각 효과와 플레이 연출, 진동에 의한 타격감까지 콘텐츠의 현실감을 높이는 데 기여한 것으로 평가되었다.

Research on Technology Production in Chinese Virtual Character Industry

  • Pan, Yang;Kim, KiHong;Yan, JiHui
    • International Journal of Internet, Broadcasting and Communication
    • /
    • 제14권4호
    • /
    • pp.64-79
    • /
    • 2022
  • The concept of Virtual Character has been developed for a long time with people's demand for cultural and entertainment products such as games, animations, and movies. In recent years, with the rapid development of concepts and industries such as social media, self-media, web3.0, artificial intelligence, virtual reality, and Metaverse, Virtual Character has also expanded new derivative concepts such as Virtual Idol, Virtual YouTuber, and Virtual Digital Human. With the development of technology, people's life is gradually moving towards digitalization and virtualization. At the same time, under the global environment of the new crown epidemic, human social activities are rapidly developing in the direction of network society and online society. From the perspective of digital media content, this paper studies the production technology of Virtual Character related products in the Chinese market, and analyzes the future development direction and possibility of the Virtual Character industry in combination with new media development directions and technical production methods. Consider and provide reference for the development of combined applications of digital media content industry, Virtual Character and Metaverse industry.

시각적 특징과 머신 러닝으로 악성 URL 구분: HTTPS의 역할 (Malicious URL Detection by Visual Characteristics with Machine Learning: Roles of HTTPS)

  • Sung-Won HONG;Min-Soo KANG
    • Journal of Korea Artificial Intelligence Association
    • /
    • 제1권2호
    • /
    • pp.1-6
    • /
    • 2023
  • In this paper, we present a new method for classifying malicious URLs to reduce cases of learning difficulties due to unfamiliar and difficult terms related to information protection. This study plans to extract only visually distinguishable features within the URL structure and compare them through map learning algorithms, and to compare the contribution values of the best map learning algorithm methods to extract features that have the most impact on classifying malicious URLs. As research data, Kaggle used data that classified 7,046 malicious URLs and 7.046 normal URLs. As a result of the study, among the three supervised learning algorithms used (Decision Tree, Support Vector Machine, and Logistic Regression), the Decision Tree algorithm showed the best performance with 83% accuracy, 83.1% F1-score and 83.6% Recall values. It was confirmed that the contribution value of https is the highest among whether to use https, sub domain, and prefix and suffix, which can be visually distinguished through the feature contribution of Decision Tree. Although it has been difficult to learn unfamiliar and difficult terms so far, this study will be able to provide an intuitive judgment method without explanation of the terms and prove its usefulness in the field of malicious URL detection.

Multimodal audiovisual speech recognition architecture using a three-feature multi-fusion method for noise-robust systems

  • Sanghun Jeon;Jieun Lee;Dohyeon Yeo;Yong-Ju Lee;SeungJun Kim
    • ETRI Journal
    • /
    • 제46권1호
    • /
    • pp.22-34
    • /
    • 2024
  • Exposure to varied noisy environments impairs the recognition performance of artificial intelligence-based speech recognition technologies. Degraded-performance services can be utilized as limited systems that assure good performance in certain environments, but impair the general quality of speech recognition services. This study introduces an audiovisual speech recognition (AVSR) model robust to various noise settings, mimicking human dialogue recognition elements. The model converts word embeddings and log-Mel spectrograms into feature vectors for audio recognition. A dense spatial-temporal convolutional neural network model extracts features from log-Mel spectrograms, transformed for visual-based recognition. This approach exhibits improved aural and visual recognition capabilities. We assess the signal-to-noise ratio in nine synthesized noise environments, with the proposed model exhibiting lower average error rates. The error rate for the AVSR model using a three-feature multi-fusion method is 1.711%, compared to the general 3.939% rate. This model is applicable in noise-affected environments owing to its enhanced stability and recognition rate.

