• Journal of Multimedia Information System
• /
• 제6권4호
• /
• pp.235-238
• /
• 2019

In real world object detection is an active research topic for understanding different objects from images. There are different models presented in past and had significant results. In this paper we are presenting vehicle logo detection using previous object detection models such as You only look once (YOLO) and Faster Region-based CNN (F-RCNN). Both the front and rear view of the vehicles were used for training and testing the proposed method. Along with deep learning an image pre-processing algorithm called perspective transformation is proposed for all the test images. Using perspective transformation, the top view images were transformed into front view images. This algorithm has higher detection rate as compared to raw images. Furthermore, YOLO model has better result as compare to F-RCNN model.

• 로봇학회논문지
• /
• 제4권3호
• /
• pp.185-191
• /
• 2009

This paper presents a new shape-based algorithm using affine category shape model for object category recognition and model learning. Affine category shape model is a graph of interconnected nodes whose geometric interactions are modeled using pairwise potentials. In its learning phase, it can efficiently handle large pose variations of objects in training images by estimating 2-D homography transformation between the model and the training images. Since the pairwise potentials are defined on only relative geometric relationship betweenfeatures, the proposed matching algorithm is translation and in-plane rotation invariant and robust to affine transformation. We apply spectral matching algorithm to find feature correspondences, which are then used as initial correspondences for RANSAC algorithm. The 2-D homography transformation and the inlier correspondences which are consistent with this estimate can be efficiently estimated through RANSAC, and new correspondences also can be detected by using the estimated 2-D homography transformation. Experimental results on object category database show that the proposed algorithm is robust to pose variation of objects and provides good recognition performance.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제11권4호
• /
• pp.179-188
• /
• 2022

기계의 주요 부품인 베어링 결함 진단에 딥러닝을 활용하는 연구가 활발하게 진행되어 좋은 성능을 달성하였으나, 학습 데이터와 테스트 데이터의 운영 환경 차이로 인해 기계가 실제로 가동되는 환경에서는 성능 저하가 발생하는 문제가 있다. 학습 데이터와 테스트 데이터의 분포 차이 문제를 다루는 방법으로 데이터 적응이 제안되어 좋은 결과를 보여주고 있으나, 각 방법이 가정하고 있는 특정 적용 시나리오를 벗어나기 어렵다는 제약이 있다. 이에 본 연구는 MFCCs를 이용한 입력 데이터의 변환과 간단한 CNN 구조를 이용해 원시 도메인 데이터로부터 생성된 모델에 대해 추가적인 학습이나 조정 없이 타겟 도메인 데이터에 대한 테스트를 강건하게 수행하는 방법을 제안하였으며, 대표적인 베어링 결함 진단 데이터셋인 CWRU 베어링 데이터를 이용해 제안한 방법에 대한 실험 및 분석을 수행하였다. 실험 결과 전이 학습 기반의 방법들과 대등한 성능을 보였으며, 입력 변환 기반의 베이스라인 방법보다는 최소 15% 정도의 높은 성능을 달성하였다.

• 대한화학회지
• /
• 제68권4호
• /
• pp.205-220
• /
• 2024

본 연구는 초임과학교사학습공동체에서 자연 발생적이며 자발적으로 일어나는 교수학적 추론 과정의 특징을 파악하는 것을 목적으로 한다. 본 연구에서는 국내 A사범대학에서 화학교육을 전공하고 중등학교교사 임용후보자 선정경쟁시험에 합격하여 임용 첫해 중학교 2학년 과학을 가르치는 교사 세 명이 과학 교수-학습 계획을 함께 하려는 목적으로 자율적으로 구성하여 일 년 동안 운영한 교사학습공동체를 연구 대상으로 하였다. 본 연구에서는 교사학습공동체 34회 모임 중 월 1-2회 녹음 및 전 사록을 선정하여 총 11회 모임의 대화를 Shulman의 교수학적 추론 및 실행 모델 즉, 교과 이해, 교수-학습 계획, 수업, 평가, 반성, 새로운 이해의 과정에 기반하여 연역적 접근과 귀납적 접근을 모두 사용하여 분석하였다. 본 연구 교사학습공동체 교사들의 협력적 교수학적 추론에서는 교수-학습 계획 단계에서 준비, 표상, 수업전략, 조정 뿐 아니라 평가, 반성, 새로운 이해가 나타났다. 교수-학습 계획을 하는 과정에서 교사학습공동체 교사들의 협력적 반성은 준비, 표상, 수업전략, 조정, 평가 각 요소의 새로운 이해로 귀결되어 교수-학습 계획에 관한 논의를 활성화하는 것으로 드러났다.

• 아동학회지
• /
• 제32권2호
• /
• pp.71-85
• /
• 2011

This study was carried out in order to better understand how cooperative learning effects the geometric understanding of young children. The geometry tasks used in the study included the geometric relationship between two dimensional shapes and three dimensional shapes, coordination, symmetry and transformation visualization and spacial reasoning. The subjects were composed of children aged five years and were taken from two kindergartens in a relatively new city close to Seoul. The experimental group of children the comparative learning in geometry. The comparative group of children were enrolled in a kindergarten that uses an the intergrated curriculum. The results indicated that cooperative learning impacted positively on the children's understanding of geometry. The specific results are as follows : The scores that the experimental acquired were higher in terms of p < .001 level. than the scores of the comparative group studying the geometric relationships between two dimensional shapes and three dimensional shapes, coordination, symmetry and transformation visualization & spacial reasoning.

