• 한국정보과학회논문지:데이타베이스
• /
• 제28권3호
• /
• pp.301-314
• /
• 2001

데이터마이닝은 대용량 데이터베이스에 내재된 유용한 지식을 탐사하는 기술로 정의된다. 데이터마이닝에 대한 연구가 진행되면서 순차 패턴, 유사 시계열 탐사, 시간 연관규칙 탐사 등과 같이 시간 값을 가진 데이터로부터 지식을 탐사하고자 하는 시간 데이터마이닝에 대한 연구가 수행되었다. 그러나 기존 연구는 트랜잭션의 발생 시점만을 가진 데이터를 다루고 있으며 시간 간격을 가진 데이터는 거의 고려하고 있지 않다. 실세계에서는 환자의 병력, 상품 구매 이력, 웹 로그 등과 같은 시간간격을 가진 다양한 데이터가 존재하며 이로부터 여러 유용한 지식을 찾아낼 수 있다. Allen은 시간간격 데이터 사이에 발생할 수 있는 시간 관계와 시간 관계를 구할 수 있는 시간간격 연산자를 정의하였다. 본 논문에서는 Allen의 정의를 기반으로 시간간격 데이터로부터 시간관계 규칙을 효율적으로 탐사하기 위한 새로운 데이터마이닝 기법을 제안하였다. 이 기법은 발생 시점을 가진 시간 데이터를 시간간격 데이터로 요약하여 일반화하는 전처리 알고리즘과 시간간격 데이터로부터 시간관계 규칙을 생성하는 규clr 탐사 알고리즘으로 구성된다. 이 기법은 기존 데이터마이닝 기법에서 찾지 못하는 유용한 시간 규칙을 탐사할 수 있다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권1호
• /
• pp.169-189
• /
• 2015

Detection of anomalous events from video streams is a challenging problem in many video surveillance applications. One such application that has received significant attention from the computer vision community is traffic video surveillance. In this paper, a Lossy Count based Sequential Temporal Pattern mining approach (LC-STP) is proposed for detecting spatio-temporal abnormal events (such as a traffic violation at junction) from sequences of video streams. The proposed approach relies mainly on spatial abstractions of each object, mining frequent temporal patterns in a sequence of video frames to form a regular temporal pattern. In order to detect each object in every frame, the input video is first pre-processed by applying Gaussian Mixture Models. After the detection of foreground objects, the tracking is carried out using block motion estimation by the three-step search method. The primitive events of the object are represented by assigning spatial and temporal symbols corresponding to their location and time information. These primitive events are analyzed to form a temporal pattern in a sequence of video frames, representing temporal relation between various object's primitive events. This is repeated for each window of sequences, and the support for temporal sequence is obtained based on LC-STP to discover regular patterns of normal events. Events deviating from these patterns are identified as anomalies. Unlike the traditional frequent item set mining methods, the proposed method generates maximal frequent patterns without candidate generation. Furthermore, experimental results show that the proposed method performs well and can detect video anomalies in real traffic video data.

• 정보처리학회논문지D
• /
• 제18D권4호
• /
• pp.215-224
• /
• 2011

U-Health는 다양한 종류의 센서로 환자 정보를 수집하며, 스트림 데이터는 시작 시점과 종료 시점을 갖는 인터벌 이벤트로 요약 가능하다. 그러나 대부분의 시간 데이터 마이닝 기법들은 이벤트 발생 시점만을 고려하며 스트림 데이터의 상태 변화는 간과하는 문제가 있다. 이 논문은 U-Health에서 이벤트 상태 변화를 고려한 시간 마이닝 기법을 제안한다. 제안 방법은 U-Health에서 관심이 있는 이벤트만을 센서에서 서버로 전송함으로써 환경의 제약 사항들을 극복하고 스트림 데이터에 대한 네 가지 이벤트 상태를 정의하여 상태 변화를 고려한 시간 마이닝을 수행한다. 최종적으로, 제안 방법은 이벤트들 사이에 존재하는 인과 관계를 시간 관계 시퀀스로 기술하여 탐사 규칙의 모호함을 제거한다.

• 전자공학회논문지B
• /
• 제30B권7호
• /
• pp.70-82
• /
• 1993

In this paper, employing the 3-dimensional discrete cosine transform (DCT) for the utilization of the temporal correlation, an adaptive motion sequence coding technique is proposed. The energy distribution in a 3-D DCT block, due to the nonstationary nature of the image data, varies along the veritical, horizontal and temporal directions. Thus, aiming an adaptive system to local variations, adaptive procedures, such as the 3-D classification, the classified linear scanning technique and the VLC table selection scheme, have been implemented in our approach. Also, a hybrid structure which adaptively combines inter-frame coding is presented, and it is found that the adaptive hybrid frame coding technique shows a significant performance gain for a moving sequence which contains a relatively small moving area. Through an intensive computer simulation, it is demonstrated that, the performance of the proposed 3-D transform coding technique shows a close relation with the temporal variation of the sequence to be code. And the proposed technique has the advantages of skipping the computationally complex motion compensation procedure and improving the performance over the 2-D motion compensated transform coding technique for rates in the range of 0.5 ~ 1.0 bpp.

