• International Journal of Computer Science & Network Security
• /
• 제22권6호
• /
• pp.364-373
• /
• 2022

Image morphing methods make seamless transition changes in the image and mask the meaningful information attached to it. This can be detected by traditional machine learning algorithms and new emerging deep learning algorithms. In this research work, scope of different Hybrid learning approaches having combination of Deep learning and Machine learning are being analyzed with the public dataset CASIA V1.0, CASIA V2.0 and DVMM to find the most efficient algorithm. The simulated results with CNN (Convolution Neural Network), Hybrid approach of CNN along with SVM (Support Vector Machine) and Hybrid approach of CNN along with Random Forest algorithm produced 96.92 %, 95.98 and 99.18 % accuracy respectively with the CASIA V2.0 dataset having 9555 images. The accuracy pattern of applied algorithms changes with CASIA V1.0 data and DVMM data having 1721 and 1845 set of images presenting minimal accuracy with Hybrid approach of CNN and Random Forest algorithm. It is confirmed that the choice of best algorithm to find image forgery depends on input data type. This paper presents the combination of best suited algorithm to detect image morphing with different input datasets.

• International Journal of Advanced Culture Technology
• /
• 제8권2호
• /
• pp.42-47
• /
• 2020

This study is to develop Korean version of the 7 Learning Style Inventory(LSI) for IT major Students by systematic translation process and to test learning style of IT major University students. Translated and developed Korean version of LSI was verified of validity by comparing with existing V.A.K. learning style model. We can develop various tactics for 7 learning styles of students. Once the learning style of each student is confirmed, customized teaching for individual and team can be done more efficiently through teaching and learning strategies according to each learning style. Developed LSI was applied to the IT major students of two classes from Chungwoon University in Incheon. Results of LSI survey show that learning styles of 24 students out of 35 students from two classes are matched with V.A.K. learning styles of same students. It was 68.6% match in learning style, and shows that validity of 7 LSI. We need to elaborate Korean questionnaires of the LSI more, and extend and apply to the non-IT major students group.

• Current Photovoltaic Research
• /
• 제10권3호
• /
• pp.77-83
• /
• 2022

Renewable energy is receiving attention again as a way to realize carbon neutrality to overcome the climate change crisis. Among renewable energy sources, the installation of Photovoltaic is continuously increasing, and as of 2020, the global cumulative installation amount is about 590 GW and the domestic cumulative installation amount is about 17 GW. Accordingly, O&M technology that can analyze the power generation and fault diagnose about PV plants the is required. In this paper, a study was conducted to diagnose fault using I-V curves of PV strings and deep learning. In order to collect the fault I-V curves for learning in the deep learning, faults were simulated. It is partial shade and voltage mismatch, and I-V curves were measured on a sunny day. A two-step data pre-processing technique was applied to minimize variations depending on PV string capacity, irradiance, and PV module temperature, and this was used for learning and validation of deep learning. From the results of the study, it was confirmed that the PV fault diagnosis using I-V curves and deep learning is possible.

• 한국전자거래학회지
• /
• 제6권3호
• /
• pp.115-126
• /
• 2001

In constructive view, acquiring knowledge is made by experiences among members or elements. The knowledge in e-learning system can be extended up to knowledge of teachers and knowledge. of operating managers. We have many difficult problems to develop and manage e-1earning system because demanders on e-learning system have various. requirements. In traditional education system demanders are learners but in constructive view demanders can be extended to learners and teachers, operating mangers on e-learning system., In this study, we design and implement e-learning system named kedu V.1. Kedu V.1 is developed considering interactions of extended demanders of e-learning system in constructive view. And this system based on Linux operation system and MySQL, PHP. Also this system have efficient transplantation and portability capabilities and reduced cost and labor in implementation of real e-learning system

• International Journal of Computer Science & Network Security
• /
• 제21권7호
• /
• pp.324-330
• /
• 2021

Agriculture is one of the essential needs of human life on planet Earth. It is the source of food and earnings for many individuals around the world. The economy of many countries is associated with the agriculture sector. Lots of diseases exist that attack various fruits and crops. Apple Tree Leaves also suffer different types of pathological conditions that affect their production. These pathological conditions include apple scab, cedar apple rust, or multiple diseases, etc. In this paper, an automatic detection framework based on deep learning is investigated for apple leaves disease classification. Different pre-trained models, VGG16, ResNetV2, InceptionV3, and MobileNetV2, are considered for transfer learning. A combination of parameters like learning rate, batch size, and optimizer is analyzed, and the best combination of ResNetV2 with Adam optimizer provided the best classification accuracy of 94%.

• 대한원격탐사학회지
• /
• 제39권3호
• /
• pp.269-282
• /
• 2023

본 연구는 딥러닝 모델(deep learning model)을 활용하여 토지피복분류를 수행하였으며 입력 이미지의 크기, Stride 적용 등 데이터세트(dataset)의 조절을 통해 토지피복분류를 위한 최적의 딥러닝 모델 선정을 목적으로 하였다. 적용한 딥러닝 모델은 3종류로 Encoder-Decoder 구조를 가진 U-net과 DeeplabV3+, 두 가지 모델을 결합한 앙상블(Ensemble) 모델을 활용하였다. 데이터세트는 RapidEye 위성영상을 입력영상으로, 라벨(label) 이미지는 Intergovernmental Panel on Climate Change 토지이용의 6가지 범주에 따라 구축한 Raster 이미지를 참값으로 활용하였다. 딥러닝 모델의 정확도 향상을 위해 데이터세트의 질적 향상 문제에 대해 주목하였으며 딥러닝 모델(U-net, DeeplabV3+, Ensemble), 입력 이미지 크기(64 × 64 pixel, 256 × 256 pixel), Stride 적용(50%, 100%) 조합을 통해 12가지 토지피복도를 구축하였다. 라벨 이미지와 딥러닝 모델 기반의 토지피복도의 정합성 평가결과, U-net과 DeeplabV3+ 모델의 전체 정확도는 각각 최대 약 87.9%와 89.8%, kappa 계수는 모두 약 72% 이상으로 높은 정확도를 보였으며, 64 × 64 pixel 크기의 데이터세트를 활용한 U-net 모델의 정확도가 가장 높았다. 또한 딥러닝 모델에 앙상블 및 Stride를 적용한 결과, 최대 약 3% 정확도가 상승하였으며 Semantic Segmentation 기반 딥러닝 모델의 단점인 경계간의 불일치가 개선됨을 확인하였다.