위모트를 활용한 시지각 장애아동 교육 콘텐츠개발 (Development of an Edutainment Contents using Wiimote Controller for Children with Visual Perception Disabilities)

  • 유상조;한경임;김봉석;박동규
    • 한국멀티미디어학회논문지
    • /
    • 제13권10호
    • /
    • pp.1547-1556
    • /
    • 2010
  • 현재까지 유아나 장애인을 위한 컴퓨터 활용 교육 콘텐츠는 지각훈련, 인지훈련, 한글 교육 등 다양한 분야에서 개발되었으나 가장 큰 문제점은 컴퓨터 모니터 앞에서 장시간 마우스를 이용하여 교육을 할 경우 활동성이 저하된다는 점이다. 이것은 특히 왕성하게 운동 능력이 발달하는 시기의 유아와 운동 장애로 인해 활동 기회가 부족한 장애 아동에게 적잖은 문제점으로 지적되어 왔다. 이와 같은 문제점을 개선하고 활동성과 협동력, 몰입성을 강화시키는 콘텐츠를 개발하기 위해서는 터치스크린과 같은 스크린 상에서 인간의 동작을 인식하여 이를 대화식으로 보여주는 기술이 필요하다 본 연구에서는 기존의 컴퓨터 활용 콘텐츠의 단점을 보완하고, 사용자의 활동성을 강화하기 위해 위모트가 가지는 센서 기술을 활용하여 실시간으로 빔 프로젝터나 컴퓨터 스크린으로 교육콘텐츠를 제공하고 신체를 직접 움직이며 적외선 펜을 사용하여 자극에 반응하는 교육 콘텐츠를 개발하였다.

빅데이터를 활용한 영상콘텐츠 스토리 리모델링 프로세스 개발 (The Development of Remodeling Process for Visual Content's Story by Big Data)

  • 이혜원;박성원;김이경
    • Journal of Information Technology Applications and Management
    • /
    • 제26권3호
    • /
    • pp.121-134
    • /
    • 2019
  • The Fourth Industrial Revolution has differentiated technologies such as artificial intelligence, IoT(Internet of things), big data, and mobile. As the civilization develops more and more, humanity enjoy the cultural activities more than economic activity for the food and shelter. The platform structure based on the advanced information technology of the present will expand the cultural contents area in a variety of ways. Cultural contents respond sensitively to changes in consumer and will be useful experiences of human activities. Therefore, it should be noted again that the contents industry should not be limited to the discussion of the application of the fourth technology, but should be produced with emphasis on useful experiences of human being. In other words, the discussion of human activities around cultural contents should be focused on how to apply beyond the use of fourth industrial technology. Therefore, it is necessary to analyze the basis of the successful storytelling of the planning stage to connect the fourth industrial technology and human useful experience as a method for developing cultural contents, and to build and propose a model as a strategic method. This study analyzes domestic and foreign cases made by using big data among the visual contents which show continuous increase of consumption among culture industry field, and draws success factors and limit points. Next, we extract what is the successful matching factor that influenced consumer 's consciousness, and find out that the structure of culture prototype has been applied in the long history of mankind, and presents it as a storytelling model. Through the above research, this study aims to present a new interpretation and creative activity of cultural contents by presenting a storytelling model as a methodology for connecting creative knowledge, away from the general interpretation of social phenomenon applied with big data.

HW/SW Co-design of a Visual Driver Drowsiness Detection System

  • Yu, Tian;Zhai, Yujia
    • 중소기업융합학회논문지
    • /
    • 제4권1호
    • /
    • pp.31-39
    • /
    • 2014
  • PID 오토 튜닝 컨트롤러는 퍼지 논리를 통해 설계되었다. 이러한 오류 및 오류 파생 의견으로 일반적인 값은 발견적 표현으로 변경, 그들은 퍼지 및 defuzzification 과정을 통해 PID 이득을 결정했다. 퍼지 절차 및 PID 제어기 설계는 개별적으로 간주하고, 그것들을 혼합하고, 분석 하였다. 퍼지 논리에 의해 획득 자동 조정 PID 컨트롤러는 3 차 플랜트 제어 이하의 능력을 보여 주었다. 또한 설계된 자동 동조 방식으로 추적 문제를 참조하는 데 적용한다.

  • PDF