• Proceedings of the Korea Contents Association Conference
• /
• 2014.11a
• /
• pp.213-214
• /
• 2014

본 논문에서는 증강현실 저작도구를 사용 할 때 마커인식을 통하여 동영상의 Tracking 데이터를 추출하는 방법을 제안한다. 실험에 이용한 마커는 직사각형모양의 특징점이 잘 나타나는 물체로서, 사각형 마커인식을 위해 CornerDetection과 Matching기법을 사용하였다. Tracking을 활용하는 방식에는 동영상의 기준프레임을 활용하여 Tracking하는 방법과 각 프레임을 순차적으로 Tracking하여 비교하는 방법, 그리고 마커를 사용하지 않고 동영상의 Tracking데이터를 추출하는 방법이 있는데 본 논문에서는 이 세 가지 방법을 비교하여, 증강현실 저작도구의 상용화를 위한 최적화된 알고리즘을 제안한다.

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.10
• /
• pp.25-34
• /
• 1999

In this paper, an algorithm is presented to recognize lane and obstacles based on highway road image. The road images obtained by a video camera undergoes a pre-processing that includes filtering, edge detection, and identification of lanes. After this pre-processing, a part of image is grouped into 27 sub-windows and fed into a three-layer feed-forward neural network. The neural network is trained to indicate the road direction and the presence of absence of an obstacle. The proposed algorithm has been tested with the images different from the training images, and demonstrated its efficacy for recognizing lane and obstacles. Based on the test results, it can be said that the algorithm successfully combines the traditional image processing and the neural network principles towards a simpler and more efficient driver warning of assistance system

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.1
• /
• pp.345-359
• /
• 2017

Optical Character Recognition (OCR) that has been a main research topic of computer vision and artificial intelligence now extend its applications to detection of text area from video or image contents taken by camera devices and retrieval of text information from the area. This paper aims to implement a binarization algorithm that removes user intervention and provides robust performance to outdoor lights by using TopHat algorithm and channel transformation technique. In this study, we particularly concentrate on text information of outdoor signboards and validate our proposed technique using those data.

• Research in Mathematical Education
• /
• v.16 no.4
• /
• pp.207-216
• /
• 2012

Through quantitative video analysis of four algebra classes and statistical analysis of various types of teacher evaluation behavior in the classroom teaching, we get: (1) Teacher evaluation behavior in classroom is close to take 1/5 of the total time of the classroom teaching, and it appears most frequently in class exercises and take the longest time; (2) There are many forms of teacher evaluation behavior in classroom, and most of the behaviors are positive assessment; (3) Recognition evaluation is relatively conservative in a single form without losing fairness; (4) Classroom assessments of teachers' behaviors are primarily concerned about students' knowledge and skills mastery, and it is less involved in student feelings, attitudes and behaviors; (5) The correct teacher evaluation behavior in classroom will inspire students to create internal motivations; and (6) The correct teacher evaluation behavior in classroom can stimulate the potential of students.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06b
• /
• pp.396-398
• /
• 2012

대부분의 실세계 데이터는 시간성을 띄고 있으므로 시간성을 지닌 데이터를 분석할 수 있는 기계 학습 방법론은 매우 중요하다. 이런 관점에서 비디오 데이터는 다양한 모달리티가 결합된 대표적인 시간 데이터 이므로 비디오 데이터를 대상으로 하는 기계 학습 방법은 큰 의미를 갖는다. 본 논문에서는 음성 채널에기반한 비디오 데이터 분석 방법의 예비 연구로 비디오 데이터에 등장하는 화자를 인식할 수 있는 간단한 방법을 소개한다. 제안 방법은 MFCC (Mel-frequency cepstrum coefficients)를 이용하여 인간 음성 특성의 분포를 분석한 후 분석 결과를 신경망에 입력하여 목표한 화자를 인식하는 복합 신경망 모델을 특징으로 한다. 실제 TV 드라마 데이터에서 가우시안 혼합모델, 가우시안 혼합 신경망 모델, 제안 방법의 화자 인식 성능을 비교한 결과 제안 방법이 가장 우수한 인식 성능을 보임을 확인하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.6 no.9
• /
• pp.2341-2350
• /
• 2012

