• Transactions on Control, Automation and Systems Engineering
• /
• v.3 no.4
• /
• pp.283-288
• /
• 2001

In this paper, we proposed a method for extracting facial characteristics of human being in an image. Given a pair of gray level sample images taken with and without human being, the face of human being is segmented from the image. Noise in the input images is removed with the help of Gaussian filters. Edge maps are found of the two input images. The binary edge differential image is obtained from the difference of the two input edge maps. A mask for face detection is made from the process of erosion followed by dilation on the resulting binary edge differential image. This mask is used to extract the human being from the two input image sequences. Features of face are extracted from the segmented image. An effective recognition system using the discrete wave let transform (DWT) is used for recognition. For extracting the facial features, such as eyebrows, eyes, nose and mouth, edge detector is applied on the segmented face image. The area of eye and the center of face are found from horizontal and vertical components of the edge map of the segmented image. other facial features are obtained from edge information of the image. The characteristic vectors are extrated from DWT of the segmented face image. These characteristic vectors are normalized between +1 and -1, and are used as input vectors for the neural network. Simulation results show recognition rate of 100% on the learned system, and about 92% on the test images.

• Journal of the Korean Society for Precision Engineering
• /
• v.10 no.4
• /
• pp.170-179
• /
• 1993

In this study, self-organized neural network is used to solve the vorrespondence problem of the axial stereo image. Edge points are extracted from a pair of stereo images and then the edge points of rear image are assined to the output nodes of neural network. In the matching process, the two input nodes of neural networks are supplied with the coordi- nates of the edge point selected randomly from the front image. This input data activate optimal output node and its neighbor nodes whose coordinates are thought to be correspondence point for the present input data, and then their weights are allowed to updated. After several iterations of updating, the weights whose coordinates represent rear edge point are converged to the coordinates of the correspondence points in the front image. Because of the feature map properties of self-organized neural network, noise-free and smoothed depth data can be achieved.

• The KIPS Transactions:PartA
• /
• v.17A no.5
• /
• pp.213-220
• /
• 2010

In this paper, a mechanism connecting all input edges with minimum length through Steiner tree is proposed. Edges are convertible into communication lines, roads, railroads or trace of moving object. Proposed mechanism could be applied to connect these edges with minimum cost. In our experiments where input edge number and maximum connections per edge are used as input parameters, our mechanism made connection length decrease average 6.8%, while building time for a connecting solution increase average 192.0% comparing with the method using minimum spanning tree. The result shows our mechanism might be well applied to the applications where connecting cost is more important than building time for a connecting solution.

• Journal of Korea Multimedia Society
• /
• v.7 no.12
• /
• pp.1745-1753
• /
• 2004

A method of estimating the pose of a three-dimensional object from a set of two-dimensioal images based on parametric eigenspace method is proposed. A Gaussian blurred edge image is used as an input image instead of the original image itself as has been used previously. The set of input images is compressed using K-L transformation. By comparing the estimation errors for the original, blurred original, edge, and blurred edge images, we show that blurring with the Gaussian function and the use of edge images enhance the data compression ratio and decrease the resulting from smoothing the trajectory in the parametric eigenspace, thereby allowing better pose estimation to be achieved than that obtainable using the original images as it is. The proposed method is shown to have improved efficiency, especially in cases with occlusion, position shift, and illumination variation. The results of the pose angle estimation show that the blurred edge image has the mean absolute errors of the pose angle in the measure of 4.09 degrees less for occlusion and 3.827 degrees less for position shift than that of the original image.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.10
• /
• pp.2464-2478
• /
• 2013

Image stitching techniques produce an image with a wide field-of-view by aligning multiple images with a narrow field-of-view. While conventional algorithms successfully stitch images with a small parallax, structure misalignment may occur when input images contain a large parallax. This paper presents an image stitching algorithm that aligns images with a large parallax by regional linear warping. To this end, input images are first approximated as multiple planar surfaces, and different linear warping is applied to each planar surface. For approximating input images as multiple planar surfaces, the concept of dominant edges is introduced. Dominant edges are defined as conspicuous edges of lines in input images, and extracted dominant edges identify the boundaries of each planar surface. Dominant edge extraction is conducted by detecting distinct changes of local characteristics around strong edge pixels. Experimental results show that the proposed algorithm successfully stitches images with a large parallax without structure misalignment.

