• 한국지능시스템학회논문지
• /
• 제20권3호
• /
• pp.348-355
• /
• 2010

음성 인식 기술의 발전에도 불구하고 잡음 환경하의 음성 인식은 여전히 어려운 분야이다. 이를 해결하기 위한 방안으로 음성 정보 이외에 시각 정보를 이용한 시각 음성인식에 대한 연구가 진행되고 있다. 하지만 시각 정보 또한 음성과 마찬가지로 주위 조명 환경이나 기타, 다른 요인에 따른 영상잡음이 존재하며, 이런 영상잡음은 시각 음성 인식의 성능 저하를 야기한다. 따라서 인식 성능 향상을 위해 시각 특징 파라미터를 어떻게 추출하느냐는 하나의 관심분야이다. 본 논문에서는 HMM기반 시각 음성인식의 인식 성능 향상을 위한 영상 기반 접근방법에 따른 시각 특징 파라미터의 추출 방법에 대하여 논하고 그에 따른 인식성능을 비교하였다. 실험을 위해 105명에 화자에 대한 62단어의 데이터베이스를 구축하고, 이를 이용하여 히스토그램 매칭, 입술 접기, 프레임 간 필터링 기법, 선형마스크, DCT, PCA 등을 적용하여 시각 특징 파라미터를 추출하였다. 실험결과, 제안된 방법에 의해 추출된 특징 파라미터를 인식기에 적용하였을 때의 인식 성능은 기본 파라미터에 비해 약21%의 성능 향상이 됨을 알 수 있다.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2001년도 춘계종합학술대회
• /
• pp.656-659
• /
• 2001

Human have a notable ability to recognize faces, which is one of the most common visual feature in our environment. In regarding face pattern, just like other natural object, a geometrical interpretation of face is difficult to achieve. In this paper, we present wavelet based approach to extract the face features. Proposed approach is similar to the feature based scheme, where the feature is derived from the intensity data without detecting any knowledge of the significant feature. Topological graphs are involved to represent some relations between facial features. In our experiments, proposed approach is less sensitive to the intensity variation.

• 한국정보통신학회논문지
• /
• 제5권2호
• /
• pp.349-355
• /
• 2001

Human have a notable ability to recognize faces, which is one of the most common visual feature in our environment. In regarding face pattern, just like other natural object, a geometrical interpretation of face is difficult to achieve. In this paper, we present wavelet based approach to extract the face features. Proposed approach is similar to the feature based scheme, where the feature is derived from the intensity data without detecting any knowledge of the significant feature. Topological graphs are involved to represent some relations between facial features. In our experiments, proposed approach is less sensitive to the intensity variation.

• 해양환경안전학회지
• /
• 제3권2호
• /
• pp.1-12
• /
• 1997

One of the main uses for ship simulators is in the field of port design, and an increasing number of simulators, of varying degrees of fidelity, are being used for this purpose. An essential feature of all such simulators is their visual scene, which must be of sufficent fidelity to convey the key visual cues adequately. This paper examines the ability of a number of experienced mariners to perceive speeds and distances correctly using Computer Generated Imagery visual scenes of different fidelity, compared with their performance at sea. From the results, it was found that the microcomputer based simulator might be considered, as far as its visual scene representation is concerned, to be as valid as the full mission ship simulator for the port design task.

• 한국컴퓨터정보학회논문지
• /
• 제20권11호
• /
• pp.183-190
• /
• 2015

This paper investigates the three main functions of a typical visual app programming environment for Novice Programming developers, and compares the features. The Scratch is a visual programming environment for education, anyone can create a story easy as possible variously interaction, games, animations and more. App inventor provides precise and professional application development capabilities as compared with scratch. App Inventor in runs independently of the computer platform, and has a feature that must be constantly connected to the server over the internet, while the Inventor app runs. M-Bizmaker is suitable for commercial application development, consists of m-BizBuilder, m-BizEngine, m-BizServer or the like, provides a cross-platform visual programming environment.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2013년도 제47차 동계학술대회논문집 21권1호
• /
• pp.197-199
• /
• 2013

