• Journal of Trauma and Injury
• /
• v.20 no.2
• /
• pp.83-89
• /
• 2007

Purpose: This research was performed to determine which clinical signs and symptoms of brain injury are sensitive indicators of skull fracture (SF) and intracranial injury (ICI) in head injured children. Methods: We conducted a prospective study of minor head trauma in children younger than 2 years of age for a 1-year period. Skull radiographs, brain computed tomography (CT), and data forms, including mechanism of injury, symptoms, physical findings, and hospital course, were completed for each child. Results: Of 137 study subjects, 17 (12.4%) had SF/ICI. Falls were the most common mechanism of injury, and heights of fall above 1 meter were associated with incidence of SF/ICI (p<0.05). Scalp abnormalities were not associated with incidence of SF/ICI. As for clinical symptoms, lethargy and a grouping of features (irritability & vomiting) were associated with incidence of SF/ICI (p<0.05). The incidence of seizure, loss of consciousness, vomiting, irritability, and scalp abnormality did not differ significantly between those with normal radiologic findings and those with SF/ICI. Among asymptomatic patients, 11 (14.5%) patients had SF/ICI, and among patients with normal scalp findings, 9 (12.7%) patients had SF/ICI. Conclusion: Clinical signs and symptoms, except for lethargy and a grouping of features (irritability & vomiting), were not sensitive predictors of SF/ICI. Nevertheless, SF/ICI occurred among normal children. In such a case, a liberal policy of CT scanning is warranted.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.5 no.2
• /
• pp.58-69
• /
• 1995

In this paper GFI (Generalized Fuzzy Isodata) and FI (Fuzzy Isodata) algorithms are studied and applied to the tire tread pattern classification problem. GFI algorithm which repeatedly grouping the partitioned cluster depending on the fuzzy partition matrix is general form of GI algorithm. In the constructing the binary tree using GFI algorithm cluster validity, namely, whether partitioned cluster is feasible or not is checked and construction of the binary tree is obtained by FDH clustering algorithm. These algorithms show the good performance in selecting the prototypes of each patterns and classifying patterns. Directions of edge in the preprocessed image of tire tread pattern are selected as features of pattern. These features are thought to have useful information which well represents the characteristics of patterns.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2001.06a
• /
• pp.128-132
• /
• 2001

본 논문에서는 객체가 포함되어 있는 칼라 영상들의 집합으로부터 유사한 시각적 특징을 가진 캑체를 공통적으로 포함하는 영상들을 추출하여, 전체 영상들을 소수의 그룹으로 묶을 수 있는 방법을 제안한다. 먼저 영상에 포함된 객체는 주로 영상의 내부 위치에 나타난다는 사실에 착안하여, 영상의 경계와 접하지 않는 비경계 영역들을 병합하여 객체가 포함된 영역으로 간주한다. 그러나, 병합된 내부 영역에는 객체가 아닌 영역들도 포함하고 있을 수 있으므로, 칼라 히스토그램만을 이용하여 내부 영역에 대한 특징으로 이용한다. 이러한 내부 영역의 칼라 특징들이 분포하는 형상을 분석하여 밀집도가 높은 클러스터를 그룹으로 추출한다. 이때, 밀집도는 일반적으로 사용하는 공간적인 분포 대신에 히스토그램 인터섹션에 의한 유사도를 이용하여 정의한다. 즉, 서로 유사도가 놓은 것들이 집중되어 분포되어 있는 경우에 밀집도가 높은 클러스터로 간주하여 추출하는 방법을 사용한다. 클러스터의 형상 및 개수를 자동적으로 결정할 수 있는 방법도 제안한다. 실험에 의해, 추출한 클러스터의 칼라 영상들이 동일한 객체를 포함하고 있음을 알 수 있었으나, 향후 보다 안정화된 방법의 개발이 필요하다. 아울러, 클러스터별로 객체의 의미를 부여할 수 있는 방법론의 개발도 필요함을 알 수 있었다.

