• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2005년도 한국컴퓨터종합학술대회 논문집 Vol.32 No.1 (B)
• /
• pp.226-228
• /
• 2005

Predicting the cellular location of an unknown protein gives valuable information for inferring the possible function of the protein. For more accurate Prediction system, we need a good feature extraction method that transforms the raw sequence data into the numerical feature vector, minimizing information loss. In this paper we propose new methods of extracting underlying features only from the sequence data by computing pairwise sequence alignment scores. In addition, we use composition based features to improve prediction accuracy. To construct an SVM ensemble from separately trained SVM classifiers, we propose specificity based weighted majority voting . The overall prediction accuracy evaluated by the 5-fold cross-validation reached $88.53\%$ for the eukaryotic animal data set. By comparing the prediction accuracy of various feature extraction methods, we could get the biological insight on the location of targeting information. Our numerical experiments confirm that our new feature extraction methods are very useful forpredicting subcellular localization of proteins.

• 정보관리학회지
• /
• 제26권1호
• /
• pp.305-320
• /
• 2009

웹을 통해 제공되는 뉴스 페이지의 경우 필요한 정보 뿐 아니라 많은 불필요한 정보를 담고 있다. 이러한 불필요한 정보는 뉴스를 처리하는 시스템의 성능 저하와 비효율성을 가져온다. 이 연구에서는 웹 페이지로부터 뉴스 콘텐츠를 추출하기 위해 문장과 블록에 기반한 뉴스 기사 추출 방법을 제시하였다. 또한 이들을 결합하여 최적의 성능을 가져올 수 있는 방안을 모색하였다. 실험 결과, 웹 페이지에 대해 하이퍼링크 텍스트를 제거한 후 문장을 이용한 추출 방법을 적용하였을 때 효과적이었으며, 여기에 블록을 이용한 추출 방법과 결합하였을 때 더 좋은 결과를 가져왔다. 문장을 이용한 추출 방법은 추출 재현율을 높여주는 효과가 있는 것으로 나타났다.

• 한국정밀공학회지
• /
• 제10권2호
• /
• pp.54-65
• /
• 1993

The paper describes and alternative method based on a new idea to measure the circular movement of machining centers. ISO has employed three testing methods for the acceptance tests of machine tools; the first is a rotating one-dimensional probe method, the second is a two-dimensional probe and a master circular ring, and the third is a kinematic ball bar. The last two methods were proposed and introduced by W. Knapp and J. B. Bryan, respectively. The newly developed method is superior to above two methods; the rotating angle can be detected and the rotating radius is variable. Circular movement errors of machining centers were investigated by the analysis of data measured by R- .THETA. method. Followint observations are obtained 1) The errors which depend on positions, i.e., periodical errors by the pitch of ball screws, errors by compensation of backlash and errors by perpendicularity of X and Y-axis, were analyzed. 2) The errors which depend on NC control system, i.e., errors by the unbalance of position-loop-gaians, errors by velocity-loop-gains and errors by feed speeds, were quantiatively analyzed. 3) The method of extracting error information, which uses moving technique of averaging angle and fourier's analysis data mesured by the R- .THETA. method, was proposed.

• 로봇학회논문지
• /
• 제19권1호
• /
• pp.65-70
• /
• 2024

Tool wear is considered an important issue in manufacturing and engineering, as worn tools can negatively impact productivity and product quality. Given that the wear status of tools plays a decisive role in the production process, measuring tool wear is a key task. Consequently, there is significant attention in manufacturing fields on the precise measurement of tool wear. Current domestic methods for measuring wear are limited in terms of speed and efficiency, with traditional methods being time-consuming and reliant on subjective evaluation. To address these issues, we developed a measurement module implementing the DeepContour algorithm, which uses image processing technology for rapid measurement and evaluation of tool wear. This algorithm accurately extracts the tool's outline, assesses its condition, determines the degree of wear, and proves more efficient than existing, subjective, and time-consuming methods. The main objective of this paper is to design and apply in practice an algorithm and measurement module that can measure and evaluate tool wear using image processing technology. It focuses on determining the degree of wear by extracting the tool's outline, assessing its condition, and presenting the measured value to the operator.

