• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1998년도 추계종합학술대회 논문집
• /
• pp.731-734
• /
• 1998

A scalable parallel algorithm is proposed for efficient image component labeling with local operatos on a mesh connected SIMD computer. In contrast to the conventional parallel labeling algorithms, where a single pixel is assigned to each PE, the algorithm presented here is scalable and can assign m$\times$m pixel set to each PE according to the input image size. The assigned pixel set is converted to a single pixel that has representative value, and the amount of the required memory and processing time can be highly reduced. For N$\times$N image, if m$\times$m pixel set is assigned to each PE of P$\times$P mesh, where P=N/m, the time complexity due to the communication of each PE and the computation complexity are reduced to O(PlogP) bit operations and O(P) bit operations, respectively, which is 1/m of each of the conventional method. This method also diminishes the amount of memory in each PE to O(P), and can decrease the number of PE to O(P2) =Θ(N2/m2) as compared to O(N2) of conventional method. Because the proposed parallel labeling algorithm is scalable, we can adapt to the increase of image size without the hardware change of the given mesh connected SIMD computer.

• 융합신호처리학회논문지
• /
• 제16권1호
• /
• pp.1-8
• /
• 2015

In many image processing tasks, connected component labeling (CCL) is performed to extract regions of interest. CCL was usually done in a sequential fashion when image resolution was relatively low and there are small number of input channels. As image resolution gets higher up to HD or Full HD and as the number of input channels increases, sequential CCL is too time-consuming to be used in real time applications. To cope with this situation, parallel CCL framework was introduced where multiple cores are utilized simultaneously. Several parallel CCL methods have been proposed in the literature. Among them are NSZ label equivalence (NSZ-LE) method[1], modified 8 directional label selection (M8DLS) method[2], and HYBRID1 method[3]. Soh [3] showed that HYBRID1 outperforms NSZ-LE and M8DLS, and argued that HYBRID1 is by far the best. In this paper we propose an improved hybrid parallel CCL algorithm termed as HYBRID2 that hybridizes M8DLS with label backtracking (LB) and show that it runs around 20% faster than HYBRID1 for various kinds of images.

• 드라이브 ㆍ 컨트롤
• /
• 제20권4호
• /
• pp.92-99
• /
• 2023

Autonomous vehicles should recognize and respond to the specified speed to drive in compliance with regulations. To recognize the specified speed, the most representative method is to read the numbers of the signs by recognizing the speed signs in the front camera image. This study proposes a method that utilizes YOLO-Labeling-Labeling-EfficientNet. The sign box is first recognized with YOLO, and the numeric digit is extracted according to the pixel value from the recognized box through two labeling stages. After that, the number of each digit is recognized using EfficientNet (CNN) learned with the virtual environment dataset produced directly. In addition, we estimated the depth of information from the height value of the recognized sign through regression analysis. We verified the proposed algorithm using the virtual racing environment and GTSRB, and proved its real-time performance and efficient recognition performance.

• 한국통신학회논문지
• /
• 제28권9C호
• /
• pp.860-867
• /
• 2003

본 논문에서는 방향 패턴 레이블링을 이용하여 지문 영상의 중심점을 검출하는 방법을 제안하였다. 중심점은 지문영상에서의 특이점들 중의 하나이며 대부분의 지문 인식 시스템에서 기준점으로 사용되고 있다. 중심점의 검출은 지문 인식 시스템에서 반드시 수행되어야할 중요한 단계로 전체 시스템의 성능에 큰 영향을 준다. 제안된 방법에서는 ridge의 분포로부터 얻어낸 방향 성분에 레이블링 방법과 중심점의 위치를 결정하는 알고리즘을 적용하여 중심점의 위치를 검출할 수 있었다. 모의 실험 결과 제안한 방법이 Poincare index와 Sine map 방법들에 비해 수행시간과 검출률 모두에서 좀더 나은 성능을 보임을 확인하였다. 특히 제안한 방법은 arch 형의 중심점 검출에 있어 Poincare index 방법의 낮은 검출률과 Sine map 방법의 긴 수행 시간이라는 단점들을 모두 극복하였다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2008년도 하계종합학술대회
• /
• pp.1117-1118
• /
• 2008

The LightScribe Technology is for printing images on the label side of recordable media using CD laser diode. By implementing Optimum Labeling Power Calibration for LightScribe, Labeling Quality can be improved. This paper proposes a new laser power calibration method using RFSUM signal. This function is implemented based on GH22LP20 of LG Electronics.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.241-241
• /
• 2000

In this Paper, we present a new efficient algorithm that can segment an object in the image. There are many algorithms for segmentation and many studies for criteria or threshold value. But, if the environment or brightness is changed, their would not be suitable. Accordingly, we apply a clustering algorithm for adopting and compensating environmental factors. And applying labeling method, we try arranging segment by the similarity that calculated with the fuzzy algorithm. we also present simulations for searching an object and show that the algorithm is somewhat more efficient than the other algorithm.

