• The KIPS Transactions:PartB
• /
• v.10B no.4
• /
• pp.443-450
• /
• 2003

In this paper, we proposed the novel hierarchical algorithm for the recognition of English calling cards that processes multiresolution images of calling cards hierarchically to extract individual characters and recognizes the extracted characters by using the enhanced neural network method. The hierarchical recognition algorithm generates multiresolution images of calling cards, and each processing step in the algorithm selects and processes the image with suitable resolution for lower processing overhead and improved output. That is, first, the image of 1/3 times resolution, to which the horizontal smearing method is applied, is used to extract the areas including only characters from the calling card image, and next, by applying the vertical smearing and the contour tracking masking, the image of a half time resolution is used to extract individual characters from the character string areas. Lastly, the original image is used in the recognition step, because the image includes the morphological information of characters accurately. And for the recognition of characters with diverse font types and various sizes, the enhanced RBF network that improves the middle layer based on the ART1 was proposed and applied. The results of experiments on a large number of calling card images showed that the proposed algorithm is greatly improved in the performance of character extraction and recognition compared with the traditional recognition algorithms.

• Journal of the Korean Regional Science Association
• /
• v.35 no.2
• /
• pp.19-31
• /
• 2019

There have been a growing number of studies that show huge discrepancies between individuals' perceptions of relative economic position and the economic position based on the reported income in the survey. If this is the case, it is expected that the impact of perceived income status on individual happiness can be different from that of objective income status based on the reported income. The aim of this study is to investigate the factors affecting the discrepancies between perception and reality with respect to relative income status, focusing on the 'Reference group theory'. This study also tries to extend existing knowledge of the relative status on the happiness level of individuals, by examining how individual's happiness can be changed by providing the accurate information about their objective income level. There are systematic biases in perceived income status. A majority of people who actually rank in the lower part of income ladder place themselves in higher positions, while a significant portion of rich individuals underestimate their actual income status. Secondly, the misperception about the income distribution is affected by a variety of individual, household characteristics and reference group income. Thirdly, providing individuals with accurate information has a considerable effect on their happiness level.

• Annual Conference on Human and Language Technology
• /
• 1990.11a
• /
• pp.183-190
• /
• 1990

본 논문에서는 인쇄체 한글문자를 실용적으로 인식하기 위하여 고안된 계층적 신경망을 소개하고, 이를 다중활자체의 한글문자를 인식하는 문제에 적용하였다. 이 신경망은 입력된 문자영상을 6가지의 유형으로 분류한 후, 해당 유형을 처리하는 신경망에서 실제 문자를 인식하도록 구성되었다. 또한 각 신경망을 모든 입력영상의 모든 출력노드에 대해 고르게 학습시키기 위하여 Backpropagation 알고리즘을 개선한 Descending Epsilon 알고리즘을 도입하였다. 그 결과 사용빈도수가 높은 한글 520자에 대해 94.4 - 98.4%의 인식률을 얻음으로써 본 논문에서 제안한 시스템이 다양한 활자체로 이루어진 실제 문서인식시스템의 문자인식부에 효과적으로 사용될 수 있음을 제시하였다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.9
• /
• pp.482-491
• /
• 2018

This study intended to identify different level of teacher-efficacy, perception of principals' leadership and peer collaboration as it pertains to a teachers' job stress and job satisfaction in Elementary school. Samples include 1,031 teachers in elementary school from Korean Children & Youth Panel Survey(KCYPS) and data were analyzed using Latent Class Analysis(LCA) to identify different patterns of teacher-efficacy and perception of principals' leadership and peer collaboration. Multivariate analysis of variance were employed to identify the influence of predictors for classification of teachers' job stress and job satisfaction among latent classes. The study found three latent classes at risk class, middle-level adaptive class, and adaptive class and results showed that each distinctive class can be identified by some of predictors. Teachers at adaptive class showed higher teacher-efficacy and positive perception of principals' leadership and peer collaboration than teachers at risk and middle-level adaptive class. Also, teachers at adaptive class showed lower job stress and higher job satisfaction than teachers at two other classes. The study suggests that help teachers based on personal profile are effective rather teacher-efficacy and perception of principals' leadership and peer collaboration.

• Convergence Security Journal
• /
• v.18 no.3
• /
• pp.53-59
• /
• 2018

In this paper, we propose a hierarchical image classification scheme for efficient object image classification. In the non-hierarchical image classification, which classifies the existing whole images at one time, it showed that objects with relatively similar shapes are not recognized efficiently. Therefore, in this paper, we introduce the image classification method in the hierarchical structure which attempts to classify object images hierarchically. Also, we introduce to the efficient class structure and algorithms considering the scalability that can occur when a deep learning image classification is applied to an actual system. Such a scheme makes it possible to classify images with a higher degree of confidence in object images having relatively similar shapes.

