• 제목/요약/키워드: k-mean segmentation

검색결과 112건 처리시간 0.042초

엔트로피 기반의 가중치와 분포크기를 이용한 향상된 FCM 알고리즘 (Improved FCM Algorithm using Entropy-based Weight and Intercluster)

  • 곽현욱;오준택;손영호;김욱현
    • 대한전자공학회논문지SP
    • /
    • 제43권4호
    • /
    • pp.1-8
    • /
    • 2006
  • 본 논문은 엔트로피 기반의 가중치와 클러스터 분포크기를 이용한 향상된 FCM(Fuzzy C-Mean)알고리즘을 제안한다. FCM 알고리즘은 영상분할에서 일반적으로 많이 사용되는 퍼지 클러스터링 방법이다. 그러나 공간정보를 포함하지 않기 때문에 잡음 등에 민감하고, 클러스터를 이루는 특정들의 분포에 따라 화소들을 정확하게 분류할 수 없다. 이러한 단점을 해결하기 위해서 FCM 알고리즘의 소속정도를 연산할 때 클러스터 분포크기와 이웃 화소의 공간정보를 이용한 엔트로피 기반의 가중치를 적용한다. 실험결과에서 제안한 방법이 기존의 방법들보다 잡음에 강건하며 분할결과를 보였다.

고객중심의 CRM 구축비교 사례연구 (Customer-Centric CRM Implementation Case Study)

  • 이호섭
    • 경영과정보연구
    • /
    • 제23권
    • /
    • pp.25-40
    • /
    • 2007
  • In the highly competitive and divers world of financial market, customer is the single most important factor to company's survival. Especially, creating a relationship with valued customers is a key to success. CRM provides the mean to retain high value customers. It takes a prospect of what customers expect. Utilizing those knowledge can help the products and service meet the customers' needs, thereby maximizing customer satisfaction and company's profit. In this report, I am going to suggest a few ways to develop successful CRM in the life insurance industry. First, CRM should innovate the way of communication to keep pace with Web 2.0 era. In other words, the customer's needs should be caught by real-time communication than traditional off-line market research. Thus, the functionality and specification of products can be decided by customer's direct choice so that the customers are able to purchase the understanding and experience of the products. Second, CRM project should consider whether the initial strategy plan can promise the stable growth of customer at the first step. When planning strategy, the project needs to identify what customer wants and how to fulfill the needs with stable growth of the customer. In addition, the CRM should be developed by realizing that customer centric benefits ultimately guarantee the growth of the organization. Third, CRM systems should enhance the organization's ability to take the customer's insight in a 360 degree view and to capture the voice of the customer directly. In order to develop the best matched product package, more precise customer segmentation should be ahead of market segmentation strategy. Forth, the biggest reward from CRM will be a customer royalty program. Many successful banks are already planning and practicing customer royalty strategy. A comprehensive analysis of customers and their behavior allow organization to identify high value potential customers' needs and determine a strategy required to meet those needs. Even life insurance companies such as Prudential Korea are developing products designed for royal customers. Fifth, understanding and managing the experience of customer called Customer Experience Management also can increase customer satisfaction. Measuring only customers' experience and adapting it to marketing strategy make products position in the gap between the customers' expectation and experience not required by market. A key component of CEM is its application across all organizational functions. At last, the direction of change and development of CRM can be defined from the conceptualization of information technology represented by Ubiquitous and Web 2.0. Instead of just managing customer information, companies should take the initiative in personalized system with customer oriented strategy. Furthermore, with the regular communication between CRM stakeholders (Sales-Marketing-IT), customer's demand should be directly reflected to enterprise strategy in real time.

