http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
객체 코너의 재배열된 특성요소의 최대 열에 의한 영상 정보 검색
강광원(Gwang-won Kang),이지민(Ji-min Lee),박종안(Jong-an Park) 한국정보기술학회 2010 한국정보기술학회논문지 Vol.8 No.12
본 연구에서는 객체 코너의 재배열된 특성요소의 최대 열에 기반하는 비주얼 정보검색 기법을 제안한다. 제안된 알고리즘은 다음 단계로 진행된다. 먼저 영상 내 객체의 코너 점을 추출하여, 이를 크기순에 따라 순환적으로 재배열하여 특성벡터를 형성한다. 그리고 재배열된 특성벡터에 따라 얻게 되는 최대 열벡터를 기반으로 유사도를 측정하게 된다. 제안된 기법은 영상 내에서 형태 구조가 분명한 객체의 실험에서 성능이 우수하였으고, 객체의 이동이나 회전에도 강인하였으며 코너 패치 히스토그램을 이용한 형태 검색에 비해 약 0.05%의 향상된 recall을 나타내었다. This paper have proposed an algorithm for retrieving a visual information based on rearranged feature vectors of the object corner. The proposed algorithm is processed as follows. First, the corner points of the object in an image are extracted and then the feature vectors are obtained. It are rearranged according to the number dimension and consist of sequence vectors. And the similarity based on the maximum of sequence vectors is measured. The proposed technique is invariant to the rotation or the transfer of the objects and more efficient in case that the objects present simple structure. In simulation that use Wang's database, the method presents that the recall property is improved by 0.05% and more than the standard corner patch histogram.
Image retrieval using local frequency maxima of histogram based color Correlogram
Gwang-won Kang(강광원),Seung-hak Rhee(이승학),Jong-hun Chun(천종훈),Seung-jin Park(박승진),Jong-an Park(박종안) 한국정보기술학회 2008 한국정보기술학회논문지 Vol.6 No.1
Color histograms is a very useful technique used for retrieving and indexing images in a content-based image retrieval (CBIR) system. They provide a graphical representation of global color distribution of an image. However, the histograms are not robust to major changes in the object appearance. There might also be a case where histogram might give same or similar representation for different images, which depends upon the color distribution in the images. Correlograms are used for efficient indexing of images. They not only extract the color distribution of pixels in images like color histogram, but also extracts the spatial information of pixels in the images. The characteristic of the intensity Correlogram are taken into account for extracting the spatial information as well as the distribution of color pixels greatly attracts the researcher for content based image retrieval. In this paper, we propose the image bin (histogram value divisions) separation technique followed by extracting maxima of frequencies and plotting a Correlogram. At first, the histogram is calculated for an image. After that, it is subdivided into four equal bins. Each bin is subdivided into four more bins and for every such subdivision the maxima of frequencies is calculated. This information is stored in the form of a Correlogram. The distance between Correlogram of the query image and the corresponding Correlogram of database images is calculated. The proposed algorithm is tested on a database comprising a large number of images and gives promising results.
RGB 칼라 특징정보와 Median 필터링을 이용한 영상검색
강광원(Gwang-Won Kang),박종안(Jong-An Park) 한국정보기술학회 2007 한국정보기술학회논문지 Vol.5 No.2
This paper presents an image retrieval technique that use RGB color feature information and median filtering to design a more effective algorithm that extracting features of color images for content-based image retrieval. First, extract middle value that use median filtering after segmentation color image into block of regular size. Extraction middle value presents an image retrieval technique using feature after color information do index using difference value and quantization by size order of R, G, B. Compose and use in small feature table by color image feature in this paper that see because need a large amount of storage space and many calculations according to used number of feature at algorithm of effective extract feature. Goodness of fit and order was obtained by comparing of features that were organized into a table, using the proposed algorithm, for the input image and database images. Confirmed that is superior comparing of features that were into a feature table by algorithm that propose and compare input image and DB image and yield fitness and order.
이중구조형 z-슬롯 안테나의 슬롯 형상에 따른 방사패턴 분석
강광원(Gwang-Won Kang),박종안(Jong-An Park),강성관(Sung-Kwan Kang),천종훈(Jong-Hun Chun),안영은(Young-Eun An) 한국정보기술학회 2012 한국정보기술학회논문지 Vol.10 No.9
In this paper, the dual-structure z-slot antenna with radiation pattern according to the shape of slot for UHF RFID Reader was analyzed. In order to analyze characteristics of the dual-structure z-slot antenna, the 12 slots on the radiation plate were shaped under different parameter conditions with diagonal length (L), line length (d), the slot gap (t). The designed antenna and the standard antenna were rotated in the horizontal and vertical directions respectively and then the isotropic pattern was measured by 48 kinds of experimental methods and we have obtained the antenna prameter with optimal isotropic pattern characteristics. It was showed to improve the efficiency of the antenna, as well as to improve the miniaturization of antenna size and the receive sensitivity as a range of the tag through characterization analysis of the dual-structure z-slot antenna for UHF RFID Reader with optimal isotropic pattern.
