• Smart Structures and Systems
• /
• 제31권4호
• /
• pp.325-334
• /
• 2023

Computer vision-based damage detection enables non-contact, efficient and low-cost bridge health monitoring, which reduces the need for labor-intensive manual inspection or that for a large number of on-site sensing instruments. By leveraging recent semantic segmentation approaches, we can detect regions of critical structural components and identify damages at pixel level on images. However, existing methods perform poorly when detecting small and thin damages (e.g., cracks); the problem is exacerbated by imbalanced samples. To this end, we incorporate domain knowledge to introduce a hierarchical semantic segmentation framework that imposes a hierarchical semantic relationship between component categories and damage types. For instance, certain types of concrete cracks are only present on bridge columns, and therefore the noncolumn region may be masked out when detecting such damages. In this way, the damage detection model focuses on extracting features from relevant structural components and avoid those from irrelevant regions. We also utilize multi-scale augmentation to preserve contextual information of each image, without losing the ability to handle small and/or thin damages. In addition, our framework employs an importance sampling, where images with rare components are sampled more often, to address sample imbalance. We evaluated our framework on a public synthetic dataset that consists of 2,000 railway bridges. Our framework achieves a 0.836 mean intersection over union (IoU) for structural component segmentation and a 0.483 mean IoU for damage segmentation. Our results have in total 5% and 18% improvements for the structural component segmentation and damage segmentation tasks, respectively, compared to the best-performing baseline model.

• 한국도서관정보학회지
• /
• 제26권
• /
• pp.235-259
• /
• 1997

Marketing can assist libraries in determining their future and identifying quality products - services, programs and materials. Marketing provides a theoretical framework addressing the specific library and information science questions facing public, school, special and academic libraries in both the public and private sectors. The purpose of this paper is to analyse information marketing. For this purpose, characteristics of information, concept of marketing are to be studied, and then the process of information marketing program is analysed by market segmentation, marketing audit, marketing mix(products, pricing, place, promotion) and marketing system development. As a result, traditionally having focused on the undifferentiated and concentrated marketing strategies in the market segmentation, now they have to develope differentiated strategies in order to compete with other information centers.

• 방송공학회논문지
• /
• 제28권3호
• /
• pp.267-274
• /
• 2023

의료영상 공개 데이터는 수집에 한계가 있어 데이터셋의 양이 부족하다는 문제점이 있다. 때문에 기존 연구들은 공개 데이터셋에 과적합 되었을 우려가 있다. 본 논문은 실험을 통해 8개의 (Unet, X-Net, HarDNet, SegNet, PSPNet, SwinUnet, 3D-ResU-Net, UNETR) 의료영상 분할 모델의 성능을 비교함으로써 기존 모델의 성능을 재검증하고자 한다. 뇌졸중 진단 공개 데이터 셋인 Anatomical Tracings of Lesions After Stroke(ATLAS) V1.2과 ATLAS V2.0에서 모델들의 성능 비교 실험을 진행한다. 실험결과 대부분 모델은 V1.2과 V2.0에서 성능이 비슷한 결과를 보였다. 하지만 X-net과 3D-ResU-Net는 V1.2 데이터셋에서 더 높은 성능을 기록했다. 이러한 결과는 해당 모델들이 V1.2에 과적합 되었을 것으로 해석할 수 있다.

• 한국병원경영학회지
• /
• 제10권4호
• /
• pp.75-97
• /
• 2005

The research objective is to empirically investigate the compensation packages maximizing the utilities of internal customers by applying the market segmentation theory. Data was collected from four Korean hospitals in Seoul, Busan and Gyunggi-do. The research is designed to seek the compensation package maximizing the utility of doctors by mixture regression model, which has been applied as latent structure and other type of finite mixture models from various academic fields since early 1980s. The mixture regression model shows the optimal segments number and fuzzy classification for each observation by EM(expectation-maximization algorism). The finite mixture regression model is to unmix the sample, to identify the groups, and to estimate the parameters of the density function underlying the observed data within each group. The doctors were segmented into 5 groups by their preference for the compensation package. The results of this study imply that the utility of doctors increases with differentiated compensation package segmented by their preference.

• 한국병원경영학회지
• /
• 제6권3호
• /
• pp.90-108
• /
• 2001

This research examines the compensation package maximizing the utilities of internal customers by applying the market segmentation theory. Data were collected from four Korean hospitals in Seoul, Pusan and Kyunggi-do. The research is designed to seek the compensation package maximizing the utility of doctors and nurses by applying the market segmentation theory. The compensation package for doctors and nurses was classified into 5 attributes which are level of salary, payment method, education, promotion, reward method. The test results were as follows. First, the relative importance of each attribute in the compensation package is different. The level of salary is the most important, reward method is the next. Second, the utility of doctors increases by 8.7%, when they are segmented on the basis. of their preference for compensation attributes while that of nurses increases by 39.8%. The results of this study imply that the utility of doctors and nurses increases with differentiated compensation package for internal customer segmented by their preference.

