• The Journal of the Korea Contents Association
• /
• v.19 no.2
• /
• pp.469-476
• /
• 2019

Robotic motion is designed by the rotation and the translation of multiple joint coordinates in a three-dimensional space. Joint coordinates are generally modeled by homogeneous transform matrix. However, the complexity of three dimensional motions prefers the visualization methods based on simulation environments in which models and generated motions work properly. Many simulation environments have the limitations of usability and functional extension from platform dependency and interpretation of predefined commands. This paper proposes the web-based three dimensional simulation environment toward high user accessibility. Also, it covers the small size web server that is linked with Python script. The non linearities of robot control apply to verify the computing efficiency, the process management, and the extendability of user scripts.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.7
• /
• pp.834-841
• /
• 2020

Among the autonomous driving systems based on visual sensors, the control method using a vanishing point is the most general method for autonomous driving. However, if the lane is lost or does not exist, it is very difficult to detect this and estimate the vanishing point. In this paper, we predict the vanishing point of the road and the vanishing point lines on the left and right sides using CNN for the camera image and design the steering controller for autonomous driving from the predicted results. As a result of the simulation, it was confirmed that the proposed method well tracked the center of the road regardless of the presence or absence of a solid lane, and was superior to the control method using a general method using the vanishing point.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.27-29
• /
• 2021

인공지능 기반 머신 비전 응용이 증가함에 따라 사람이 아닌 기계에서 소비되는 영상 정보를 전송하는 요구가 발생하고 있다. 일반적으로 영상 정보를 전송할 때는 전송 비용을 고려하여 정보를 압축하며 기존 영상 압축 방법은 사람의 시각 인지적 특성을 반영하여 설계되었다. 따라서 기존 영상 압축 방법은 기계에서 소비되는 영상 정보를 압축하는 방법으로 적절하지 않다고 판단하여 2019년 7월, 기계를 위한 영상 부호화 기술의 표준화가 시작되었다. 본 논문에서는 머신 비전 태스크 중, 객체 탐지를 수행하는 네트워크의 피처 맵을 압축하는 방법을 제안한다. 제안하는 방법은 피처 맵의 채널 간 중복성을 제거하기 위해 PCA 기반의 변환을 적용하여 피처 맵의 차원을 축소하며 특히 해상도 계층 구조를 갖는 네트워크의 피처 맵을 압축하기 위해 각 해상도 계층간 변환 기저를 예측하여 추가로 압축률을 높인다. 제안하는 방법을 적용하여 객체 탐지 결과의 큰 성능 하락 없이 약 92.3%에 데이터양 감소를 달성하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.01a
• /
• pp.281-282
• /
• 2021

본 논문에서는 항공 수요예측을 위한 빅데이터 기반 플랫폼의 설계 및 실증 결과를 제시한다. 항공 수요예측 통합 플랫폼은 항공산업 관련 데이터를 Open API, RSS Feed, 웹크롤러(Web Crawler) 등을 이용하여 수집 및 분석하여 자체 개발한 항공 수요예측 알고리즘을 기반으로 결과를 시각화하여 보여주도록 구현되어 있다. 또한, 제안하는 플랫폼의 사용자 인터페이스를 통해 변수 설정을 하여 단위별(Global, National 등), 기간별(단기, 중장기 등), 유형별(여객, 화물 등) 예측 통계 자료를 도출할 수 있다. 플랫폼의 성능 검증을 위해 정형화된 데이터를 비롯하여 소셜네트워크서비스(SNS), 검색엔진 등에서 수집한 비정형 데이터까지 활용하여 특정 키워드의 빈도와 특정 노선에 대한 항공 수요간 상관관계를 분석하였다. 개발한 통합 플랫폼의 지능형 항공 수요예측 알고리즘을 통해 전반적인 공항 운영 및 공항 운영 정책 수립에 기여할 것으로 예상한다.

