• Proceedings of the IEEK Conference
• /
• 2000.09a
• /
• pp.901-904
• /
• 2000

A direct authentication from gray-scale image, instead of the conventional multi-step preprocessing, is proposed using Gabor filter-based features from the gray-scale fingerprint around core point. The core point is located as a reference point for the translation invariant matching. And its principal symmetry axis is detected for the rotation invariant matching from its neighboring region centered at the core point. And then fingerprint is divided into non-overlapping blocks with respect to the core point and features are directly extracted form the blocked gray level fingerprint using Gabor filter. The proposed fingerprint authentication is based on the Euclidean distance between the corresponding Gabor features of the input and the template fingerprints. Experiments are conducted on 300${\times}$300 fingerprints obtained from a CMOS sensor with 500 dpi resolution, and the proposed method could lower the False Reject Rate(FRR) to 18.2% under False Acceptance Rate(FAR) of 0%.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.22 no.5
• /
• pp.597-606
• /
• 1998

The purposes of the study were 1) to evaluate the visual image of colored Korean traditional dress for woman 2) to analyze the colors and, color combinations effect on the image perception using gestalt theory. The research method was a quasi-experimental with a between subjects design. The experimental materials developed for the study were a set of stimuli and a response scale. The stimuli was consisted of 17 drawings of females wearing Korean tradinational dress, by using CAD simulation. A response scale consisted of semantic differential scales. The subjects were 1138 undergraduate students of Taejon city, Chungnam province and Chungbuk province. Their responses to the semantic differential scales were analyzed using factor analysis, one-way ANOVA, Duncan's multiple range test, 1-test. Results were as follows; 1) The image of the stimulus was consisted of the 4 different dimensions.(sociability, evaluation, visibility, attractiveness) 2) Clothing colors had significant effects on image perception of the evaluation dimension, visibility dimension and attractiveness dimension in the mono-color set. The blue showed the most positive image on the evaluation dimension, and the yellow and the gray showed negative image on the same dimension. The yellow showed the most salient image and the gray showed the least salient image on the visibility dimension. The red showed the most attractive image and the green showed the least attractive image on the attractiveness dimension. 3) In hi-color set stimulus, the perceived image was influenced by color combinations. The yellow blouse-the red skirt set showed the most sociable image on the sociability dimension. The blue blouse-the green skirt set showed the most positive image on the evaluation dimension. The yellow blouse-the red skirt set showed the most salient image and the blue blouse-the green skirt set showed the least salient image on the visibility dimension. And the red blouse-the yellow skirt set showed the most attractive image on the attractiveness dimension. On conclusion the visual image of Korean traditional dress wearer was affected by dress colors and color combinations.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.7
• /
• pp.569-573
• /
• 2001

An approximate representation of discrete functions {f$_{i,j}\mid$｜i, j=-1, 0, 1, …, N+1}in two variables by a fuzzy system is described. We use the cubic B-splines as fuzzy sets for the input fuzzification and spike functions as the output fuzzy sets. The ordinal number of f$_{i,j}$ in the sorted list is taken to be the out put fuzzy set number in the (i, j) th entry of the fuzzy rule table. We show that the fuzzy system is an exact representation of the cubic spline function s(x, y)=$\sum_{N+1}^{{i,j}=-1}f_{i,j}B_i(x)B_j(y)$ and that the approximation error S(x, y)-f(x, y) is surprisingly O($h^2$) when f(x, y) is three times continuously differentiable. We prove that when f(x, y) is a gray scale image, then the fuzzy system is a smoothed representation of the image and the original image can be recovered exactly from its fuzzy system representation when it is a digitized image.e.

• Proceedings of the IEEK Conference
• /
• 2002.07c
• /
• pp.1811-1814
• /
• 2002

In this paper, we propose an improved contrast control method f9r image improvement of multi-gray scale image. The proposed contrast control method can improve contrast of image by changing gradient of weight as the type of input image. In addition, the proposed method does not require field and frame memory for large amount of computed data. And, the proposed method can be easily applied to the FPD fur real-time processing because of its less hardware complexity than that of the conventional methods. Also it can flexibly control the contrast of input gray level by varying the weight values that control the contrast range. The operation and performance of the proposed controller have been verified using computer simulation.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.05a
• /
• pp.631-634
• /
• 2003