  • PDF

라이프스타일 유형에 따른 레스토랑 실내이미지 선호도 차이에 관한 연구 (Preference Differences in Interior Images of Restaurants according to Lifestyles)

  • 김태희;박영석
    • 대한가정학회지
    • /
    • 제43권10호
    • /
    • pp.69-79
    • /
    • 2005
  • The purpose of this study was to determine restaurant patrons' preference differences in interior design style of restaurants according to their lifestyles. Written questionnaires were handed out to 500 adults in Seoul and surroundings and the results were sampled by convenience sampling. The questionnaire was composed of respondents' general characteristics, lifestyles, and preference for 10 types of interior design style. A total of 415 questionnaires were usable for data analysis, resulting in a response rate of $83\%$. To analyze the collected data, frequency, factor, reliability, quick clustering K- means and One-Way ANOVA analysis were conducted using SPSS 10.0. The results showed that there were preference differences in 10 types of interior design style of restaurants according to lifestyle types which were categorized into 4 groups. The conservative and self-convinced group showed the lowest preference scores in the 10 types of interior design style which are Romantic, Ethnic, Classic, High-Tech, Elegant, Country, Modem, Minimal, Natural, and Casual style. The quality life pursuing group and extroverted individuality groups showed the high preference scores in most of the styles, especially in the Classic and Elegant styles. The realistic self-centered group showed the highest preference scores in Casual style among the 4 groups. These study findings indicate that restaurants should take into account their patrons' lifestyles as a mean of market segmentation, and respond to their taste and preference when they have established suitable servicescape.

영상구도에 근거한 영상내의 주요객체 자동추출 기법 (Automatic Extraction of Major Object in the Image based on Image Composition)

  • 강선도;유헌우;신영근;장동식
    • 한국콘텐츠학회논문지
    • /
    • 제8권3호
    • /
    • pp.8-17
    • /
    • 2008
  • 본 논문에서는 새로운 주요객체 자동추출 알고리즘을 제안한다. 제안된 알고리즘은 크게 2단계 과정으로 요약될 수 있다. 1단계로 객체와 배경을 분리하는 영상분할 작업을 수행한다. 우선적으로 '인간은 3또는 4개의 주요 색상으로 축약하여 사물을 인식한다'는 연구 결과에 따라 K-means 알고리즘을 이용하여 3구역으로 분할하고, 분할된 영상 간 히스토그램 유사도를 계산하여 임계값 이상으로 유사하면 병합하는 과정을 수행한다. 2단계로 영상구도에 근거해 분할된 영상 중에 객체라고 지정하는 작업을 수행한다. 사람이 사진을 잘 찍기 위해서는 '주요객체의 위치를 영상구도에 맞추어 촬영하는 것이 바람직하다는 사실'에 근거하여 삼각구도를 바탕으로 가중치 마스크를 설계하여 객체위치를 역 추정하였다. 제안된 방법의 우수성을 보이기 위해 약 400개의 영상에 대해 실험하였으며, 최근에 발표된 KMCC, GBIS방법과도 비교하였다.

파노라마 방사선사진상에서 경상돌기의 변이에 관한 연구 (A STUDY ABOUT THE VARIATION OF STYLOID PROCESSES IN PANORAMIC RADIOGRAPHS)

  • 오숙희;김종열
    • 치과방사선
    • /
    • 제18권1호
    • /
    • pp.193-201
    • /
    • 1988
  • The purpose of this study was to investigate the variation in the length and shape of styloid processes and the relationships between the elongated styloid processes and the styloid process syndrome, using panoramic radiogrpahs and questionaires. The subjects were 342 patients consisted of 170 males and 172 females aged from 11 to 78 years, not showing facial asymmetry, who visited infirmary of dental college of Yonsei University. Also, the accuracy was determined for measuring the length of styloid processes from panoramic radiographs, using dried skulls. The results were as follows: 1. The length of styloid processes was magnified approximately 1.1 times, but there was no statistically significant differencies in the magnification rate of length between right and left side. 2. The mean radiographic length of styloid processes was 29.72±7.92㎜ in males, 27.93±6.69㎜ in females, and 28.82±7.37㎜ in total. And elongated styloid process (>30㎜) was seen in 31.3% of total subjects. 3. The growth in the length of styloid processes was completed in the third decade. 4. The most common shape of styloid processes was straight followed by segmented and bent form. The incidence of segmentation was reduced with increasing age. 5. The 74% of subjects with elongated styloid process (>30㎜) showed symptoms of the styloid process syndrome. The most frequent symptom was headache followed by discomfort in the neck when turning the head from left to right, tinnitus or earache, vague facial pain, discomfort or pain when swallowing, feeling that an object is caught in throat.

