• Journal of KIISE
• /
• v.42 no.2
• /
• pp.203-212
• /
• 2015

This paper proposes a traffic sign recognition method in real road environments. The video stream in driving environments has two different characteristics compared to a general object video stream. First, the number of traffic sign types is limited and their shapes are mostly simple. Second, the camera cannot take clear pictures in the road scenes since there are many illumination changes and weather conditions are continuously changing. In this paper, we improve a modified census transform(MCT) to extract features effectively from the road scenes that have many illumination changes. The extracted features are collected by histograms and are transformed by the dense descriptors into very high dimensional vectors. Then, the high dimensional descriptors are encoded into a low dimensional feature vector by Fisher-vector coding and Gaussian Mixture Model. The proposed method shows illumination invariant detection and recognition, and the performance is sufficient to detect and recognize traffic signs in real-time with high accuracy.

• Journal of Service Research and Studies
• /
• v.4 no.2
• /
• pp.37-47
• /
• 2014

Service industry is a fundamental field that takes sustainable development of state economy. However the development of service industry in our economy doesn't yet compare to level of developed countries. It is supported by the recent statistics of our service industry. This paper shows regional comparative analysis about the current status of the service industry based on the number of establishments and workers. Using the data of census on establishments in 16 cities and provinces, we analyze the relative ratio for the number of establishments and workers in service industry among total industry in Seoul metropolitan, Gangwon, Chungcheong, Honam, Youngnam and Jeju region. Also we analyze the relative ratio by provinces in each region. We used the data of census on establishment for 7 years from 2006 to 2012, we calculated the relative ratio of the number of establishments and workers in service industry for the 6 regions. And we analyzed the relative ratio by provinces in each region.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.2
• /
• pp.33-44
• /
• 2010

This paper propose the structure of real-time face detection hardware architecture for robot vision processing applications. The proposed architecture is robust against illumination changes and operates at no less than 60 frames per second. It uses Modified Census Transform to obtain face characteristics robust against illumination changes. And the AdaBoost algorithm is adopted to learn and generate the characteristics of the face data, and finally detected the face using this data. This paper describes the face detection hardware structure composed of Memory Interface, Image Scaler, MCT Generator, Candidate Detector, Confidence Comparator, Position Resizer, Data Grouper, and Detected Result Display, and verification Result of Hardware Implementation with using Virtex5 LX330 FPGA of Xilinx. Verification result with using the images from a camera showed that maximum 32 faces per one frame can be detected at the speed of maximum 149 frame per second.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.11
• /
• pp.3302-3307
• /
• 2009

The objective of the present study lies in providing the method to construct the precision population-statistical map for statistical demographics making full advantage of GIS and the national census data in an attempt to improve accuracy and reliability of population estimation applicable for a variety of location analysis. More specifically, it adopts the multiple regression analysis by segmented land use type(biotope) taking into account that the land use diversified as residence, commercial and office areas has the close connectivity and interdependence with population. Based on the analyzed result above, the study finalizes the modeling to construct demographic map with higher precision by prioritizing the population density by weight value and then re-distributing the population according to jurisdictional dong's and types of use for the land. The study consequently is expected to be conducive to improving precision and reliability rather than the existing method for population estimation widely acceptable thus far.

• Journal of IKEEE
• /
• v.26 no.4
• /
• pp.686-692
• /
• 2022

In this paper, the matching error rates of modified area-based, energy-based algorithms, and learning-based structures were compared for stereo image matching. Census transform (CT) based on region and life propagation (BP) algorithm based on energy were selected, respectively.Existing algorithms have been improved and implemented in an embedded processor environment so that they can be used for stereo image matching in mobile systems. Even in the case of the learning base to be compared, a neural network structure that utilizes small-scale parameters was adopted. To compare the error rates of the three matching methods, Middlebury's Tsukuba was selected as a test image and subdivided into non-occlusion, discontinuous, and disparity error rates for accurate comparison. As a result of the experiment, the error rate of modified CT matching improved by about 11% when compared with the existing algorithm. BP matching was about 87% better than conventional CT in the error rate. Compared to the learning base using neural networks, BP matching was about 31% superior.