• 정보과학회 논문지
• /
• 제41권9호
• /
• pp.699-706
• /
• 2014

XML 문서가 점차 복잡해지면서 XML문서의 구조를 알 필요 없이 키워드로만 검색을 하는 키워드 검색 방식이 많이 사용되고 있다. XML문서 내에서 키워드 검색 방식을 사용하기 위해서는 문서 내의 모든 키워드에 레이블을 부여해야 하며, 구조적인 정보 또한 레이블 내에 충분히 표현해야한다. 하지만 기존 레이블링 방법들은 색인을 위한 단순정보만 레이블링 하거나, 증가하는 XML문서의 크기에 대응하기 어려운 형태로 구조적인 정보를 표현한다. 이는 XML문서가 커질수록 키워드검색성능이 떨어지거나, 공간 사용량이 기하급수적으로 증가하는 문제를 야기한다. 따라서 본 논문에서는 대용량 XML문서에 대한 키워드 검색 시 기존 레이블링 방식이 가지고 있던 문제점을 보완하는 새로운 레이블링 방식인 RPLS(Repetitive Prime Labeling Scheme)을 소개한다. 이 방법은 기존 소수 레이블방식을 개선하여 상위 레벨의 소수를 하위 레벨에서 반복 사용할 수 있도록 하여 레이블링을 위해 생성해야하는 소수의 수를 감소시키도록 한 것이다. 본 논문에서는 대용량 XML 문서의 키워드검색에 대한 RPLS 스킴의 효율성 검증을 위해 기존 레이블링 기법들과의 성능 비교 실험 결과도 제시한다.

• Archives of Pharmacal Research
• /
• 제27권9호
• /
• pp.961-967
• /
• 2004

The development of an antibody labeling method with $^{99m}$Tc is important for cancer imaging. Most bifunctional chelate methods for $^{99m}$Tc labeling of antibody incorporate a $^{99m}$Tc chelator through a linkage to lysine residue. In the present study, a novel site-specific $^{99m}$Tc labeling method at carbohydrate side chain in the Fc region of 2 antibodies (T101 and rabbit anti-human serum albumin antibody (RPAb)) using dihydrazinophthalazine (DHZ) which has 2 hydrazino groups was developed. The antibodies were oxidized with sodium periodate to pro-duce aldehyde on the Fc region. Then, one hydrazine group of DHZ was conjugated with an aldehyde group of antibody through the formation of a hydrazone. The other hydrazine group was used for labeling with $^{99m}$Tc. The number of conjugated DHZ was 1.7 per antibody. $^{99m}$Tc labeling efficiency was 46-85％ for T101 and 67∼87％ for RPAb. Indirect labeling with DHZ conjugated antibodies showed higher stability than direct labeling with reduced antibodies. High immunoreactivities were conserved for both indirectly and directly labeled antibodies. A biodistribution study found high blood activity related to directly labeled T1 01 at early time point as well as low liver activity due to indirectly labeled T101 at later time point. However, these findings do not affect practical use. No significantly different biodistribution was observed in the other organs. The research concluded that DHZ can be used as a site-specific bifunctional chelating agent for labeling antibody with $^{99m}$Tc. Moreover, $^{99m}$Tc labeled antibody via DHZ was found to have excellent chemical and biological properties for nuclear medicine imaging.edicine imaging.

• 한국생활과학회지
• /
• 제6권2호
• /
• pp.115-120
• /
• 1997

To investigate the cognition of labeling system and its problems, problems in laundry, and consumer's dissatisfaction and to decrease problems in management of clothing products, total 476 subjects were surveyed in Taejon. The major results were as follows ; 1. The cognition of labeling system was understood by 93.1% of respondents, and 72.9% of them prefer to recognized labeling by figures and letters than by either of them. Most of respondents got the knowledge of labeling system from school. The most frequently experienced mislabeling was the label for management. 2. In cleaning, 60.4% of respondents made their decisions of the laundry method after seeing labeling system. When the label recommanded either of hand washing or dry cleaning, they usually laundered with hand washing after a few times of dry cleaning. The first consideration factor for laundry was fiber composition of textiles. 3. About 70% of respondents understood ironing, laundry, and drying mark on labeling system, and 53.8% understood fiber composition and bleaching. 4. More than 90% of respondents experienced dissatisfaction in handling clothing products. The reason of dissatisfaction was deformation and decoloring after laundry. Most of respondents experienced change of tactile sensation, too.

• 정보과학회 논문지
• /
• 제44권7호
• /
• pp.686-691
• /
• 2017

텔레마케팅은 지식정보화 사회가 되면서 기업 마케팅 활동의 중심축으로 발전하였다. 최근 금융 데이터에 기계학습을 적용하는 연구가 활발하게 진행되고 있으며 좋은 성과를 내고 있다. 하지만 지도학습법이 대부분이어서 많은 양의 클래스가 있는 데이터가 필요하다. 본 논문에서는 텔레마케팅의 목표 고객을 선정하는데 클래스가 없는 금융 데이터에 자동으로 클래스를 부여하는 방법을 제안한다. 준지도 학습법 중 레이블 전파와 의사결정나무 기반의 협동 학습으로 클래스가 없는 데이터를 레이블링한다. 신뢰도가 낮은 데이터를 제거한 후 두 방법이 같은 클래스로 예측한 데이터만 추출한다. 이를 학습 데이터에 추가한 후 의사결정나무를 학습하여 테스트 데이터로 평가한다. 제안하는 방법의 유용성을 입증하기 위해 실제 포르투갈 은행의 텔레마케팅 데이터를 이용하여 실험을 수행하였다. 비교 실험 결과, 정확도가 83.39%로 1.82% 향상되고, 정밀도가 19.37%로 2.67% 향상되었으며, t-검증을 통해 유의미한 성능 향상이 있음을 입증하였다.