• International Journal of Internet, Broadcasting and Communication
• /
• 제16권2호
• /
• pp.22-30
• /
• 2024

Image compression-decompression methods have become increasingly crucial in modern times, facilitating the transfer of high-quality images while minimizing file size and internet traffic. Historically, early image compression relied on rudimentary codecs, aiming to compress and decompress data with minimal loss of image quality. Recently, a novel compression framework leveraging colorization techniques has emerged. These methods, originally developed for infusing grayscale images with color, have found application in image compression, leading to colorization-based coding. Within this framework, the encoder plays a crucial role in automatically extracting representative pixels-referred to as color seeds-and transmitting them to the decoder. The decoder, utilizing colorization methods, reconstructs color information for the remaining pixels based on the transmitted data. In this paper, we propose a novel approach to image compression, wherein we decompose the compression task into grayscale image compression and colorization tasks. Unlike conventional colorization-based coding, our method focuses on the colorization process rather than the extraction of color seeds. Moreover, we employ the Denoising Diffusion Null-Space Model (DDNM) for colorization, ensuring high-quality color restoration and contributing to superior compression rates. Experimental results demonstrate that our method achieves higher-quality decompressed images compared to standard JPEG and JPEG2000 compression schemes, particularly in high compression rate scenarios.

• 한국실내디자인학회논문집
• /
• 제21권6호
• /
• pp.204-211
• /
• 2012

The purpose of this study is to analyze characteristics and expression methods of contemporary art museums from the perspective of 'Rhizome', which is considered under similar context as 'indeterminacy' that represents the characteristics of contemporary public spaces. Methods of this study are as follows. First, necessity of this study is verified by investigating literature data and preceding studies, and by examining the fact that indeterminate characteristic is a new approach to contemporary public spaces. Second, in order to analyze Rhizomic expression in public spaces, analysis criteria are suggested by extracting components of public spaces and spatial characteristics of R150hizome from the literature. Third, Rhizomic characteristics and expression methods used in contemporary art museums are understood based on such analysis criteria. Fourth, when each spatial characteristic is expressed in art museums, the study found out that two factors among 'program', 'circulation', and 'form' are planned out as a mixture. Therefore, 'cohesion' is expressed as 'program' and 'form' in contemporary public spaces viewed from the perspective of 'Rhizome'. Also, 'diversity' is actively expressed through 'program' and 'circulation', and 'non-hierarchy' through 'form' and 'circulation'. Such methods are positive methods of expressing indeterminate contemporary public spaces. Since this study conducted analysis on characteristics and expression methods of public spaces from Rhizomic perspective of contemporary society, such results are deemed valuable in planning out public spaces that reflect the characteristics of contemporary society.

• 디지털콘텐츠학회 논문지
• /
• 제16권4호
• /
• pp.583-593
• /
• 2015

오늘날 컴퓨터와 통신 기술의 발달로 멀티미디어(이미지 데이터)는 다양한 응용 분야에서 사용되고 있다. 여기에 가장 널리 사용되고 있는 JPEG2000는 관심영역(ROI) 기술을 제공한다. ROI의 추출은 사용자에게 우선적으로 보여져야 하기 때문에 빠르게 수행되어야 하고 큰 이미지에서 자동적으로 추출되어야 한다. 이를 위해, 본 연구는 JPEG2000의 코드 블록 내에 있는 에지 분포를 이용하여 ROI의 자동 추출과 우선적 처리에 관한 방법을 제안한다. 먼저 에지 추출을 위한 처리와, 다음으로 에지 정보를 이용해 ROI를 자동적으로 추출한다. 그리고, 추출된 ROI 블록을 이용하여 ROI를 그룹핑 하고, ROI 블록의 마스크를 생성한다. 이후에는 양자화를 하고 우선적 처리를 하는 ROI 코딩을 하고 EBCOT를 실행한다. 제안 방법의 유효성을 보이기 위하여 JPEG2000에서 사용되는 다른 ROI 추출 기법들과 비교하고 ROI 코딩을 하지 않는 기법과 ROI 코딩이 포함된 기법 간의 PSNR을 평가하여 품질을 비교한다.