• Journal of the Korean Regional Science Association
• /
• v.36 no.3
• /
• pp.3-16
• /
• 2020

The causes and effects of intra-urban spatial inequality have received much scholarly attention. However, the effects of urban spatial inequality on resident perceptions and the mechanisms through which it is sustained and reproduced remain mostly unknown. This study aimed to investigate whether regional deprivation, the relative socioeconomic standing of a region, affects the residents' perceptions of upward social mobility. By employing the ordinal logistic multilevel model to analyze nested data collected from Seoul, South Korea, this study found that the regional deprivation has a significant negative effect on residents' perception of upward social mobility. The results of this study suggest that one way in which spatial inequality is sustained and reproduced is by the effects of regional deprivation, having negative impacts on the aspirations and socioeconomic activities of residents. This study is expected to provide meaningful implications for planning and policy aimed to combat spatial inequality.

• Journal of the Korea Computer Industry Society
• /
• v.2 no.8
• /
• pp.1045-1054
• /
• 2001

To improve classification accuracy in this paper, we proposed an algorithm for the chromosome image reconstruction in the image preprocessing part and also proposed the pattern classification method using the hierarchical multilayer neural network(HMNN) to classify the chromosome karyotype. It reconstructed chromosome images for twenty normal human chromosome by the image reconstruction algorithm. The four morphological and ten density feature parameters were extracted from the 920 reconstructed chromosome images. The each combined feature parameters of ten human chromosome images were used to learn HMNN and the rest of them were used to classify the chromosome images. The experimental results in this paper were composed to optimized HMNN and also obtained about 98.26％ to recognition ratio.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.4
• /
• pp.346-355
• /
• 2020

The purpose of this study was to suggest measures to improve safety and health training by analyzing trainees who participated in the training programs. A survey was administered to 226 trainees who participated in and completed the safety and health training programs provided by the Korea Occupational Safety and Health Agency. The study results showed that the level of satisfaction of employers for the training programs was the lowest compared to the middle managers and employees. Moreover, the middle managers' job competencies were increased, and they applied newly acquired knowledge and skills to their jobs. However, there were significant differences between industries. Finally, although all levels of trainee agreed on the effectiveness of safety and health training, the perceptions of employers were lower than those of employees. Based on these study results, several implications to improve the safety and health training were suggested.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04c
• /
• pp.325-327
• /
• 2003

본 논문에서는 인쇄체 한글을 실용적으로 인식할 수 있는 계층적으로 구성한 신경망을 제안하고, 이를 이용해서 주민등록증의 성명을 인식하는데 적용하였다. 문자영상을 신경망을 이용하여 한글의 6가지 유형으로 먼저 분류한 후, 분류된 문자영상을 각 형식에 따라 자소단위로 분할해서 각 형식에 따른 신경망으로 인식하는 구조로 만들었다. 훈련용 데이터는 각 형식 별로 자소를 분리해서 얻은 영상들을 자소별 평균이미지로 만들어서 이를 조합하여 만든 글자로 사용하였다. 그래서 같은 형식의 같은 자음이라도 글자의 모양과 위치가 조금 다른것에 대해서 강인한 훈련을 할 수 있었다. 또한 입력단에서의 잡음을 줄이기 위해 히스토그램의 국부 평균을 적용하였다. 100명의 주민등록증을 컴퓨터 카메라를 이용하여 입력받아서 테스트한 결과 98.1%의 높은 인식률을 얻을 수 있었다.

• 한국HCI학회:학술대회논문집
• /
• 2007.02a
• /
• pp.750-754
• /
• 2007

본 논문은 실내의 로비나 복도에 설치된 방범 카메라로부터 얻어진 일련의 영상으로부터 '걷기', '뛰기', '앉기', '일어서기', '넘어짐'의 비교적 짧은 시간에 일어나는 인간 행동들을 실시간으로 인식하는 시스템의 구현에 관해 다룬다. 먼저 입력으로 받은 영상을 계층적 색인 구조를 갖는 다중 가우시안 기반의 배경 모델을 이용하여 윤곽을 추출하고 객체를 인식하여 시간차에 의한 가중치로 누적하여 시간 템플릿을 만든다. 만들어진 시간 템플릿으로부터 특징을 추출하여 신경망 모델에 적용하여 5가지 인간행동을 구분한다. 구현된 시스템으로 인간행동 인식 실험을 수행하였는데, 실험 참가자들의 행동 방식이 약간씩 달랐음에도 불구하고 높은 인식률을 보여주었다.