• Journal of Broadcast Engineering
• /
• v.27 no.5
• /
• pp.794-807
• /
• 2022

Although registration of dogs is mandatory according to the revision of the Animal Protection Act, the registration rate is low due to the inconvenience of the current registration method. In this paper, a performance improvement study was conducted on the dog face recognition technology, which is being reviewed as a new registration method. Through deep learning learning, an embedding vector for facial recognition of a dog was created and a method for identifying each dog individual was experimented. We built a dog image dataset for deep learning learning and experimented with InceptionNet and ResNet-50 as backbone networks. It was learned by the triplet loss method, and the experiments were divided into face verification and face recognition. In the ResNet-50-based model, it was possible to obtain the best facial verification performance of 93.46%, and in the face recognition test, the highest performance of 91.44% was obtained in rank-5, respectively. The experimental methods and results presented in this paper can be used in various fields, such as checking whether a dog is registered or not, and checking an object at a dog access facility.

• Journal of Internet Computing and Services
• /
• v.23 no.4
• /
• pp.21-33
• /
• 2022

Recently, the recommender system has been widely used in various fields such as movies, music, online shopping, and social media, and in the meantime, the recommender model has been developed from correlation analysis through the Apriori model, which can be said to be the first-generation model in the recommender system field. In 2005, many models have been proposed, including deep learning-based models, which are receiving a lot of attention within the recommender model. The recommender model can be classified into a collaborative filtering method, a content-based method, and a hybrid method that uses these two methods integrally. However, these basic methods are gradually losing their status as methodologies in the field as they fail to adapt to internal and external changing factors such as the rapidly changing user-item interaction and the development of big data. On the other hand, the importance of deep learning methodologies in recommender systems is increasing because of its advantages such as nonlinear transformation, representation learning, sequence modeling, and flexibility. In this paper, among deep learning methodologies, RNN, CNN, and GAN-based models suitable for sequential modeling that can accurately and flexibly analyze user-item interactions are classified, compared, and analyzed.

• Journal of Korean Home Economics Education Association
• /
• v.32 no.4
• /
• pp.149-170
• /
• 2020

The purpose of this study was to develop and evaluate a Smart learning-based middle school home economics education plan to improve the online home economics education classes. The educational plan in this study was completed through the process of analysis, design, development, and evaluation. The results of this study are as follows. First, as a result of analyzing consumer life units in the middle school textbooks based on 2015-revised curriculum, Smart learning activities were presented in only two out of the 12 textbooks analyxed. Second, a Smart learning-based middle school home economics education plan was developed in this study with the following characteristics: the topics and contents are structured so that to help learners actively engage in the teaching and learning activities; the education plan to reflects various media and current issues that learners may be interested in; the lesson plans were structured with the premise of online classes; softwares that enable real-time discussion and collaboration are used; and the evaluation method are composed of online activities. Third, the expert evaluation scores for the educational plan and activity materials developed were 4.52 (5-point Likert scale), when averaged across subject, goal, content, teaching/learning activity, and evaluation, and the overall content validity index(CVI) was 0.95. The adequacy of execution, benefit, attractiveness, usefulness, and feasibility were highly with an average of 4.62. Based on the experts' comments, the education plan and activity materials were revised and completed. This study is meaningful in that it developed teaching and learning activities based on online classes after the COVID-19 outbreak, overcoming the limitations of offline classes. It has implications for face-to-face home economics classes due to COVID-19, as it suggests ways to blend online and offline teaching/learning activities depending on the situation.

• English Language & Literature Teaching
• /
• no.6
• /
• pp.1-24
• /
• 2000

Listening comprehension is an integrative and creative process of interaction through which listeners receive speakers' production of linguistic or non-linguistic knowledge. Compared with reading comprehension, it may arouse difficulties and thus impose more burdens on foreign learners. The Audio-Lingual Method focused primarily on speaking. Mimicry, repetition, rote memory, and transformation drills actually interfered with listening comprehension. So learners lost interest and were not highly motivated. Improving listening comprehension requires continual attentiveness and interest. Listening skill can be extended systematically only when students are frequently exposed to a wide range of listening materials with an affective, cultural, social, and psycholinguistic approach. Therefore, teachers should help students learn how to comprehend intactly the overall meaning of intended messages. The literature on teaching listening skill suggests various useful activities: TPR, dictation, role playing, singing, picture recognition, completion, prediction, seeking specific information, summarizing, labeling, humor, jokes, cartoons, media, and so on. Practical classroom teaching necessitates a systematic procedure in which students should take part in meaningful tasks/activities. In addition to this, learners must practice listening comprehension trough a self-study process.