• Journal of Multimedia Information System
• /
• v.6 no.4
• /
• pp.251-258
• /
• 2019

In the image methodology, cinemagraphs are needed to find out how to think through the empirical studies of the audiences in actual empirical theory. Therefore, this study analyzes what cinemagraph images represent and what groups are inclined to see them. The analytical study aims to reveal the grouping categorized as a preference survey according to the subjectivity study by Q methodology. Research reveals how images should be read and how disparities tend to differ from or agree with those in understanding images. Therefore, four types categorized into Q analysis through 1: 1 interview using a sample group and using samples showed that human interpretation and cinemagraph images were represented in understanding cinemagraphs.

• 한국데이터정보과학회:학술대회논문집
• /
• 2005.10a
• /
• pp.59-69
• /
• 2005

Data mining technique is used to find hidden knowledge by massive data, unexpectedly pattern, relation to new rule. The methods of data mining are decision tree, association rules, clustering, neural network and so on. Clustering is the process of grouping the data into clusters so that objects within a cluster have high similarity in comparison to one another. It has been widely used in many applications, such that pattern analysis or recognition, data analysis, image processing, market research on off-line or on-line and so on. We analyze Gyeongnam social indicator survey data by 2001 using twostep clustering technique for environment information. The twostep clustering is classified as a partitional clustering method. We can apply these twostep clustering outputs to environmental preservation and improvement.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.5
• /
• pp.931-936
• /
• 2009

Classification problem plays a major role in grouping of the records in database systems, detection of faults in VLSI circuits, image processing, and so on. In this paper, we propose the algorithm constructing D1-MACA as a two-class classifier with minimum memory for given pattern sets using the concepts of subspace. Also we analyze the condition that is designed a two-class classifier D1-MACA with two attractors.

• Proceedings of the IEEK Conference
• /
• 1999.11a
• /
• pp.360-363
• /
• 1999

Video data compression is a major key technology in the field of multimedia applications. Variable-length coding is the most popular data compression technique which has been used in many data compression standards, such as JPEG, MPEG and image data compression standards, etc. In this paper, we present memory efficient VLC decoder architecture for MPEG-2 application which can achieve small memory space and higher throughput. To reduce the memory size, we propose a new grouping, remainder generation method and merged lookup table (LUT) for variable length decoders (VLD's). In the MPEG-2, the discrete cosine transform (DCT) coefficient table zero and one are mapped onto one memory whose space requirement has been minimized by using efficient memory mapping strategy The proposed memory size is only 256 words in spite of mapping two DCT coefficient tables.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.1
• /
• pp.1-11
• /
• 2006

Data mining technique is used to find hidden knowledge by massive data, unexpectedly pattern, relation to new rule. The methods of data mining are decision tree, association rules, clustering, neural network and so on. Clustering is the process of grouping the data into clusters so that objects within a cluster have high similarity in comparison to one another. It has been widely used in many applications, such that pattern analysis or recognition, data analysis, image processing, market research on off-line or on-line and so on. We analyze Gyeongnam social indicator survey data by 2001 using twostep clustering technique for environment information. The twostep clustering is classified as a partitional clustering method. We can apply these twostep clustering outputs to environmental preservation and improvement.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.467-469
• /
• 2003

This research aimed to develop forest type classification technique for the mixed forest with coniferous and broad-leaved species using the high resolution satellite data. QuickBird data was used as satellite data. The method of this research was to extract satellite data for every single tree crown using image segmentation technique, then to evaluate the accuracy of classification by changing grouping criteria such as tree species, families, coniferous or broad-leaved species, and timber prices. As a result, the classification of tree species and families level was inaccurate, on the other hand, coniferous or broad-leaved species and timber price level was high accurate.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.6 no.1
• /
• pp.8-15
• /
• 2011

This paper presents an effective method for detecting vehicles in front of the camera-assisted car during nighttime driving. The proposed method detects vehicles based on detecting vehicle headlights and taillights using techniques of image segmentation and clustering. First, in order to effectively extract spotlight of interest, a pre-signal-processing process based on camera lens filter and labeling method is applied on road-scene images. Second, to spatial clustering vehicle of detecting lamps, a grouping process use light tracking method and locating vehicle lighting patterns. For simulation, we are implemented through Da-vinci 7437 DSP board with visible light mono-camera and tested it in urban and rural roads. Through the test, classification performances are above 89% of precision rate and 94% of recall rate evaluated on real-time environment.