Face detection is the first step of vision-based driver fatigue detection method. Traditional face detection methods have problems of high false-detection rates and long detection times. A space-time restrained Adaboost method is presented in this paper that resolves these problems. Firstly, the possible position of a driver's face in a video frame is measured relative to the previous frame. Secondly, a space-time restriction strategy is designed to restrain the detection window and scale of the Adaboost method to reduce time consumption and false-detection of face detection. Finally, a face knowledge restriction strategy is designed to confirm that the faces detected by this Adaboost method. Experiments compare the methods and confirm that a driver's face can be detected rapidly and precisely.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.683-684
• /
• 2006

본 논문은 비디오기반의 얼굴인식에 있어서 환경의 변화나 왜곡, 노이즈 등으로 발생할 수 있는 부분적인 가림현상(Partial Occlusion)에 대한 처리기법을 제시한다. 인증되는 각 사람은 하나의 Manifold 를 구성하며 각 Manifold 는 m 개의 pose-Manifold 로 구성된다. Pose-Manifold 를 구성하기 위한 학습데이터는 매우 유사한 포즈들로 구성되기 때문에 얼굴을 이루는 영역의 픽셀에 대한 Intensity 의 변화는 크지 않다. 입력되는 이미지의 Intensity 를 학습데이터의 Intensity 의 변화량을 고려한 Intensity Distortion 을 이용하면 Occlusion 이 발생한 영역을 찾을 수 있고, Occlusion 이 발생한 정도에 따라 가중치를 부여할 수 있다. 이렇게 Occlusion 에 따라 영역에 중요도를 다르게 하여 얼굴인식률을 높이고자 한다. 실험에서는 제시하는 Mask 를 사용하지 않았을 경우와 기존에 제시된 알고리즘과의 성능을 비교한다.

• Journal of information and communication convergence engineering
• /
• v.11 no.4
• /
• pp.298-306
• /
• 2013

The scale invariant feature transform (SIFT) is an effective algorithm used in object recognition, panorama stitching, and image matching. However, due to its complexity, real-time processing is difficult to achieve with current software approaches. The increasing availability of parallel computers makes parallelizing these tasks an attractive approach. This paper proposes a novel parallel approach for SIFT algorithm implementation using a block filtering technique in a Gaussian convolution process on the SIMD Pixel Processor. This implementation fully exposes the available parallelism of the SIFT algorithm process and exploits the processing and input/output capabilities of the processor, which results in a system that can perform real-time image and video compression. We apply this implementation to images and measure the effectiveness of such an approach. Experimental simulation results indicate that the proposed method is capable of real-time applications, and the result of our parallel approach is outstanding in terms of the processing performance.

• Proceedings of the KIEE Conference
• /
• 2000.07d
• /
• pp.3041-3043
• /
• 2000

GUI(graphical user interface) has been the dominant platform for HCI(human computer interaction). The GUI-based style of interaction has made computers simpler and easier to use. However GUI will not easily support the range of interaction necessary to meet users' needs that are natural, intuitive, and adaptive. In this paper we study an approach to track a hand in an image sequence and recognize it, in each video frame for replacing the mouse as a pointing device to virtual reality. An algorithm for real time processing is proposed by estimating of the position of the hand and segmentation, considering the orientation of motion and color distribution of hand region.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1993.06a
• /
• pp.903-906
• /
• 1993

Lipreading through visual processing techniques help provide some useful systems for the hearing impaired to learn communication assistance. This paper proposes a method to understand spoken words by using visual images taken by a camera with a video-digitizer. The image is processed to obtain the contours of lip, which is approximated into a hexagon. The pattern lists, consisting of lengths and angles of hexagon, are compared and computed to get the fuzzy similarity between two lists. By similarity matching, the mouth shape is recognized as the one which has the pronounced voice. Some experiments, exemplified by recognition of the Japanese vowels, are given to show feasibilities of this method.