• Archives of Plastic Surgery
• /
• 제34권6호
• /
• pp.777-784
• /
• 2007

Purpose: For a minor degree of mandibular prognathism, mandibular anterior segmental osteotomy (ASO), usually extracting the bilateral premolars, has been performed frequently to correct malocclusion of the anterior teeth. Preoperative planning using cephalometry and a dental model is very important for such a orthognathic surgery. Depending on the specific preoperative mock surgery with the dental model, ASO, with ipsilateral unitooth extraction, is defined to be feasible and performed for ten patients. The comparisons of its preoperative and postoperative analysis of clinical photographs, dental casts, and lateral cephalograms, for soft tissue profiles, skeletal and dental relationships are described in the following, and its clinical applications are noted. Methods: From March 1, 2004, to March 31, 2006, We performed 10 mandibular ASO by extraction of ipsilateral unitooth to improve their lower facial profiles and the lip relationships. Patient age ranged from 19 to 33 years, with a mean age of 25.6 years. Two were males and eight were females. Results: All patients were satisfied with aesthetic and occlusal changes postoperatively. Significant and persistent decrease in the SNB and interincisal angle were observed in the postoperative cephalometries. The soft tissue profiles also were improved and near Ricketts's esthetic line. Other combined procedures include nine genioplasties, two rhinoplasties, and one blepharoplasty. One patient complained of transient unilateral inferior mental nerve paresthesia. There were no other significant complications or relapses throughout the follow-up period(6-20 months). Conclusion: Mandibular ASO, extracting the ipsilateral unitooth, was performed for ten patients to correct mild mandibular prognathism. The amount of setback of the mandibular anterior portion was 2 to 3 mm, and satisfactory results were obtained combined with genioplasties.

• 대한본초학회지
• /
• 제25권4호
• /
• pp.113-120
• /
• 2010

Objectives : This study was conducted to compare Chaenomelis Sinensis Fructus with Chaenomelis Lagenariae Fructus, and to examine Fructus chaenomelis having an influence on Intestinal Motility. Methods : We compared Chaenomelis Sinensis Fructus with Chaenomelis Lagenariae Fructus, by observing their effects on Intestinal Motility. Oral administration of water extracts of Chaenomelis Sinensis Fructus and Chaenomelis Lagenariae Fructus into albino rats was followed by dealing with carbachol or loperamide, injecting charcoal meal and measuring the moving length in the intestine. Results : By Chaenomelis Sinensis Fructus and Chaenomelis Lagenariae Fructus, the Intestinal Motility of normal albino rats did not change significantly. Chaenomelis Sinensis Fructus and Chaenomelis Lagenariae Fructus controled the accelerated Intestinal Motility of albino rats. Chaenomelis Sinensis Fructus was not different from the extracting method, Chaenomelis Lagenariae Fructus had effects of control on extracting by methanol significantly. When Chaenomelis Sinensis Fructus was compared with Chaenomelis Lagenariae Fructus, the former was proved to have control effects on Intestinal Motility in decoction extracts and extracts by ethyl ether higher than the latter. By both Chaenomelis Sinensis Fructus and Chaenomelis Lagenariae Fructus, the declined Intestinal Motility was not changed significantly. Conclusion : Chaenomelis Fructus suppressed the exasperated Intestinal Motility by carbachol, but did not influence the dropped Intestinal Motility by loperamide. In addition, Chaenomelis Sinensis Fructus was more excellent than Chaenomelis Lagenariae Fructus in its control effects of Intestinal Motility. inhibitory effects by DKT represent a potential therapeutic approach to the treatment of inflammatory diseases.

• 정보처리학회논문지B
• /
• 제15B권5호
• /
• pp.493-500
• /
• 2008

개인 통신 방법의 수단으로 전자 메일이 널리 사용되고 있으나, 업무에 직접 관련이 없는 쓸모없는 상업용 메일도 대량으로 유포되고 있다. 본 연구에서는 사용자가 작성한 프로파일을 이용하여 메일을 자동으로 그룹핑(grouping) 하는 방법을 제안하고자 한다. 기존의 연구 방법은 단어의 빈도수만을 이용하는 단일 속성을 이용하므로 높은 정확률을 얻을 수 없었다. 그러나 본 논문에서 제안하는 방법은 기존 사용자의 폴더에 수신된 메일의 분류 체계에서 추출된 사용자 프로파일을 이용하여 그룹핑 되는 메일의 정확률을 높이고자 한다. 본 논문에서 적극적으로 이용하는 정보는 다중 속성(송신처, 문서의 주제, 문서의 유형 정보, 시간제한 표현의 어구 등) 값이다. 사용자의 프로파일을 이용함으로써 새로 도착한 메일이 사용자에게 중요한가 혹은 중요하지 않은가의 자동 판단이 가능하도록 시스템을 설계하였다. 학습 데이터를 네 가지 형태로 나누어 실험한 결과 Rocchio와 Widrow-Hoff의 상관계수가 각각 0.40과 0.43인 것 보다 본 논문의 방법이 0.52로 보다 높은 상관계수를 나타내 빈도수만을 이용하는 기존의 연구보다 정확한 방법임을 알 수 있었다.