• KEPCO Journal on Electric Power and Energy
• /
• v.8 no.2
• /
• pp.137-141
• /
• 2022

To inspect the power facility faults using artificial intelligence, it need that improve the accuracy of the diagnostic model are required. Data augmentation skill using generative adversarial network (GAN) is one of the best ways to improve deep learning performance. GAN model can create realistic-looking fake images using two competitive learning networks such as discriminator and generator. In this study, we intend to verify the effectiveness of virtual data generation technology by including the fake image of power facility generated through GAN in the deep learning training set. The GAN-based fake image was created for damage of LP insulator, and ResNet based normal and defect classification model was developed to verify the effect. Through this, we analyzed the model accuracy according to the ratio of normal and defective training data.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.521-524
• /
• 2023

본 연구에서는 AI 인공지능을 활용한 통일신라 석탑 '경주 불국사 삼층석탑' 복원을 위해 석탑 3D모델링 과정에 대한 연구를 수행하였다. 산성비로 부식되어 갈라진 더 이상 원본의 형태를 알아 볼 수 없는 현재 통일신라 석탑 형태를 3D모델링 작업을 통하여 AI로 하여금 원활한 교육이 실시 되도록 하는 것을 목표로한다. 본래 제작 되어있는 3D 모델링은 많은 버텍스와 페이스로 학습 데이터가 많아 실제 활용하기에 어려움을 가지고 있다. 때문에 적은 양의 버텍스와 페이스로 새로운 3D 모델링 제작에 대한 필요성에 대해 확인하였다. 본 연구는 그에 필요한 석탑 모델링 과정에 대해 서술한다. 이를 위해 본 논문은 석탑에 대한 구조를 살피고 모델링에 활용된 프로그램의 장단점과 분석을 도출하였다. 본 연구를 통해 석탑 복원에 필요한 3D모델링 프로그램 활용의 전망과 더불어 인공지능 AI의 한계점을 3D 모델링의 정확도와 세밀함을 통하여 타파하고자 하였다.

• ETRI Journal
• /
• v.45 no.5
• /
• pp.862-873
• /
• 2023

Determining whether an autonomous self-driving agent is in the middle of an intersection can be extremely difficult when relying on visual input taken from a single camera. In such a problem setting, a wider range of views is essential, which drives us to use three cameras positioned in the front, left, and right of an agent for better intersection recognition. However, collecting adequate training data with three cameras poses several practical difficulties; hence, we propose using data collected from one camera to train a three-camera model, which would enable us to more easily compile a variety of training data to endow our model with improved generalizability. In this work, we provide three separate fusion methods (feature, early, and late) of combining the information from three cameras. Extensive pedestrian-view intersection classification experiments show that our feature fusion model provides an area under the curve and F1-score of 82.00 and 46.48, respectively, which considerably outperforms contemporary three- and one-camera models.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.12
• /
• pp.485-496
• /
• 2018

So far, the deep learning, a field of artificial intelligence, has achieved remarkable results in solving problems from unstructured data. However, it is difficult to comprehensively judge situations like humans, and did not reach the level of intelligence that deduced their relations and predicted the next situation. Recently, deep neural networks show that artificial intelligence can possess powerful relational reasoning that is core intellectual ability of human being. In this paper, to analyze and observe the performance of Relation Networks (RN) among the neural networks for relational reasoning, two types of RN-based deep neural network models were constructed and compared with the baseline model. One is a visual question answering RN model using Sort-of-CLEVR and the other is a text-based question answering RN model using bAbI task. In order to maximize the performance of the RN-based model, various performance improvement experiments such as hyper parameters tuning have been proposed and performed. The effectiveness of the proposed performance improvement methods has been verified by applying to the visual QA RN model and the text-based QA RN model, and the new domain model using the dialogue-based LL dataset. As a result of the various experiments, it is found that the initial learning rate is a key factor in determining the performance of the model in both types of RN models. We have observed that the optimal initial learning rate setting found by the proposed random search method can improve the performance of the model up to 99.8%.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2006.04a
• /
• pp.111-118
• /
• 2006

The LBS (Location Based Service) location information is the element which is important decides the application of form service, Recently the location accuracy with the method which applies the wireless Internet DGPS improved inside 1 meter Like this. application of the LBS which corresponds In the location accuracy which Improves accurate the Location Information output system development which is accurate stands is essential. In this study which it sees it used 1:5000 digital map MS visual 6.0 Active X controls of the country geography intelligence Personnel and it developed the location information output module of web browser base. The location information output system which is developed it will be able to secure the convenient characteristic of precision and visual program of the digital map simultaneously and the maximum anger of LBS application is expected.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.21 no.48
• /
• pp.101-111
• /
• 1998