• The Journal of Bigdata
• /
• v.6 no.2
• /
• pp.161-168
• /
• 2021

In this study, we propose a cardiovascular disease prediction model using machine learning. First, a multidimensional analysis of various differences between the two groups is performed and the results are visualized. In particular, we propose a predictive model using cost-sensitive learning that can improve the sensitivity for cases where there is a high class imbalance between the normal and patient groups, such as diseases. In this study, a predictive model is developed using CART and XGBoost, which are representative machine learning technologies, and prediction and performance are compared for cardiovascular disease patient data. According to the study results, CART showed higher accuracy and specificity than XGBoost, and the accuracy was about 70% to 74%.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.253-257
• /
• 2023

최근 다양한 대화 시스템이 스마트폰 어시스턴트, 자동 차 내비게이션, 음성 제어 스피커, 인간 중심 로봇 등의 실세계 인간-기계 인터페이스에 적용되고 있다. 하지만 대부분의 대화 시스템은 텍스트 기반으로 작동해 다중 모달리티 입력을 처리할 수 없다. 이 문제를 해결하기 위해서는 비디오와 같은 다중 모달리티 장면 인식을 통합한 대화 시스템이 필요하다. 기존의 비디오 기반 대화 시스템은 주로 시각, 이미지, 오디오 등의 다양한 자질을 합성하거나 사전 학습을 통해 이미지와 텍스트를 잘 정렬하는 데에만 집중하여 중요한 행동 단서와 소리 단서를 놓치고 있다는 한계가 존재한다. 본 논문은 이미지-텍스트 정렬의 사전학습 임베딩과 행동 단서, 소리 단서를 활용해 비디오 기반 대화 시스템을 개선한다. 제안한 모델은 텍스트와 이미지, 그리고 오디오 임베딩을 인코딩하고, 이를 바탕으로 관련 프레임과 행동 단서를 추출하여 발화를 생성하는 과정을 거친다. AVSD 데이터셋에서의 실험 결과, 제안한 모델이 기존의 모델보다 높은 성능을 보였으며, 대표적인 이미지-텍스트 자질들을 비디오 기반 대화시스템에서 비교 분석하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.908-909
• /
• 2023

Recently, several generative models have emerged and are being used in various industries. Among them, Cycle GAN is still used in various fields such as style transfer, medical care and autonomous driving. In this paper, we propose two methods to improve the performance of these Cycle GAN model. The ReLU activation function previously used in the generator was changed to Leaky ReLU. And a new loss function is proposed that considers the semantic level rather than focusing only on the pixel level through the VGG feature extractor. The proposed model showed quality improvement on the test set in the art domain, and it can be expected to be applied to other domains in the future to improve performance.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.6
• /
• pp.1259-1268
• /
• 2023

This study proposes a throttle and steering control technology using visual sensors based on deep learning's convolutional and recurrent neural networks. It collects camera image and control value data while driving a training track in clockwise and counterclockwise directions, and generates a model to predict throttle and steering through data sampling and preprocessing for efficient learning. Afterward, the model was validated on a test track in a different environment that was not used for training to find the optimal model and compare it with a CNN (Convolutional Neural Network). As a result, we found that the proposed deep learning model has excellent performance.

• KIPS Transactions on Computer and Communication Systems
• /
• v.9 no.2
• /
• pp.37-44
• /
• 2020

In this paper, we proposed that a fully automatic multi-class whole heart segmentation algorithm using deep learning. The proposed method is based on U-Net architecture which consist of recurrent convolutional block, residual multi-dilated convolutional block. The evaluation was accomplished by comparing automated analysis results of the test dataset to the manual assessment. We obtained the average DSC of 96.88%, precision of 95.60%, and recall of 97.00% with CT images. We were able to observe and analyze after visualizing segmented images using three-dimensional volume rendering method. Our experiment results show that proposed method effectively performed to segment in various heart structures. We expected that our method can help doctors and radiologist to make image reading and clinical decision.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.697-702
• /
• 2008

Mobile display panel is small so that users are often difficult to perceive images clearly. About image we perceive much through colors and therefore we propose color fitting approach for clear perception even on the small and low quality LCD panels. Various color modifications have been studied and used in commercial software packages. For mobile usage, our approach instantly enhances color images by modifying colors in a way to contrast differences of them. The method includes tone enhancements (which contrast dark and bright sides) and color enhancements (which reduce saturation for pure colorants). Based on color theory, our method also modifies color values towards specified complementary and preference colors. We term this color fitting. This approach enables displaying photos, multimedia messages, videos and digital media broadcasting (DMB) for better perception in real-time on mobile devices. Index Terms.) color fitting, visualization on small display, mobile graphics, visual perception.