• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.03a
• /
• pp.157-160
• /
• 1998

본 논문에서는 동영상으로부터 에지 정보와 형판 변형을 통해 얼굴의 추출하고 그 특징을 기반으로 하는 추적 기법을 제안한다. 본 논문에서 제안하는 추적기법은 추출된 특징에 기반을 추적 지법으로 초기에 모자익 영상을 이용하여 얼굴 부분을 찾고 찾아진 얼굴 부분에 에지 연산자를 적용하여 에지를 추출한다. 에지 영상이 얻어지면 에지 영상에서 영역의 크기와 모양, 그리고 관계 검증을 통해 대략적인 눈 영역을 추출한다. 눈 영역이 찾아지면 이를 바탕으로 입 영역에 대한 후보 영역에 대하여 이진화를 수행하고 히스토그램 프로젝션을 통해 대략적인 입 영역을 추출한다. 추출된 눈 영역과 입 영역에 각각의 형판을 사용해 형판 변형을 하고 초기 매개변수를 추출한다. 추출된 매개변수는 다음 프레임에서 형판의 초기 값으로 사용된다. 그리고 나서 형판에 대하여 변형(deformation) 과정을 수행한다. 이 과정을 반복함으로써 추적 과정을 수행한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.11a
• /
• pp.699-702
• /
• 2002

본 논문에서는 anisotropic diffusion 방법의 일종인 평균곡률 확산 (Mean Curvature Diffusion) 방법을 이용하여 영상에 포함된 잡음은 제거하고 동시에 에지는 보존하는 기법을 제안한다. 평균곡률 확산은 2 차원 영상의 밝기값을 3 차원 공간상의 z 좌표에 대응시켜 영상의 밝기값에 대응하는 공간 상의 곡면을 구성하고 이 곡면을 평균곡률에 비례하는 속도로 확산시킨다. 확산이 진행되면서 평균곡률이 영이 되는 에지에서는 확산이 발생하지 않고 잡음 등의 영향이 많은 에지 이외의 영역에서는 확산이 빠른 속도로 진행된다. 기존의 평균곡률 확산 방법의 성능을 개선하기 위해 최소/최대 흐름 방법을 평균곡률 확산 방법과 결합시키고 영상의 2 차 도함수를 사용하여 d얇은 에지를 보존하였다. 실험을 통해 제안한 방법이 기존의 방법보다 잡음 제거와 에지 보존 성능이 우수함을 확인할 수 있었다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.3
• /
• pp.748-753
• /
• 2004

Edge includes variety of information about image's location and shape etc. So a lot of researches for detecting those edges have been continuing even now and at the early stage of the research, edge detection operators searched neighborhood pixels and used relation among pixels. However, these methods do not have excellent performance in the image which exists noise and can not detect edges selectively. For this reason, in this paper we detected image's line-edge element from degraded image by AWGN (additive white gaussian noise) with wavelet function which is independent of line's width.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.244-246
• /
• 2014

Edge detection is a technique that obtains the particular information of the image using the brightness variation of pixel values and utilized for preprocessing in various image processing sectors. The conventional edge detection methods such as Sobel, Prewitt and Roberts are processed by applying the same weighted value to the entire pixels regardless of pixel distribution and provides somewhat insufficient edge detection results. Therefore, this paper has proposed an edge detection algorithm considering the direction and size of pixels by applying a modified directional mask.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.3
• /
• pp.147-153
• /
• 2004

This paper proposes an efficient edge detection using Van der Waerden statistic in original and noisy images. An edge is where the intensity of an image moves from a low value to a high value or vice versa. We describe a nonparametric Wilcoxon test and a parametric T test based on statistical hypothesis testing for the detection of edges. We use the threshold determined by specifying significance level $\alpha$, while Bovik, Huang and Munson consider the range of possible values of test statistics for the threshold. From the experimental results of edge detection, the T and Wilcoxon method perform sensitively to the noisy image, while the proposed Waerden method is robust over both noisy and noise-free images under $\alpha$=0.0005. Comparison with our statistical test and Sobel, LoG, Canny operators shows that Waerden method perform more effectively in both noisy and noise-free images.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.7 no.5
• /
• pp.1033-1039
• /
• 2012