• Current Optics and Photonics
• /
• 제4권4호
• /
• pp.286-292
• /
• 2020

Imaging through multicore fiber (MCF) is of great significance in the biomedical domain. Although several techniques have been developed to image an object from a signal passing through MCF, these methods are strongly dependent on the surroundings, such as vibration and the temperature fluctuation of the fiber's environment. In this paper, we apply a new, strong technique called deep learning to reconstruct the phase image through a MCF in which each core is multimode. To evaluate the network, we employ the binary cross-entropy as the loss function of a convolutional neural network (CNN) with improved U-net structure. The high-quality reconstruction of input objects upon spatial light modulation (SLM) can be realized from the speckle patterns of intensity that contain the information about the objects. Moreover, we study the effect of MCF length on image recovery. It is shown that the shorter the fiber, the better the imaging quality. Based on our findings, MCF may have applications in fields such as endoscopic imaging and optical communication.

• 한국컴퓨터정보학회논문지
• /
• 제26권6호
• /
• pp.183-193
• /
• 2021

본 논문에서 저자는 교실 및 실험실 등에서 진행되던 전통적인 교육 방식이 온라인으로 바뀌게 됨으로 발생하는 문제점들을 찾아 분석하였다. 주요한 문제점으로 분석된 것들을 살펴보면, 첫째, 모든 환경과 시설이 인터넷으로 연결되지 못하였음을 포함하여, 다양한 기술적 문제가 있었다. 둘째, 갑작스럽게 온라인으로 전환되어 시행되는 가상 수업의 효과도 의심받을 수 있었다. 마지막으로, 새로운 환경에 직면하여, 새로운 교육 방법론에 빠른 속도로 적응해야 한다는 교사들의 스트레스가 문제였다. 저자는 이러한 문제점들을 해결하기 위한 방법으로 디지털 전환을 제안하였다. 저자는 디지털 변환이 가능하도록 하기 위한 교육 변화, 학습 양식 및 다양한 기술 도구, 그리고 다양한 과제에 대해 분석하였다. 먼저, 저자는 온라인으로 바뀌게 되는 교실 환경을 효율적으로 운영하기 위해 필요한 요소들을 조사하고 분석하여 제시하였다. 다음으로 저자는 학생들의 수업을 내실있고 효율적인 것으로 만들기 위해 필요한 요소 및 문제점들을 분석하였으며, 해결방법을 제안하였다. 마지막으로, 저자는 온라인 강의가 진행되면서 학습의 책임이 교사로부터 학생에게 과도하게 전가된다는 문제점을 지적하였으며, 이에 대한 해결 방법을 제안하였다. 이후 저자는 향후 연구를 제안하였다.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 1999년도 추계학술대회-지능형 정보기술과 미래조직 Information Technology and Future Organization
• /
• pp.323-335
• /
• 1999

In this paper, we propose a new hybrid model of artificial neural networks(ANNs) and genetic algorithm (GA) to optimal feature transformation and feature weighting. Previous research proposed several variants of hybrid ANNs and GA models including feature weighting, feature subset selection and network structure optimization. Among the vast majority of these studies, however, ANNs did not learn the patterns of data well, because they employed GA for simple use. In this study, we incorporate GA in a simultaneous manner to improve the learning and generalization ability of ANNs. In this study, GA plays role to optimize feature weighting and feature transformation simultaneously. Globally optimized feature weighting overcome the well-known limitations of gradient descent algorithm and globally optimized feature transformation also reduce the dimensionality of the feature space and eliminate irrelevant factors in modeling ANNs. By this procedure, we can improve the performance and enhance the generalisability of ANNs.

• 디지털융복합연구
• /
• 제6권2호
• /
• pp.135-143
• /
• 2008

The remarkable growth of the Internet since mid-l990s has expanded the e-learning market and brought the transformation of educational environments and methodology. It can be said that the e-learning has changed the educational paradigm. Korean government is firmly determined to support the diffusion of e-learning because of the benefits of e-learning. People seem to accept the e-learning when its contents have high quality. A lot of research have been conducted on e-learning, however, it was mostly about user's usage intention, satisfaction and educational effect. It can't seem that sufficient research efforts have been put into figuring out the role of e-learning contents quality in the expansion of e-learning. In this paper, we present the empirical study on the influence of e-learning contents quality on user's satisfaction and educational effect. We conducted an questionnaire survey on college students to collect data and found that the quality of e-learning contents has significant influence on the users' satisfaction.

• 지식경영연구
• /
• 제20권4호
• /
• pp.149-168
• /
• 2019

In the era of the Fourth Industrial Revolution, digital transformation is emerging in the education and learning fields. As the use of the mobile Internet and mobile devices has become a daily life, mobile learning that supports a variety of learning in a mobile environment is drawing attention. Mobile learning applications (apps) are expected to expand their use by providing a convenient learning environment anytime, anywhere. This study investigates the use of mobile learning apps in English education, which is one of the most popular learning areas, and empirically examines the factors that influence the continuous use of mobile learning apps. In particular, it analyzes the differences between Chinese and Korean users. The results of this study provide theoretical and practical implications to promote the development of mobile apps suitable for mobile learning environments and the sustainable user growth in mobile learning.