  • PDF

코로나바이러스 감염증19 데이터베이스에 기반을 둔 인공신경망 모델의 특성 평가 (Evaluation of Deep-Learning Feature Based COVID-19 Classifier in Various Neural Network)

  • 홍준용;정영진
    • 대한방사선기술학회지:방사선기술과학
    • /
    • 제43권5호
    • /
    • pp.397-404
    • /
    • 2020
  • Coronavirus disease(COVID-19) is highly infectious disease that directly affects the lungs. To observe the clinical findings from these lungs, the Chest Radiography(CXR) can be used in a fast manner. However, the diagnostic performance via CXR needs to be improved, since the identifying these findings are highly time-consuming and prone to human error. Therefore, Artificial Intelligence(AI) based tool may be useful to aid the diagnosis of COVID-19 via CXR. In this study, we explored various Deep learning(DL) approach to classify COVID-19, other viral pneumonia and normal. For the original dataset and lung-segmented dataset, the pre-trained AlexNet, SqueezeNet, ResNet18, DenseNet201 were transfer-trained and validated for 3 class - COVID-19, viral pneumonia, normal. In the results, AlexNet showed the highest mean accuracy of 99.15±2.69% and fastest training time of 1.61±0.56 min among 4 pre-trained neural networks. In this study, we demonstrated the performance of 4 pre-trained neural networks in COVID-19 diagnosis with CXR images. Further, we plotted the class activation map(CAM) of each network and demonstrated that the lung-segmentation pre-processing improve the performance of COVID-19 classifier with CXR images by excluding background features.

Lung Ventilation SPECT에서 LEAP Collimator의 유용성에 관한 연구 (The Study on Usefulness of LEAP Collimator in Lung Ventilation SPECT)