In this paper, a simple and robust algorithm is proposed for detecting each steel plate from a image which contains several steel plates. Steel plate is characterized by line edge, so line detection is a fundamental task for analyzing and understanding of steel plate images. To detect the line edge, the proposed algorithm uses the small eigenvalue analysis. The proposed approach scans an input edge image from the top left corner to the bottom right corner with a moving mask. A covariance matrix of a set of edge pixels over a connected region within the mask is determined and then the statistical and geometrical properties of the small eigenvalue of the matrix are explored for the purpose of straight line detection. Using the detected line edges, each plate is determined based on the directional information and the distance information of the line edges. The results of the experiments emphasize that the proposed algorithm detects each steel plate from a image effectively.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.1C
• /
• pp.140-148
• /
• 2004

Natural images can be classified into edge or flat region. Edges have also various shapes such as long edge, texture and so on. Because the conventional deinterlacing methods commonly use one specific algorithm, they are faced with the difficulty that does not adapt various shapes of images. In this paper, a selective deinterlacing method based on the characteristics of local region of image is proposed. An input image is classified into three regions; flat region, complex edge, long edge. And then for each region, the proper method is assigned according to the characteristic of the local feature. For long edge region, the modified $NEDI(New Edge Directed Interpolation)^{[1]}$ method that interpolates long edge very well is used. The linear $filter^{[2]}$ that enhances high frequency components is used for complex edge, and the bilinear interpolation method is applied to flat region. The proposed method shows improved performance in PSNR and subjective evaluation compared with previous algorithms.

• Journal of the Institute of Convergence Signal Processing
• /
• v.15 no.4
• /
• pp.135-140
• /
• 2014

As the digital image devices are widely used, interests in the software- and the hardware-related image processing become higher and the image processing techniques are applied in various fields such as object recognition, object detection, fingerprint recognition, and etc. For the edge detections Sobel, Prewitt, Laplacian, Roberts and Canny detectors are used and these existing methods can excellently detect the edges of the images without noise. However, in the images corrupted by the impulse noise, these methods are insufficent in noise elimination characteristics, showing unsatisfactory edge detection. Therefore in this paper, in order to obtain excellent edge detection characteristics in the corrupted image by the impulse noise, an detection algorithm is porposed, which uses the central pixel of mask divided by four regions along the axis, calculates the estimated mask according to the representing pixel values in each regions, and detects the final edges by applying the estimates mask and the new directional one.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.04b
• /
• pp.292-294
• /
• 2002

본 연구에서는 영상에서 효과적인 에지 연결(edge linking)을 위하여 기존의 셀룰러 신경망 구조에서 두 가지 유형의 시냅스 구조를 고려한 활성화 특성을 제안한다 제시하는 모델에서 노드들간의 측면 연결에 의한 상호 작용은 노이즈에 의한 에지 및 영상에서 추출된 비최대점(non-maximum)의 에지를 억제할 문만 아니라, 특정 노드의 원형 이웃(circular neighborhood)으로 그려되는 특징들 간의 상호 연관도를 반영하여 에지의 연결 효과를 이를 수 있게 한다. 이러한 과정은 에지를 표현하는 벡터형식의 각 성분에 대한 활성화 특성으로부터 정형화된 에너지 함수로 모델링하고 이에 대한 최적화 과정으로써 구현될 수 있다.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.4 s.42
• /
• pp.9-15
• /
• 2006

In this paper we propose Wilcoxon test, Median test and Van der Waerden test such as linear rank tests in two-sample location problem for detecting edges effectively in noisy images. These methods are based on detecting image intensity changes between two pixel neighborhoods using an edge-height model to perform effectively on noisy images. The neighborhood size used here is small and its shape is varied adaptively according to edge orientations. We compare and analysis the performance of these statistical edge detectors on both natural images and synthetic images with and without noise.