• Journal of the Korean Geographical Society
• /
• v.43 no.2
• /
• pp.235-252
• /
• 2008

Many cartographers have for many years searched for a way to construct cartograms in which the sizes of geographic areas such as states, counties or census tracts are reseated in proportion to their population or some other socio-economic properties. While many techniques and algorithms for creating cartograms have been proposed, some of them are still extremely complex to generate in a proper manner, and many of them suffer either from this lack of readability or from seamless integration with GIS software. This paper, therefore, presents a simple population cartogam technique based on the Circular Cartogram Algorithm(CCA) by Dorling(1996) to tackle these drawbacks by drawing the areas as simple circles for use as a base map and linking the construction with GIS mapping processes. For an automated approach in the cartogram generation, this paper proposes a close coupling method of ArcView GIS 3.3. package in order for users to access to the cartopam algorithm. Then, they will be available through an interface that the ArcView GIS system allows user-written routines to be accessed easily. The CCA and its coupling architecture ensure to improve the potential applicability of the use of cartograms to census mapping at practical levels. As the cartogram examples, cartograms of population and property types in 2005 Korea census data sets are illustrated in the end, by which viewers can easily identify the residential concentration and their relative ratio in Seoul metropolitan area.

• Korean Journal of Community Nutrition
• /
• v.21 no.5
• /
• pp.463-476
• /
• 2016

Objectives: Environmental, social and personal factors influence eating patterns. This study aimed to investigate the relationship between unhealthy food outlets within a residential area and obesity using nationally representative Korean survey data and data from the Census on Establishments. Methods: Data on the food intakes and socioeconomic variables of a total of 9,978 adults aged ${\geq}19$ years were obtained from the 2013-2014 Korea National Health and Nutrition Examination Survey. Geographic locations of restaurants were obtained from the 2013 Census on Establishments in Korea. Administrative area was categorized into tertiles of count of unhealthy food outlets based on the distribution of number of unhealthy food outlets among all urban (Dong) and rural (Eup or Myun) administrative districts in Korea. Multilevel logistic regressions model were used to assess the association between the number of unhealthy food outlets and obesity. Results: People living in the district with the highest count of unhealthy food outlets had higher intakes of fat (45.8 vs. 44.4 g/day), sodium (4,142.6 vs. 3,949.8 mg/day), and vitamin A (753.7 vs. $631.6 {\mu}gRE/day$) compared to those living in the district with the lowest count of unhealthy food outlets. A higher count of unhealthy food outlets was positively associated with frequent consumption of instant noodles, pizza, hamburgers and sandwiches, sweets and sour pork or pork cutlets, fried chicken, snacks, and cookies. Higher exposure to unhealthy food outlets was associated with increased odds of obesity (1st vs. 3rd tertile; OR 1.689; 95% CI 1.098-2.599). Conclusions: A high count of unhealthy food outlets within a residential area is positively associated with the prevalence of obesity in Korea. The results suggest that food environmental factors affects the health outcomes and interventions aiming to restrict the availability of unhealthy food outlets in local neighborhoods may be a useful obesity prevention strategy.

• Korean Journal of Environmental Biology
• /
• v.37 no.1
• /
• pp.82-87
• /
• 2019

Global warming has a significant impact on diverse ecosystems including agroecosystem through; changing of phenology, physiology and distribution. Monitoring of biological responses emanating from global warming is required to understand the challenges of biological diversity conservation posed by climate change. The Korean government selected four butterfly species as indicators of climate change in agroecosystem: Papilio xuthus, Pieris rapae, Colias erate, and Eurema mandarina. The aim of this study was to investigate the different monitoring methods of the butterflies in Korea and suggest a suitable monitoring method to track the population trends of butterflies in the agroecosystem. Butterfly monitoring was conducted in eight sites throughout Korea from April to October, 2018 using three survey methods: point census at rice paddy area, point census at the border between rice paddy and hill and line transect along the rice paddy and hill. Each method took approximately 30 min. to count the butterflies. A total of 4,691 butterflies and 92 species were counted: The most dominant species was Pieris rapae with a total count of 1,205 individuals followed by Polygonia c-aureum, Zizeeria maha, Colias erate, Cupido argiades and Papilio xuthus. Among the three census methods, the total number of species and individuals when using line transect method was statistically higher than in the other methods. However, the numbers of the four butterflies indicators showed no difference throughout three census methods. Based on the number of species and the total individuals butterflies in agroecosystem, we advocate for the application of line transect method as it can find more butterflies in agroecosystem. In addition, we advised for the implementation of education programs on the line transect method in butterfly identification to participants of the national monitoring program.

• The Korean Nurse
• /
• v.23 no.5 s.128
• /
• pp.78-89
• /
• 1984

This study is designed to investigate the nursing educational system and nursing curriculum. Data for analysis were based on the 1983 educational census report of Korean Nurses Association, curriculum of 44 Schools of Nursing sampled for comparison and li

• Proceedings of the Korea Water Resources Association Conference
• /
• 2005.09a
• /
• pp.68-69
• /
• 2005