• 한국방사선학회논문지
• /
• 제15권4호
• /
• pp.455-461
• /
• 2021

심장비대증은 흉부 X선 영상에서 흔히 보이는 질병 중 하나이지만 조기에 발견을 하지 못하면 심각한 합병증을 유발할 수도 있다. 이러한 점을 고려하여 최근에는 여러 과학기술 분야의 발전으로 인공지능을 이용한 딥러닝 알고리즘을 의료에 접목시키는 영상 분석 연구들이 많이 진행되고 있다. 본 논문에서는 Inception V3 딥러닝 모델을 흉부 X선 영상을 이용하여 심장비대증의 분류에 유용한 모델인지 평가하고자 한다. 사용된 영상의 경우 총 1026장의 경북대학교병원 내 정상 심장 진단을 받은 환자와 심장비대증 진단을 받은 환자의 흉부 X선 영상을 사용하였다. 실험결과 Inception V3 딥러닝 모델의 심장비대증 유무에 따른 분류 정확도와 손실도 결과값은 각각 96.0%, 0.22%의 결과값을 나타내었다. 연구결과를 통해 Inception V3 딥러닝 모델은 흉부 영상 데이터의 특징 추출 및 분류에 있어 우수한 딥러닝 모델인 것을 알 수 있었다. Inception V3 딥러닝 모델의 경우 흉부 질환의 분류에 있어 유용한 딥러닝 모델이 될 것으로 판단되며 조금 더 다양한 의료 영상 데이터를 이용한 연구를 진행하여 이와 같은 우수한 연구결과를 얻게 된다면 향후 임상의의 진단 시 많은 도움을 줄 수 있을 것으로 사료된다.

• 한국응급구조학회지
• /
• 제28권2호
• /
• pp.131-142
• /
• 2024

Purpose: This study aimed to explore the use of web-based virtual simulations to improve the learning flow, problem-solving ability, and academic self-efficacy of paramedicine students. Methods: This study used a pre-post comparison design with 40 pamedicine students at one university. the participants were third grade students and participated in web based virtual simulation(vSim®). The difference between the pre-post was analyzed using a Paired Sample t-test with SPSS 26.0. Results: After web based virtual simulation using vSim® program than before, significantly higher learning flow(t=12.74, p<.001), problem-solving process(t=5.78, p<.001), academic self-efficacy(t=2.77, p<.001), Paramedic-Care Skills Confidence(t=6.30, p<.001). Conclusion: The results indicated the positive effects of web-based virtual simulation(vSim®) learning method and suggests implications for learners and instructors.

• 반도체디스플레이기술학회지
• /
• 제22권4호
• /
• pp.136-141
• /
• 2023

The continuous shrinking of transistors in integrated circuits leads to difficulties in improving performance, resulting in the emerging transistors such as nanosheet field-effect transistors. In this paper, we propose a TCAD-machine learning framework of nanosheet FETs to model the current-voltage characteristics. Sentaurus TCAD simulations of nanosheet FETs are performed to obtain a large amount of device data. A machine learning model of I-V characteristics is trained using the multi-layer perceptron from these TCAD data. The weights and biases obtained from multi-layer perceptron are implemented in a PSPICE netlist to verify the accuracy of I-V and the DC transfer characteristics of a CMOS inverter. It is found that the proposed machine learning model is applicable to the prediction of nanosheet field-effect transistors device and circuit performance.

• 한국컴퓨터정보학회논문지
• /
• 제26권12호
• /
• pp.69-75
• /
• 2021

최근 인공지능 분야에서 가장 많이 사용하는 딥러닝은 그 구조가 점차 크고 복잡해지고 있다. 딥러닝 모델이 커질수록 이를 학습시키기 위해서는 대용량의 데이터가 필요하지만 데이터가 여러 소유 주체별로 분산되어 있고 보안 문제로 인해 이를 통합하여 학습시키기 어려운 경우가 발생한다. 우리는 동일한 딥러닝 모형이 필요하지만 보안 문제로 인해 데이터가 여러곳에 분산되어 처리될 수 밖에 없는 상황에서 데이터를 소유하고 있는 주체별로 분산 학습을 수행한 후 이를 통합하는 방법을 연구하였다. 이를 위해 보안 상황을 V-환경과 H-환경으로 가정하여 소유 주체별로 분산학습을 수행했으며 Average, Max, AbsMax를 사용하여 분산학습된 결과를 통합하였다. mnist-fashion 데이터에 이를 적용해 본 결과 V-환경에서는 정확도 면에서 데이터를 통합시켜 학습한 결과와 큰 차이가 없음을 확인할 수 있었으며, H-환경에서는 차이는 존재하지만 의미있는 결과를 얻을 수 있었다.