CNS (Car Navigation System) is the most compatible candidate among various in-vehicle information systems as a provider of ITS (Intelligence Transport Systems) information. It generally consists of remote controller, display, CD-changer, GPS receiver and so on. Among them, display is the most important and critical element of the HMI (Human-Machine Interface) suggesting the digital map to the driver. Therefore, it is certain that the display gives cognitive, physical, mental and visual workloads to the driver which are directly related with the driver's and road safety with the success of ITS. Until now, various human factors techniques have been developed and applied to estimate the driver's workload and to collect the driver's requirements of the CNS digital map, for example, mental workload assessment, visual activity analysis, cognitive analysis and so on. In addition to these kinds of techniques, this research performed the human sensibility ergonomics approach to directly investigate and evaluate the driver's requirements and sensibilities of the real products.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2008.04a
• /
• pp.89-91
• /
• 2008

엄청나게 쏟아져 나오는 데이터 홍수 속에서 오늘날의 업체와 연구기관에서는 신속하게 의사 결정을 해야 한다. 당면한 문제점들을 해결하기 위하여 접근할 수 있는 수많은 다양한 데이터 속에서 정확하게 경향을 파악하고 그 근본 원인을 찾아내어 신속하고 action을 행하는 것은 어떠한 회사에서도 성공에 있어서 가장 중요한 인자들 중의 하나이다. 초기 아이디어 도출, 연구 개발에서부터 제품의 생산, 판매 및 서비스에 이르기까지 모든 팀원들은 아주 빠르게 고도의 정확성으로 중요한 결정을 할 필요가 있다. 오늘날의 경쟁 시장에서 기업의 성공은 다른 경쟁자들보다 더 빠르게 결정을 할 수 있는 능력에 달려 있다. 이에 Sporfire에서는 사용자가 쉽고 빠르게 데이터를 분석하여 의사 결정을 할 수 있도록 다양한 기능을 제공하고 있다. 사용자가 SQL같은 전문 언어를 사용하지 않고도 다양한 데이터 source에서 쉽게 데이터를 가져오도록 Information Library를 이용할 수 있으며, 데이터베이스에 들어 있는 숫자들의 집합체를 다양한 차트와 도표들을 이용, 그래픽 적으로 제공해 줌으로써 데이터에 대하여 직관적으로 파악하여 신속하게 대응할 수 있도록 도와준다. 또한 그 결과물들을 MS 파워포인트, 엑셀시트, xml 등으로 저장하여 다른 용도로 사용할 수 있도록 하고 있다.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.474-474
• /
• 2000

This paper shows that a system with two-link arm can obtain hand reaching movement to a target object projected on a visual sensor by reinforcement learning using a layered neural network. The reinforcement signal, which is an only signal from the environment, is given to the system only when the hand reaches the target object. The neural network computes two joint torques from visual sensory signals, joint angles, and joint angular velocities considering the urn dynamics. It is known that the trajectory of the voluntary movement o( human hand reaching is almost straight, and the hand velocity changes like bell-shape. Although there are some exceptions, the properties of the trajectories obtained by the reinforcement learning are somewhat similar to the experimental result of the human hand reaching movement.

• Current Optics and Photonics
• /
• v.4 no.3
• /
• pp.200-209
• /
• 2020

We describe an approach that uses a bilateral filter to reduce the ringing artifact in motion-blurred restoration image. It takes into account the specific physical structure of the ringing artifact combined with the properties of the human visual system. To properly reduce the ringing artifact, each of the adjacent pixels is limited in a straight line which has a given direction. To protect the edges and the texture regions of an image, our algorithm divides the image into texture regions and flat regions, and the artifact reduction algorithm is only applied to the flat region. Finally, we use 8 typical images and 5 objective quality evaluation indices to evaluate our algorithm. Experimental results show that our algorithm can obtain better results in subjective visual effect and in objective image quality evaluation.

• Korean Institute of Interior Design Journal
• /
• no.29
• /
• pp.211-219
• /
• 2001

Dematerialization of consumption is in a rapid process, in relation to the flood of information. Consumers do not want to buy a material goods any more, in the condition of excessive supply. In the past industrial society with massive production and consumption, only the unified consumers were produced. In informational society, goods are not regarded as a simple material, but as an Intelligence Merchandise. Specialty shops for life articles should reflect these consumer's trend and should keep the value of brand and goods. Visual Merchandising strategy which connect the expectation of consumers and company images, should be considered as a communication tools, and be presented in diverse and specified ways.