• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제26권4호
• /
• pp.263-279
• /
• 2022

This paper presents an automatic inspection of defects in semiconductor images. We devise a statistical method to find defects on homogeneous background from the observation that it has a log-normal distribution. If computer aided design (CAD) data is available, we use it to construct a signed distance function (SDF) and change the pixel values so that the average of pixel values along the level curve of the SDF is zero, so that the image has a homogeneous background. In the absence of CAD data, we devise a hybrid method consisting of a model-based algorithm and two neural networks. The model-based algorithm uses the first right singular vector to determine whether the image has a linear or complex structure. For an image with a linear structure, we remove the structure using the rank 1 approximation so that it has a homogeneous background. An image with a complex structure is inspected by two neural networks. We provide results of numerical experiments for the proposed methods.

• 한국멀티미디어학회논문지
• /
• 제19권2호
• /
• pp.479-486
• /
• 2016

In this paper, the algorithm which can convert a 2D image into a 3D Model will be discussed. The 2D picture drawn by a user is scanned for image processing. The Canny algorithm is employed to find the contour. The waterfront algorithm is proposed to find foreground image area. The foreground area is segmented to decompose the complex shapes into simple shapes. Then, simple segmented foreground image is converted into 3D model to become a complex 3D model. The 3D conversion formular used in this paper is also discussed. The generated 3D model data will be useful for 3D animation and other 3D contents creation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권3호
• /
• pp.1650-1669
• /
• 2017

In this study, each pixel in an ear is used as a centroid to generate a cake. Subsequently the major axis length of this cake is computed and obtained. This obtained major axis length serves as a feature to recognize an ear. Later, the ear hole is used as a centroid and a 16-circle template is generated to extract the major axis lengths of the ear. The 16-circle template extracted signals are used to recognize an ear. In the next step, a ring-to-line mapping technique is used to map these major axis lengths to several straight-line signals. Next, the complex plane vector computing technique is used to determine the similarity of these major axis lengths, whereby a solution to the image-rotating problem is achieved. The aforementioned extracted signals are also compared to the ones that are extracted from its neighboring pixels, whereby solving the image-shifting problem. The algorithm developed in this study can precisely identify an ear image by solving the image rotation and image shifting problems.

• 복식
• /
• 제50권4호
• /
• pp.131-146
• /
• 2000

The starting point of this study was that the Orientalism reflected in the modern clothing is realized the different view by Western and Eastern. Based on this regard, this study semiotically analyse that clothing is creating the cultural values and signification and that the western designers are reproducing an ideology as orientalism being the western-oriented thought when expressing the Asian image in the modern clothing. The theoretical sides are based on the semiotic analyses of F. Saussure, C. Peirce, and R. Barthes. By the result observed through this study, the Orientalism reflected in the modern clothing designed by Western designers are discriminative image, mystic image, past-oriented image, natural image, and complex-eclectic image. The discriminative image is reproduced as a savage and uncivilized image. In addition, it regarded as a decorative and superficial image is expressed as if it is an uncharacterized and underdeveloped image in the modern clothing. The mystic image is reproduced as a romantic image to stimulate Westerners'curiosity. That is, Asian image is regarded as a dream and magical image rather than a scientific and rational image. The past-oriented image is expressed through restricting Asian positive image in Asian historical glory. The natural image differs from Western image based on science in the modern clothing. The modern clothing regards the natural image as a primitive and instinctive image. The complex-eclectic image is the pastish pattern through selecting and distributing heterogeneous materials in the modern clothing. This study critically analyzes the western-oriented thought of orientalism in the modern clothing. The results of the study imply that it is critical that we provide a new contact point for the East and the West in the context of our Asian values and traditions.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2008년도 추계종합학술대회 B
• /
• pp.455-458
• /
• 2008

이전의 정보화는 글이나 혹은 음성에 의존했다면, 현대사회의 정보전송은 다양한 영상 매체를 이용하여 전송하고 있다. 휴대폰과 TV, 컴퓨터는 대표적인 영상신호를 이용하는 매개체로서 현대사회를 이루는 큰 축이라고 할 수 있다. 이러한 이유로 중요성이 부각되어지는 영상 신호의 개발은 크게 압축 및 인식 그리고 복원 등 많은 부분에서 연구가 되어지고 있다. 노이즈는 이러한 신호를 이용함에 따라 필연적으로 발생되며, 발생되는 노이즈로서는 임펄스 노이즈(Impulse Noise)와 AWGN(Additive White Gaussian Noise)가 대표적이다. 이러한 노이즈를 줄이기 위하여 다양한 필터가 개발되고 있으며, 각기 그 잡음의 성향에 따라 다른 필터가 사용되어진다. 그러나 잡음은 신호에서 독립적으로 발생되어지는 것이 아니라 중첩되어 발생되어진다. 본 논문은 이러한 중첩된 잡음을 제거하고자 영상필터를 제안하였으며, 이를 기존의 다른 필터와 비교하였다.