• Journal of the Semiconductor & Display Technology
• /
• v.15 no.3
• /
• pp.67-71
• /
• 2016

This paper addresses a problem of defocus map recovery from single image. We describe a simple effective approach to estimate the spatial value of defocus blur at the edge location of the image. At first, we perform a re-blurring process using Gaussian function with input image, and calculate a gradient magnitude ratio with blurring amount between input image and re-blurred image. Then we get a full defocus map by propagating the blur amount at the edge location. Experimental result reveals that our method outperforms a reliable estimation of depth map, and shows that our algorithm is robust to noise, inaccurate edge location and interferences of neighboring edges within input image.

• The KIPS Transactions:PartA
• /
• v.16A no.6
• /
• pp.509-518
• /
• 2009

In this paper, a mechanism that produces an approximation edges minimum spanning tree swiftly using virtual nodes called portals dividing given edges into same distance sub-edges. The approximation edges minimum spanning tree can be used in many useful areas as connecting communication lines, road networks and railroad systems. For 3000 random input edges, when portal distance is 0.3, tree building time decreased 29.74% while the length of the produced tree increased 1.8% comparing with optimal edge minimum spanning tree in our experiment. When portal distance is 0.75, tree building time decreased 39.96% while the tree length increased 2.96%. The result shows this mechanism might be well applied to the applications that may allow a little length overhead, but should produce an edge connecting tree in short time. And the proposed mechanism can produce an approximation edge minimum spanning tree focusing on tree length or on building time to meet user requests by adjusting portal distance or portal discard ratio as parameter.

• Journal of Korea Multimedia Society
• /
• v.8 no.3
• /
• pp.313-321
• /
• 2005

Digital halftoning is a process to produce a binary image so that the original image and its binary counterpart appear similar when observed from a distance. Among digital halftoning methods, error diffusion is a procedure for generating high quality bilevel images from continuous-tone images but blurs the edge information in the bilevel images. To solve this problem, we propose the improved error diffusion using local spatial information of the original images. Based on the fact that the human vision perceives not a pixel but local mean of input image, we compute edge enhancement information(EEI) by appling the ratio of a pixel and its adjacent pixels to local mean. The weights applied to local means is computed using the ratio of local activity measure(LAM) to the difference between input pixels of 3$\times$3 blocks and theirs mean. LAM is the measure of luminance changes in local regions and is obtained by adding the square of the difference between input pixels of 3$\times$3 blocks and theirs mean. We add the value to a input pixel of quantizer to enhance edge. The performance of the proposed method is compared with conventional methods by measuring the edge correlation. The halftone images by using the proposed method show better quality due to the enhanced edge. And the detailed edge is preserved in the halftone images by using the proposed method. Also the proposed method improves the quality of halftone images because unpleasant patterns for human visual system are reduced.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.6B
• /
• pp.993-1001
• /
• 2000

This research features a method that quantitatively evaluates the performance of edge detection algorithms. Contrary to conventional methods that evaluate the performance of edge detection as a function of the amount of noise added to he input image, the proposed method is capable of assessing the performance of edge detection algorithms based on chosen parameters that influence the performance of edge detection. We have proposed a quantitative measure, called average performance index, that compares the average performance of different edge detection algorithms. We have applied the method to the commonly used edge detectors, Sobel, LOG(Laplacian of Gaussian), and Canny edge detectors for noisy images that contain straight line edges and curved line edges. Two kinds of noises i.e, Gaussian and impulse noises, are used. Experimental results show that our method of quantitatively evaluating the performance of edge detection algorithms can facilitate the selection of the optimal dge detection algorithm for a given task.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.1
• /
• pp.10-14
• /
• 1998

In this paper a double edge triggered (DET) filp-flop is proposed which changes its output state at both the positive and the negative edge transitions of the triggering input. DET filp-flop has advantages in terms of speed and power dissipation over single edge triggered (SET) filp-flop has proposed DET flip-flop needs only 12 MOS transistors and can operate at clock speed of 500 MHz. Also, the power dissipation has decreased about 33% in comparison to SET flip-flop.