  • 김정수;김수미;김진의;이재성;이동수
    • 핵의학기술
    • /
    • 제16권2호
    • /
    • pp.18-24
    • /
    • 2012
  • Lung SPECT는 planar scan 보다 깊이 위치한 병변을 명확히 보여주며 보다 예민한 것으로 보고되고 있다. 그러나 ventilation SPECT는 정맥주사를 통해 방사성의약품을 주입하는 perfusion SPECT와 다르게 환자의 호흡을 통해 흡입시키므로, 원활한 호흡이 불가능한 환자의 경우 목적 장기에 방사성의약품의 섭취율이 저하되고 검사에 유효한 적정count/rate가 낮아지므로 필연적으로 검사시간이 길어지게 되어 이에 수반되는 여러 문제로 인하여 실제 임상에 적용하기 어려운 실정이다. 따라서 본 연구는 고감도(high sensitivity) 특성을 지닌 AP (All Purpose) 혹은 GP (General Purpose) collimator를 사용하여 low count/rate 환자의 검사시간을 단축시킴으로써 lung ventilation SPECT 임상적용 가능성과 유용성을 평가하였다. 정상 성인 남녀 4명($24.3{\pm}0.95$세)을 실험대상으로 Technegas 370 MBq을 흡입시키고, E.cam SPECT를 사용하여 count/rate를 측정한 후, LEUHR, LEHR, LEAP collimator를 장착하여 300counts acquisition 조건으로 ANT, POST, LPO, RPO, both-LAT 총 6 view의 lung ventilation planar scan 하였다. 동일 실험 대상자를 본원에서 사용하던 기존 프로토콜을 적용하여 70 kcounts/view acquisition 조건으로 high counting mode SPECT를 시행하고, 같은 방법으로 7kcounts/view acquisition하여 비교 실험인 low counting mode SPECT를 시행한 후 모든 실험의 결과를 lung segmentation ratio와 검사 소요시간을 통해 비교하여 분석하였다. 각 실험군 사이의 상관관계를 분석하기 위해 paired t-test를 시행하였고, p값이 0.05를 초과할 때 실험의 통계적 의미가 있는 것으로 판단하였다. 모든 자료의 통계처리는 SPSS ver. 20을 사용하였으며, 각 실험 결과는 산술 및 기하 평균(mean)을 구하고${\pm}$표준편차(SD)를 산출하였다. 실험 결과 lung planar scan은 300 kcounts/6 iew를 acquisition하여 총 1,800 kcounts를 획득하였다. 반면 7 kcounts/view로 acquisition한 low counting mode SPECT의 경우 collimator 종류에 따라서 최소 773 kcounts에서 최대 1,106 kcounts를 획득하였고, 70 kcounts/view로 acquisition한 high counting mode SPECT의 경우 collimator 종류에 따라서 최소 7,984 kcounts에서 최대 10,222kcounts를 획득하였다. 모든 실험 자료를 통해 실험대상 4명의 lung segmentation ratio를 산출하고, 평균값과 표준편차를 계산한 결과, LT:RT lung ratio는 순서대로 각각 45.5:54.5 (${\pm}0.26$), 45.4:54.6 (${\pm}0.40$), 50.1:49.9 (${\pm}0.28$), 49.4:50.6 (${\pm}0.57$)이었다. 따라서 정량분석한 결과 planar scan과 low, high counting mode SPECT사이 collimator 종류에 의한 lung segmentation ratio는 오차 범위내에 모두 존재하였다. 보다 자세한 분석을 위해 기존의 LEHR collimator를 장착한 planar scan과 비교 실험인 LEAP collimator를 장착한 low counts mode SPECT를 통계처리한 결과 p-value가 0.10이상으로 두 실험의 결과는 통계적으로 유의하였다. 실험에 소요된 시간을 분석한 결과 planar scan은 LEUHR, LEHR, LEAP collimator 순서로 각각 $1803{\pm}268$ sec, $915{\pm}162$ sec, $501{\pm}134$ sec이었고, 비교실험인 low counting mode SPECT는 $1125.5{\pm}103.6$ sec, $900{\pm}64$ sec, $740{\pm}0$ sec 이었으며, 본원에서 사용하던 기존 프로토콜인 high counting mode SPECT는 $5794.5{\pm}946.5$ sec, $3620{\pm}397.9$ sec, $2020{\pm}104.5$ sec이었다. 관심실험인 LEAP collimator를 장착한low counting mode SPECT의 검사 소요시간 740 sec에 비하여, 기존 본원의 프로토콜인 LEHR collimator를 장착한 high counting mode SPECT의 검사 소요시간 평균은 3620sec이므로, 약 5배 정도 시간 단축이 가능한 것으로 조사되었으며, 모든 실험의 정량분석 결과도 임상적으로 유의한 것으로 분석되었다. 결국 high sensitivity LEAP collimator를 장착하여 lung SPECT를 시행하면, 임상적으로 유의한 정량분석 결과를 얻을 수 있으며, 동시에 검사시간을 효과적으로 단축시킬 수 있음을 의미한다.

  • PDF

패션 앱을 이용한 모바일 쇼핑 태도 및 사용의도 영향요인 연구 -성별과 연령집단별 차이 비교- (A Study on the Determinants of Attitude toward and Intention to Use Mobile Shopping through Fashion Apps -Comparisons of Gender and Age Group Differences-)

