• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.6
• /
• pp.709-713
• /
• 2005

A new algorithm is proposed for the extraction of line segments to construct 3D wire-frame models of building from the high-resolution aerial images. The purpose of this study Is the accurate and effective extraction of line segments, considering the problems such as discordance of lines and blurred edges existing in the conventional methods. Using the edge map extracted from aerial images, chain code tracking of edges was performed. Then, we extract the line segments considering the strength of edges and the direction of them. SUSAN (Smallest Uni-value Segment Assimilating Nucleus) algorithm proposed by Smith was used to extract an edge map. The proposed algorithm consists of 4 steps: removal of the horizontal, vertical and diagonal components of edges to reduce non-candidate point of line segments based on the chain code tracking of the edge map, removal of contiguous points, removal of the same angle points, and the extraction of the start and end points to be line segments. By comparing the proposed algorithm with Boldt algorithm, better results were obtained regarding the extraction of the representative line segments of buildings, having relatively less extraction of unnecessary line segments.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.2C
• /
• pp.124-131
• /
• 2007

This paper presents a block loss recovery technique for the image block data corrupted by transmission losses through the employment of fine directional interpolation (FDI). The proposed algorithm introduces a spatial direction vector (SDV). The SDVs are extracted from the edge information of the neighboring image data. Subsequently, the SDVs are adaptively applied to interpolate lost pixels on a pixel-by-pixel basis. This approach improves the capability to more reliably recover high-detailed contents in the corrupted block. Experimental results demonstrate that the FDI method performs better as compared to previous techniques.

• Journal of Korea Multimedia Society
• /
• v.1 no.2
• /
• pp.194-203
• /
• 1998

In this paper, a new DCT-VQ method is proposed which can solve the problems of VQ such as the degradation of edge and enormous calculations. VQ is carried in DCT domain but spatial domain in order to protect the degradation of edge. DCT makes high correlated image data decorrelated and the energy concentrated on a few coefficients. In DCT domain, the DC coefficient is quantized with 8 bits uniform scalar quantizer and the AC coefficients are divided to three regions and coded with vector qiantizer for considering edge components. For the decrease of the calculation and memory, the vectors for three region have small dimension of $1{\times}7$ and use the same codebook. Thus, the proposed method can fully express the edge components by considering AC coefficients in DCT domain and decrease the calculation and memory be reducing the dimension of vectors.

• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.2938-2940
• /
• 1999

도로 색과 대비를 이루는 계수선을 도로위에 그려놓고 영상을 이용하여 통과 차량을 계수 하는 영상기반 교통량 측정 기술을 제안하였다. 기존의 컴퓨터 영상을 이용한 차량 계수 장치들은 대부분 도로와 차량의 gray level차이를 이용하는 방식이기 때문에 도로와 차량간의 구분이 쉽지 않고 그림자와 차량의 식별도 어려우며 조명에 의해 착오를 일으키기 쉬운 문제를 안고 있었다. 제안한 계수방법은 도로 위에 도로 색과 뚜렷한 대비를 이루는 띠를 도로를 가로질러 그려 놓은 다음 이를 통과하는 차량에 의해 띠의 형상이 절단되는지의 여부를 판단하여 차량을 계수 하는 방법이다. 이를 위해서 계수선을 포함하는 영상의 창을 설치하고 수시로 에지를 추출한 후, 에지의 절단 여부를 검사한다. 이때 계수선의 색과 동일한 차량이 통과할 경우에는 계수선의 에지는 절단되어 도로방향으로 연장되게 되며 계수선의 색과 다른 차량인 경우에는 계수선의 에지가 계수선의 내부로 연장되게 된다. 계수선 상에 그림자가 있는 경우에도 도로와 계수선 간의 색상변화가 뚜렷하므로 계수선의 에지 검출에는 영향을 받지 않게 된다. 제안한 알고리즘을 실험실에 설치된 모형도로와 모형 차량들을 이용하여 실험하고 결과를 제시하였다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.100-103
• /
• 2014

Currently, edge detection is utilized in various areas. Edge detection is the preprocessing process for image processing in general, and this is a technology that is considered essential for image processing. According, research on this subject is carried out incessantly. Edge has important image related elements such as size, direction and location of the object of an image. Numerous methods were proposed for the detection. Among them, the representative methods are Sobel, Prewitt, Roberts, Laplacian. However, these existing methods are rather lacking when it comes to the edge detection characteristics in case of the image with mixed noise. Therefore, this study presented edge detection method that utilizes median and average values for the elements depending on the size and location of local mask.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.377-378
• /
• 2021

방향 네트워크 임베딩 문제는 주어진 방향 네트워크의 노드들을 그들 간의 비대칭 관계를 보존할 수 있는 저 차원 벡터들로 표현하는 것이다. 최근, 이 문제에 대한 다양한 방법들이 제안되어왔다. 본 논문에서 우리는 네 가지 실세계 방향 네트워크 데이터셋와 세 가지 에지 예측 시나리오를 이용한 실험을 통해, 최신 방향 네트워크 임베딩 방법들의 성능을 종합적으로 비교 분석한다.

• Journal of Navigation and Port Research
• /
• v.28 no.3
• /
• pp.233-238
• /
• 2004

Comer point is a very important information to a pattern recognition of image processing. And so, many researchers develope various detecting comer point algoritms. But, there are some problems to get comer points by 8 directional chain code when the degree of edge line is not integer multiplication of 45 degree. So, we propose a new algorithm which is combined with morphology and human visual conception for optimal comer points without the above defects. We get a good simulation result by this proposed algorithm Ana so, we think this algorithm is very useful to FA(factory automation} and ship's radar system to know some coastal area from its image.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.9C
• /
• pp.704-711
• /
• 2008

This paper proposes a new de-interlacing method which results in efficient visual improvement. In the proposed algorithm, the distance weight was considered and the previously developed the EDI (Edge Dependent Interpolation) algorithm and the NAL (New Adaptive Linear interpolation) algorithm were used as a basis. The do-interlacing method was divided into two main parts. First, the edge direction was found by using information of closer pixels. Then, missing pixels were interpolated along with the decided edge direction. In this paper, after predicting the edge through the EDI algorithm, missing pixels were interpolated by using the weighted distance based on the NAL algorithm. Experimental results indicate that the proposed algorithm be superior to the conventional algorithms in terms of the objective and subjective criteria.

• Journal of KIISE:Software and Applications
• /
• v.33 no.10
• /
• pp.862-867
• /
• 2006

To restore image degraded by blur and additive noise in the optical and electrical system, a regularized iterative restoration is used. A regularization operator is usually applied to all over the image without considering the local characteristics of image in conventional method. As a result, ringing artifacts appear in edge regions and the noise is amplified in flat regions. To solve these problems we propose an adaptive regularization iterative restoration considering the characteristic of edge and flat regions using directional regularization operator. Experimental results show that the proposed method suppresses the noise amplification in flat regions, and restores the edge more sharply in edge regions.

• Journal of KIISE
• /
• v.42 no.1
• /
• pp.33-43
• /
• 2015

Recently, patent images are growing importance and thus patent image retrieval is a growing area of research. However, most existing patent image retrieval systems use edges extracted in the images, whose performance is affected by the quality of edge detection in the image pre-processing step. To overcome this disadvantage, we propose a SURF-based patent image retrieval method which uses the morphological characteristics of the images. The proposed method detects SURF interest points with directions and computes regional histograms. We apply the proposed method to a patent image database with 2000 binary images and we show the proposed retrieval system achieves excellent results, even when the images have some loss or degradation.