• 대한교통학회지
• /
• 제28권4호
• /
• pp.19-30
• /
• 2010

정보화의 진전에 따라 공공 서비스인 대중교통 정보서비스 제공의 올바른 방향 제시가 필요하게 되었다. 이에 본 연구는 상황에 따른 의사결정 특성이 강한 대중교통 정보서비스의 내용과 매체 이용 행태를 분석하였다. 이를 위해 본 연구는 주로 개인의 사회 인구학적 특성과 정보이용의 특정 행태 간 상관관계를 분석하는 전통적 접근 방법론을 보완하여, 상황에 따라 가변적인 의사결정 특성 분석이 가능한 접근 방법론을 채택하였다. 보다 구체적으로, 인지심리학적 의사결정 테이블(decision table)을 바탕으로 하는 분석도구인 CHAID의 귀납적 의사결정트리 추론(decision tree induction)을 이용하여 대중교통 이용자의 활동패턴과 정보이용 행태에 대해 분석하였다. 주요 연구결과로, 대중교통 정보서비스 이용은 사회 인구학적 변수 이외에도 정보 이용 당시의 상황 변수가 큰 영향을 미침을 확인하였다. 본 연구의 결과는 효율적 대중교통 정보서비스 제공을 위한 시장세분화(market segmentation)에 대한 중요한 시사점을 제공할 것으로 기대된다.

• 한국멀티미디어학회논문지
• /
• 제22권11호
• /
• pp.1269-1279
• /
• 2019

In medical image analysis, segmentation is considered as a vital process since it partitions an image into coherent parts and extracts interesting objects from the image. In this paper, we consider automatic segmentations of OCT retinal images to find six layer boundaries using convolutional neural networks. Segmenting retinal images by layer boundaries is very important in diagnosing and predicting progress of eye diseases including diabetic retinopathy, glaucoma, and AMD (age-related macular degeneration). We applied well-known CNN architecture for general image segmentation, called Segnet, U-net, and CNN-S into this problem. We also proposed a shortest path-based algorithm for finding the layer boundaries from the outputs of Segnet and U-net. We analysed their performance on public OCT image data set. The experimental results show that the Segnet combined with the proposed shortest path-based boundary finding algorithm outperforms other two networks.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권3호
• /
• pp.1104-1120
• /
• 2020

The chest X-rays are a common way to diagnose lung cancer or pneumonia. In particular, the finding of a lung nodule is the most important problem in the early detection of lung cancer. Recently, a lot of automatic diagnosis algorithms have been studied to find the lung nodules missed by doctors. The algorithms are typically based on segmentation network like U-Net. However, the occurrence of false positives that similar to lung nodules present outside the lungs can severely degrade performance. In this study, we propose a multi-task learning method that simultaneously learns the lung region and nodule-labeled data based on the prior knowledge that lung nodules exist only in the lung. The proposed method significantly reduces false positives outside the lung and improves the recognition rate of lung nodules to 83.8 F1 score compared to 66.6 F1 score of single task learning with U-net model. The experimental results on the JSRT public dataset demonstrate the effectiveness of the proposed method compared with other baseline methods.

• 방송공학회논문지
• /
• 제27권1호
• /
• pp.13-19
• /
• 2022

동영상 객체 분할(VOS) 기술은 연속된 레이블링 데이터를 필요로 하며, 현재 공개된 데이터셋으로 훈련된 VOS방법은 그 성능이 제한된다. 이 문제를 해결하기 위해 본 논문에서는 간단하면서도 효과적인 동영상 데이터 증강 기술들을 제안한다. 첫번째 증강 기술은 영상 내에서 객체를 제외한 배경을 다른 영상의 배경으로 대체하는 기법이고, 두번째 기술은 학습될 동영상 데이터의 순서를 무작위 확률로 뒤집어 역 재생되는 영상을 학습시키는 기법이다. 두 증강 기술은 객체 분할 시 배경 정보에 강인한 추정을 가능하게 하였고, 추가 데이터 없이 기존 모델의 성능을 향상시킬 수 있음을 보였다.

• 한국도서관정보학회지
• /
• 제44권1호
• /
• pp.279-303
• /
• 2013

본 연구의 목적은 공공도서관 서비스이용자를 충성도에 따라 세분화하고 각 세분시장을 목표시장으로 하는 공공도서관서비스 포지셔닝전략을 탐색하는데 있다. 연구 목적을 달성하기위해 공공도서관 이용자를 대상으로 인터뷰를 진행하였으며 인터뷰결과는 질적 분석 방법으로 분석하였다. 분석결과를 바탕으로 공공도서관서비스에 대한 충성도가 낮은 집단에 제시된 포지셔닝전략은 첫째, 이동도서관서비스의 확대실시 및 셔틀버스의 활용, 둘째, 정기적이고 체계적인 이용자조사, 셋째, 공공도서관 서비스 홍보방식의 다양화, 넷째, 이용자조사를 통해 이용자가 원하는 자료 구비 등 이었다. 공공도서관 서비스에 대한 충성도가 높은 집단에 대한 포지셔닝전략은 첫째, 내부마케팅, 둘째, 도서관 시설 및 설비 품질의 일정 수준 유지, 셋째, 취학 전 아동 및 초, 중, 고등학생과 학부모를 위한 독서프로그램 기획 및 제공이었다.