  • 성희원
    • 한국의류학회지
    • /
    • 제37권7호
    • /
    • pp.1000-1014
    • /
    • 2013
  • This study identifies the determinants that influence attitude toward and the intention to use mobile shopping services through fashion applications (apps) based on the technology acceptance model. In addition, gender and age group differences were examined. Data were collected from subjects who have used smartphone fashion related apps; subsequently, a total of 327 data were analyzed. About 46% of respondents were males, with a mean age of 34.4 years that ranged from 20 to 49 years old. Multiple regression models were developed based on the research model. Perceived usefulness, perceived ease of use, perceived enjoyment, perceived risks (security risk and quality risk), fashion involvement, and fashion app attributes (product attributes and service attributes) were employed as predictors of attitudes towards mobile shopping. Attitudes towards mobile shopping and subjective norms with the aforementioned variables measured the intention to use. Attitudes towards mobile shopping were predicted by perceived enjoyment, perceived usefulness, and service attributes. Attitudes toward mobile shopping and subjective norms were the most important predictors of the intention to use. Gender differences were found in that service attributes were significant for attitudes towards mobile shopping only in the male model. Age differences were also found and perceived usefulness was the most important predictor of attitudes toward mobile shopping among those in their 20's; however, perceived enjoyment was the most important among those in their 30's and 40's. Quality risk was only significant to explain intention to use among those in their 40's. The findings of this study are useful to understand the possibility of the adoption of mobile shopping though fashion apps and provide basic insight into market segmentation.

Volumetric accuracy of cone-beam computed tomography

  • Park, Cheol-Woo;Kim, Jin-ho;Seo, Yu-Kyeong;Lee, Sae-Rom;Kang, Ju-Hee;Oh, Song-Hee;Kim, Gyu-Tae;Choi, Yong-Suk;Hwang, Eui-Hwan
    • Imaging Science in Dentistry
    • /
    • 제47권3호
    • /
    • pp.165-174
    • /
    • 2017
  • Purpose: This study was performed to investigate the influence of object shape and distance from the center of the image on the volumetric accuracy of cone-beam computed tomography (CBCT) scans, according to different parameters of tube voltage and current. Materials and Methods: Four geometric objects(cylinder, cube, pyramid, and hexagon) with predefined dimensions were fabricated. The objects consisted of Teflon-perfluoroalkoxy embedded in a hydrocolloid matrix (Dupli-Coe-Loid TM; GC America Inc., Alsip, IL, USA), encased in an acrylic resin cylinder assembly. An Alphard Vega Dental CT system (Asahi Roentgen Ind. Co., Ltd, Kyoto, Japan) was used to acquire CBCT images. OnDemand 3D (CyberMed Inc., Seoul, Korea) software was used for object segmentation and image analysis. The accuracy was expressed by the volume error (VE). The VE was calculated under 3 different exposure settings. The measured volumes of the objects were compared to the true volumes for statistical analysis. Results: The mean VE ranged from -4.47% to 2.35%. There was no significant relationship between an object's shape and the VE. A significant correlation was found between the distance of the object to the center of the image and the VE. Tube voltage affected the volume measurements and the VE, but tube current did not. Conclusion: The evaluated CBCT device provided satisfactory volume measurements. To assess volume measurements, it might be sufficient to use serial scans with a high resolution, but a low dose. This information may provide useful guidance for assessing volume measurements.

개 전신성염증반응증후군 및 패혈증의 진단적 표지자로서 CD11b의 활용 (CD11b as a Biomarker for Canine Systemic Inflammatory Response Syndrome and Sepsis)

  • 유도현;노동호;송루희;김준환;이다미;김수희;박철;박진호
    • 한국임상수의학회지
    • /
    • 제27권6호
    • /
    • pp.627-630
    • /
    • 2010
  • 본 연구는 중환자의 개에서 전신성염증반응증후군 및 패혈증에 의하여 유발되는 백혈구 활성화 지표를 분석하기 위하여 실행되었다. 분석된 지표로는 백혈구 수치, 혈소판 수치, 혈액 도말 표본의 백혈구 변화상 및 백혈구 표면의 CD11b발현 정도 등을 검사하였다. 그 결과, 패혈증군에서 백혈구의 CD11b 발현 값 (유세포 분석)이 현저하게 높게 관찰되었으며(P<0.05), 퇴행성 변화 정도도 같은 군에서 현저하게 높게 관찰 되었다(P<0.05). 또한 비 생존군의 CD11b 발현 값도 생존군에서의 값보다 현저하게 높았다. 이러한 연구 결과로 볼 때, 백혈구의 CD11b 발현은 중환자의 개에서 그 예후를 평가하는데 유용하다고 여겨진다.