신경망의 학습은 학습 샘플의 품질뿐만이 아니라 입력으로 사용되는 특징에도 영향을 받으므로 신경망의 출력을 결정하는데 있어 연관성이 높은 특징을 입력으로 사용함으로써 학습된 신경망의 전체적인 성능을 높일 수 있다. 이 논문에서는 신경망의 입력으로 사용되는 특징과 출력의 연관성 파악하고 연관성이 낮은 특징을 학습 과정에서 배제함으로써 신경망의 전체적인 성능을 높일 수 있는 방법을 제시하였다. 토석류 데이터를 위한 의사 샘플 신경망에 제안한 방법을 적용한 경우 연관성이 낮은 특징 하나를 제외함으로써 약 6%의 오류 감소 효과를 얻을 수 있었다.

• 한국멀티미디어학회논문지
• /
• 제23권10호
• /
• pp.1250-1257
• /
• 2020

The early detection of diseases is important in agriculture because diseases are major threats of reducing crop yield for farmers. The shape and color of plant leaf are changed differently according to the disease. So we can detect and estimate the disease by inspecting the visual feature in leaf. This study presents a vision-based leaf classification method for detecting the diseases of tomato crop. ResNet-50 model was used to extract the visual feature in leaf and classify the disease of tomato crop, since the model showed the higher accuracy than the other ResNet models with different depths. We propose a new ensemble approach using several DCNN classifiers that have the same structure but have been trained at different ranges in the DCNN layers. Experimental result achieved accuracy of 97.19% for PlantVillage dataset. It validates that the proposed method effectively classify the disease of tomato crop.

• International Journal of Contents
• /
• 제4권2호
• /
• pp.24-28
• /
• 2008

In this paper, we propose a new approach for content-based video retrieval using non-parametric based motion classification in the shot-based video indexing structure. Our system proposed in this paper has supported the real-time video retrieval using spatio-temporal feature comparison by measuring the similarity between visual features and between motion features, respectively, after extracting representative frame and non-parametric motion information from shot-based video clips segmented by scene change detection method. The extraction of non-parametric based motion features, after the normalized motion vectors are created from an MPEG-compressed stream, is effectively fulfilled by discretizing each normalized motion vector into various angle bins, and by considering the mean, variance, and direction of motion vectors in these bins. To obtain visual feature in representative frame, we use the edge-based spatial descriptor. Experimental results show that our approach is superior to conventional methods with regard to the performance for video indexing and retrieval.

• 한국항행학회논문지
• /
• 제7권2호
• /
• pp.162-170
• /
• 2003

In this paper, we propose a robust controller for trajectory control of n-link robot manipulators using feature based on visual feedback. In order to reduce tracking error of the robot manipulator due to parametric uncertainties, integral action is included in the dynamic control part of the inner control loop. The desired trajectory for tracking is generated from feature extraction by the camera mounted on the end effector. The stability of the robust state feedback control system is shown by the Lyapunov method. Simulation and experimental results on a 5-link robot manipulator with two degree of freedom show that the proposed method has good tracking performance.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년 학술대회 논문집 정보 및 제어부문
• /
• pp.434-436
• /
• 2006

In a visual driver-assistance system, separating moving objects from fixed objects are an important problem to maintain multiple hypothesis for the state. Color and edge-based tracker can often be "distracted" causing them to track the wrong object. Many researchers have dealt with this problem by using multiple features, as it is unlikely that all will be distracted at the same time. In this paper, we improve the accuracy and robustness of real-time tracking by combining a color histogram feature with a brightness of Optical Flow-based feature under a Sequential Monte Carlo framework. And it is also excepted from Tracking as time goes on, reducing density by Adaptive Particles Number in case of the fixed object. This new framework makes two main contributions. The one is about the prediction framework which separating moving objects from fixed objects and the other is about measurement framework to get a information from the visual data under a partial occlusion.