• Geomechanics and Engineering
• /
• v.9 no.4
• /
• pp.465-481
• /
• 2015

Three-dimensional simulation of flow through dam foundation is performed using finite element (Seep3D model) and artificial neural network (ANN) models. The governing and discretized equation for seepage is obtained using the Galerkin method in heterogeneous and anisotropic porous media. The ANN is a feedforward four layer network employing the sigmoid function as an activator and the back-propagation algorithm for the network learning, using the water level elevations of the upstream and downstream of the dam, as input variables and the piezometric heads as the target outputs. The obtained results are compared with the piezometric data of Shahid Abbaspour's Dam. Both calculated data show a good agreement with available measurements that demonstrate the effectiveness and accuracy of purposed methods.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.237-239
• /
• 2019

최근 인터넷 영상 매체가 발전하고 대중화되며 이를 통한 광고 효과가 커지고 있다. 이들 영상에 관련된 광고를 자동으로 연결할 수 있다면 효과적일 것이다. 본 논문은 딥러닝 검출 툴을 적용한 영상 카테고리 분류 기법을 제안한다. 이 기법은 주어진 영상을 몇 가지 카테고리로 분류하고, 분류 정보를 바탕으로 관련성이 높은 광고를 연결지어, 결과적으로 영상 시청자에게 맞춤형 광고를 제시한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.235-236
• /
• 2019

본 논문에서는 초해상도, 압축 열화 제거 등 영상 화질 복원 연구에서 영상의 다운샘플링에 딥러닝을 적용한 연구들에 대해 소개한다. 첫 번째 연구는 두 개의 컨볼루셔널 신경망과 영상 압축 코덱을 이용하여 압축 영상의 화질을 향상시켰다. 두 번째 연구는 초해상도 문제를 해결함에 있어 다운샘플링 역시 딥러닝을 통해 학습하여 복원 영상의 화질을 향상시켰다. 두 연구를 통해 영상 화질 개선 문제 해결에 있어 적절한 딥러닝 학습 방법을 영상 다운샘플링에 적용하여 좋은 결과를 얻을 수 있다는 것을 확인할 수 있다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.208-209
• /
• 2019

저해상도의 홀로그램을 Bilinear및 Bicubic 등의 알고리즘을 이용하여 업 스케일링을 하는 방법도 있다. 하지만, 홀로그램 데이터의 손실이 매우 크게 발생하며, 이로 인한 화질 저하가 발생하게 된다. 본 논문에서는 기존에 요구되던 파라미터와 연산량, 메모리를 대폭 감소시키면서도 준수한 성능을 보이는 RCI 구조를 제안한다. 제안한 네트워크 구조는 준수한 성능을 보이면서도 기존 2D 이미지에 대한 SISR 네트워크보다 더 빠르고 더 적은 메모리를 사용하였다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.162-164
• /
• 2019

본 논문에서는 물체인식 딥러닝 모델 생성에 필요한 라벨링(Labeling)과정에서 사용자가 다양한 기능을 활용하여 효과적인 학습 데이터를 구성할 수 있는 GUI 프로그램을 구현했다. 프로그램의 인터페이스는 파이썬 기반의 GUI 모듈인 Tkinter 를 활용하여, 실시간으로 이미지 데이터를 수집할 수 있는 크롤링(Crawling)기능과 미리 학습된 Retinanet 을 통해 이미지 데이터를 인식함으로써 자동으로 주석(Annotation) 과정을 수행할 수 있는 기능을 구성했다. 또한, 수집한 이미지 데이터를 다양한 효과와 노이즈, 변형 등으로 Augmentation 기능을 추가함으로써, 사용자가 모델을 학습하기 위한 데이터 전처리 단계를 하나의 GUI 프로그램에서 수행할 수 있도록 했다. 또한 사용자가 직접 학습한 모델을 추정 모델(Inference Model)로 변환하여 프로그램에 입력할 수 있도록 설계한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.143-144
• /
• 2019

자율 주행 차량의 상용화를 위해서는 차량의 정교한 위치 추정이 필수적이다 특히 실내공간의 경우 다중 경로 등 복잡한 경로를 주행 중인 차량의 위치를 추적해야 한다. 이 경우 정밀한 위치 추정을 위해 이동체가 주행하는 경로를 정확히 판별하는 것이 필수적이다. 본 논문에서는 다중 경로가 존재하는 복잡한 실내공간을 주행하는 이동체의 경로 추정을 위해 딥러닝 기법을 이용한다. 특히 딥러닝 기법이 주행 차량의 영상 정보를 활용하는 방식을 기술한다. 본 논문에서 딥러닝 방식은 주행 차량의 영상 정보를 이용하여 이동체가 주행하게 될 경로를 예측한다. 모의실험은 적용된 딥러닝 방식이 이동체의 주행 경로를 정확하게 예측함을 보인다.