• 디지털융복합연구
• /
• 제15권4호
• /
• pp.267-273
• /
• 2017

각종 디지털 기기와 통신 기술의 발전으로 다양한 멀티미디어 콘텐츠의 생산과 유통이 폭발적으로 증가하고 있다. 이미지와 동영상 등의 멀티미디어 데이터의 검색을 위해서는 기존의 문자 위주의 검색과는 다른 접근 방식이 필요하다. 이미지의 여러 가지 물리적인 특징들을 정량화 하여 분석하고 이를 비교하여 유사한 이미지를 검색하는 내용기반 이미지 검색에서 색상과 모양은 주요 물리적 특징들이다. 지금까지는 색상과 모양을 서로 독립적인 특징으로 분리하여 이용하였지만, 인지적 관점에서 두 특징은 밀접한 관련이 있다. 본 논문에서는 색상과 모양 특징을 동시에 표현하기 위하여 3차원 색상 정보를 2차원 복소수 형식으로 표현하는 복소 색상 모델을 이용하여 색상의 공간적 분포 모양을 기술하는 방법을 제안한다. 복소 이미지를 주파수 변환한 후 저주파 영역의 소수의 계수만으로 복원하는 실험을 통하여 제안된 방법이 색상의 공간적 분포 모양을 효율적으로 표현할 수 있음을 보였다.

• 한국농업기계학회:학술대회논문집
• /
• 한국농업기계학회 1996년도 International Conference on Agricultural Machinery Engineering Proceedings
• /
• pp.773-779
• /
• 1996

Dried oak mushroom have complex and various visual features. Grading and sorting of dried oak mushrooms has been done by the human expert. Though actions involved in human grading looked simple, a decision making underneath the simple action comes from the result of the complex neural processing of the visual image. Through processing details involved in human visual recognition has not been fully investigated yet, it might say human can recognize objects via one of three ways such as extracting specific features or just image itself without extracting those features or in a combined manner. In most cases, extracting some special quantitative features from the camera image requires complex algorithms and processing of the gray level image requires the heavy computing load. This fact can be worse especially in dealing with nonuniform, irregular and fuzzy shaped agricultural products, resulting in poor performance because of the sensitiveness to the crisp criteria or specific ules set up by algorithms. Also restriction of the real time processing often forces to use binary segmentation but in that case some important information of the object can be lost. In this paper, the neuro net based real time recognition algorithm was proposed without extracting any visual feature but using only the directly captured raw gray images. Specially formated adaptable size of grids was proposed for the network input. The compensation of illumination was also done to accomodate the variable lighting environment. The proposed grading scheme showed very successful results.

• 한국주거학회논문집
• /
• 제8권3호
• /
• pp.141-151
• /
• 1997

This study was conducted to examine the environmental factors of apartments and therefore to suggest elements necessary to plan environments by evaluating and analysing both reside environments and conditions with residents at high-rise apartments in Kumi city. Kyongbuk. Korea. The result of this study is summarized as follows: 1) The motive of residence at apartment showed mainly the better environment of nature. and that the locational condition of those apartments was considered as a good standard for evaluating the apartment. 2) The evaluation of satisfaction showed that residents residing at apartment A tended to be satisfaied with all factors other than ‘complex’ although there was wholly dissatisfaction. In addition. it was shown that apartment B was satisfaied with housing and complex. apartment C with comfort. respectively. and apartment D with housing. 3) The evaluation of image showed that apartment C indicated the lowest variety of its complex while other factors were highly considered. Their image concerning residence was highest in apartment A. most lowest in apartment C. Then. apartment B showed the highest familiarity with and recognition of apartment complex.

• Journal of Electrical Engineering and Technology
• /
• 제12권4호
• /
• pp.1600-1604
• /
• 2017

A new Complex Image Method based on Genetic Algorithm (GA) is proposed to calculate the Green's functions of CPW (coplanar waveguide)-type microwave components and antennas. The closed-forms of the spectral-domain integrals are obtained by the GA, avoiding the conventional procedures of the tedious linear algebra and the sampling conditions sensitive to the complex-variable sampling paths adopted in the Prony's and GPOF methods. The proposed method is compared with the numerical Sommerfeld Integral, which results in good agreement.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2013년도 추계학술대회
• /
• pp.97-100
• /
• 2013

본 논문은 복잡한 배경 영상에서 필기체가 아닌 수평, 수직 성분이 많이 포함된 문자 검출 방법을 제안한다. 본 논문에서 검출하고자 하는 문자는 코너 성분이 많이 밀집되어 있으며 배경 영상은 그에 비해 코너 성분이 적고 드문드문하다는 특징을 이용하여 먼저 해리스 코너 검출기를 이용하여 전체 영상에서 코너를 검출한다. 검출된 코너들의 위치 정보를 이용해 밀집되어 있는 코너들을 클러스터링 함으로써 문자 영역을 검출한다. 검출된 문자 영역간의 위치 정보와 히스토그램 분포를 비교하여 비슷한 특징을 갖는 영역들을 합치고 문자 성분의 특징을 갖지 않는 영역은 필터링 하여 문자 영역을 개선한다. 문자 영역에서 R채널, G채널, B채널 각각의 채널에 대한 히스토그램 분포를 분석하여 문자를 검출한다.