오늘날 CT나 MR등을 통한 의학 영상 기술과 컴퓨터 성능의 향상으로 인체 내부 장기의 영상을 비교적 용이하게 얻을 수 있으며 얻어진 영상 정보는 컴퓨터로 수치화 되므로 데이터의 조작 및 가공이 용이하다. 그러나, 이렇게 얻어진 의학 영상들은 보통 2 차원적 슬라이스 image 형태로 얻어진다. 일반적으로 슬라이스 사이의 간격은 조사량 등 여러 문제 때문에, 항상 동일한 간격을 유지하고 있지 않은 경우가 많으며 슬라이스 사이 간격이 슬라이스 내의 픽셀 간격보다 큰 경우가 대부분이다. 이러한 image로부터 3 차원적 디스플레이나, 조작, 분석을 하기 위해서는 같은 간격의 image를 얻어야 한다. 이러한 이유로 인하여 보간(Interpolation) 기법이 의학 영상 분야에서 많이 사용된다. 본 논문은 형태-기반 보간 방법을 gray-scale image 에 적용이 가능하도록 확장한 그레이 수준 형태 기반 보간 알고리즘을 구현하였다. 그리하여, 본 논문이 제안한 알고리즘을 슬라이스 간격이 큰 2차원 복부 CT 영상에 적용시켜 다른 보간 법들보다 향상된 결과를 확인할 수 있었다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.3
• /
• pp.708-713
• /
• 2004

A method of individual vehicle detection using gray scale image acquired from a high position is proposed for guidance of incoming vehicles to vacant cells in a parking lot and other similar purposes. With the proposed method, each image region corresponding to a cell is fragmented according to density(gray level), and the distribution of segment area is analyzed to decide if a vehicle is present. The proposed method was tested on an actual outdoor parking lot during 2 days with different weather conditions from sunrise through sunset.

• Journal of KIISE:Software and Applications
• /
• v.31 no.6
• /
• pp.792-798
• /
• 2004

In this paper, we will introduce a new corner detector for the wafer die position recognition. The die position recognition procedure is necessary for WSCSP(Wafer Scale Chip Scale Packaging) technology, decide the accuracy of post-procedure. We present a hierarchical gray level corner detection method for the recognition of the die position from a wafer image. The new corner detector divides the corner region into many homocentric circles, and calculates the comer response and the angle of direction about each circle to get an accurate toner point. The new corner detector has a hierarchical structure so it can detect comer point more quickly than general gray level corner detector.

• Journal of Information Display
• /
• v.2 no.2
• /
• pp.13-21
• /
• 2001

We propose a universal image processor for a-Si TFT LCD of UXGA class that can display the full screen on the LCD panel with low resolution of video sources such as NTSC, VGA, SVGA, XGA, and SXGA by using the proposed interpolation filter. In addition, we propose a real-time contrast controller for image improvement of multi-gray scale image. The operation of the proposed methods has been verified using Synopsys VHDL and computer simulation. Results show that the proposed methods might be suitable for a UXGA LCD controller for real-time image improvement.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.9
• /
• pp.75-81
• /
• 2014

The integral image is the sum of input image pixel values. It is mainly used to speed up processing of a box filter operation, such as Haar-like features. However, large memory for integral image data can be an obstacle on an embedded hardware environment with limited memory resources. Therefore, an efficient method to store the integral image is necessary. In this paper, we propose a memory size reduction hardware design for integral image. The hardware design is used two methods. It is the new integral image memory and modulo calculation for reducing integral image data. The new integral image memory has additional calculation overhead, but it is not obstacle in hardware environment that parallel processing is possible. In the Xilinx Virtex5-LX330T targeted experimental result, integral image memory can be reduced by 50% on a $640{\times}480$ 8-bit gray-scale input image.

• Journal of Information Processing Systems
• /
• v.16 no.6
• /
• pp.1424-1436
• /
• 2020

Although huge progress has been made in current image segmentation work, there are still no efficient segmentation strategies for tree image which is taken from natural environment and contains complex background. To improve those problems, we propose a method for tree image segmentation combining adaptive mean shifting with image abstraction. Our approach perform better than others because it focuses mainly on the background of image and characteristics of the tree itself. First, we abstract the original tree image using bilateral filtering and image pyramid from multiple perspectives, which can reduce the influence of the background and tree canopy gaps on clustering. Spatial location and gray scale features are obtained by step detection and the insertion rule method, respectively. Bandwidths calculated by spatial location and gray scale features are then used to determine the size of the Gaussian kernel function and in the mean shift clustering. Furthermore, the flood fill method is employed to fill the results of clustering and highlight the region of interest. To prove the effectiveness of tree image abstractions on image clustering, we compared different abstraction levels and achieved the optimal clustering results. For our algorithm, the average segmentation accuracy (SA), over-segmentation rate (OR), and under-segmentation rate (UR) of the crown are 91.21%, 3.54%, and 9.85%, respectively. The average values of the trunk are 92.78%, 8.16%, and 7.93%, respectively. Comparing the results of our method experimentally with other popular tree image segmentation methods, our segmentation method get rid of human interaction and shows higher SA. Meanwhile, this work shows a promising application prospect on visual reconstruction and factors measurement of tree.