RGB 최대 주파수 인덱싱과 BW 클러스터링을 이용한 콘텐츠 기반 영상 검색
강지영(Kang, Ji-Young),백정욱(Beak, Jung-Uk),강광원(Kang, Gwang-Won),안영은(An, Young-Eun),박종안(Park, Jong-An) 한국정보전자통신기술학회 2008 한국정보전자통신기술학회논문지 Vol.1 No.2
칼라 코렐로그램은 계산량이 많아지고 저장 공간이 커져서 검색하는 시간이 길어지므로 일반적으로 64*64 로 양자화 하여 사용되어지는데, 본 논문에서는 메디안 픽셀 특징에 공간정보를 이용하여 9*9 로 양자화 하였다. 기존 알고리즘의 경우 메디안 값이 중복되는 경우 중복된 값들을 정렬하여 그 중 가운데 값을 특징자 값으로 정하였으나, 제안된 알고리즘에서는 중복된 값들을 정렬하여 그 중 공간정보가 가장 작은 값을 특징자 값으로 정하였다. 그리고 코렐로그램을 적용하여 특징자 테이블을 구성하고 이를 이용하여 비교하였다. 제안된 알고리즘은 시뮬레이션을 통해 테스트 하였고 그 결과 기존 알고리즘 보다 더 나은 검색성능을 나타내게 되었다. This study proposed a content-based image retrieval system that uses RGB maximum frequency indexing and BW clustering in order to deal with existing retrieval errors using histogram. We split RGB from RGB color images, obtained histogram which was evenly split into 32 bins, calculated and analysed pixels of each area at histogram of R, G, B and obtained the maximum value. We indexed the color information obtained, obtained 100 similar images using the values, operated the final image retrieval system using the total number and distribution rate of clusters. The algorithm proposed in this study used space information using the features obtained from R, G, and B and clusters to obtain effective features, which overcame the disadvantage of existing gray-scale algorithm that perceived different images as same if they have the same frequencies of shade. As a result of measuring the performances using Recall and Precision, this study found that the retrieval rate and priority of the proposed algorithm are more outstanding than those of existing algorithm.
프로그래머블 LNA 기반 RFID 수신감도 파라미터 분석
문구선(Gu-seon Mun),강광원(Gwang-won Kang),장민혁(Min-hyuk Chang),천종훈(Jong-hun Chun),박종안(Jong-an Park) 한국정보기술학회 2008 한국정보기술학회논문지 Vol.6 No.6
This study designed a programmable LNA at the reader receptor to improve performance of UHF bandwidth RFID reader based on a passive transponder. The receiving set is designed to be open when there is a transmit output while when there is no transmit output, a short-cut is done. It is a system that uses the same antenna both for transmit and receiving systems while using physically separate channels. This study analyzed parameters of the cascade noise elements and found that the suggested system showed increased noise elements while noise index of the entire system decreased, 12.56dB and its sensitivity was improved, more than 19.75dB. This study also provided power for the receiving set of the reader and measured signal power shown in the output of a demodulator. It found that when the same input power was used, sensitivity of the suggested system increased, more than 20.81dB, in comparison with input power.
Run Length Coding 기반 정규화 체인코드의 특징벡터 추출
박종안(Jong-An Park),강광원(Gwang-Won Kang) 한국정보기술학회 2012 한국정보기술학회논문지 Vol.10 No.7
The chain code is extensively used for extracting the feature vector in object recognition or image retrieval. It is used to extract the feature vector because the algorithm is not only simple but also includes the contour information. But as the traditional chain code is very sensitive at the transfer/rotation/scaling of the object, it is very difficult to get the efficient result of the recall or the precision. In this paper, we have proposed a novel algorithm that the feature vectors, which is based on the voting numbers of the rearranged chain code, is extracted. The process is as follows. The object contour based chain code is obtained and rearranged and normalized according to the maximum voting number. After round off, The voting value based chain code is used as the feature vector. By using the MPEG-7 database, the proposed method experimented. It could confirm the robustness for the movement of the experimental result object or the rotation and reduction and enlargement than the standard chain code method.