• 한국언어정보학회지:언어와정보
• /
• 제4권2호
• /
• pp.93-105
• /
• 2000

I argue in this paper that the sequence of tense (SOT) phenomenon can be accounted for without positing a SOT rule, focusing on the contrast between the past under-past sentences which lead to ambiguity and those sentences which do not. The different interpreta- tion of past under past sentences depends on whether the stative or then non-stative predicates occur in the complement clauses in the propositional attitude verbs. Based on this, I also argue that the embedded past tense does not contribute to the seman- tics past tense in the complement clause. Instead, it is due to the occurrence of the stative or non-stative predicates in the complement clauses. The stative predicates are associated with the temporal precedence or the overlap relation, whereas the non-stative predicates the precedence relation only. This fact triggers the contrast in past- under- past sentences.(Korea University)

• 전자공학회논문지
• /
• 제51권8호
• /
• pp.51-57
• /
• 2014

본 논문은 비디오에서 안개 제거로 인한 플리커링(깜빡임) 현상을 효율적으로 제거하기 위한 방법을 제시한다. 단일 영상에서 사용하는 안개 제거 방법을 비디오 시퀀스에 그대로 적용하면, 안개제거를 위한 대기값이 매 프레임 마다 개별적으로 계산되기 때문에 플리커링 현상이 발생할 수 있다. 기존의 플리커링 현상을 제거하는 방법으로는 인접한 프레임간의 전달맵의 높은 상관관계 계수를 이용하는 방법이 존재한다. 하지만, 비디오 시퀀스에서 빈번하게 변경되는 대기값에 의해서도 플리커링 현상이 발생할 수 있기 때문에 인접한 프레임간의 상관관계를 효율적으로 고려한 대기값 추정방법이 필요하다. 본 논문에서는 안정적인 대기값 추정을 위해 평균값을 이용해 플리커링 현상을 제거하는 효율적인 안개 제거 기법을 제안한다. 실험결과 제안하는 방법이 기존의 방법보다 플리커링 현상을 줄여 좋은 성능결과를 보여 주는 것을 확인하였다.

• 한국인터넷방송통신학회논문지
• /
• 제12권4호
• /
• pp.201-209
• /
• 2012

시점기반 복합 이벤트 처리는 각 이벤트에 하나의 타임스탬프를 사용하여 즉각적인 이벤트를 처리한다. 하지만, 시점기반의 이벤트 처리로는 이벤트의 활동 기간이 중요한 역할을 하는 금융, 멀티미디어, 의학, 기상학 같은 분야에서 복합적인 시제 관계를 표현하기에는 불충분하다. 실세계의 애플리케이션 분야에서, 이벤트는 기간을 가지며, 두 종류 이상의 이벤트는 시간적으로 겹쳐질 수도 있고, 하나의 이벤트가 다른 이벤트를 포함할 수도 있다. 이런 종류의 이벤트들에 대한 관계는 시점기반 이벤트처럼 연속적이지 않을 수도 있다. 본 논문에서는 기간기반 이벤트를 사용하여 복합 이벤트의 패턴을 검출하는 방법을 설계하고 구현한다. 기간기반 이벤트는 시점기반 이벤트가 다룰 수 없는 이벤트들 사이의 겹침과 포함관계를 표현할 수 있다. 기간기반 이벤트 연산자는 시작 끝점과 종료 끝점을 사용하여 이벤트의 기간을 나타내고, 기간기반 이벤트의 시퀀스를 표현하여 복합 이벤트 패턴을 검출할 수 있다. 본 논문에서는 복합 이벤트 패턴 검출의 효율성을 높이기 위해 활성 인스턴스 스택을 사용하는 알고리즘을 제시하며, 이벤트의 시퀀스를 구성할 때 중간 결과의 개수를 줄이기 위해 윈도우 푸시다운 기법을 적용하여 수행시간과 메모리의 효율을 높인다.

• 말소리와 음성과학
• /
• 제5권1호
• /
• pp.109-121
• /
• 2013

This study examined several kinematic properties of the primary articulator (the tongue dorsum) and the supplementary articulator (the jaw) in the articulation of the voiceless velar stop (/k/) within nonassimilating contexts. We examined in particular the spatiotemporal properties (constriction duration and constriction maxima) from the constriction onset to the constriction offset by analyzing a velar (/k/) followed by the coronal fricative (/s/), the coronal stop (/t/), and the labial (/p/) in across-word boundary conditions (/k#s/, /k#t/, and /k#p/). Along with these measurements, we investigated intergestural temporal coordination between C1 and C2 and the jaw articulator in relation to its coordination with the articulation of consonant sequences. The articulatory movement data was collected by means of electromagnetic midsagittal articulometry (EMMA). Four native speakers of Seoul Korean participated in the laboratory experiment. The results showed several characteristics. First, a velar (/k/) in C1 was not categorically reduced. Constriction duration and constriction degree of the velar (/k/) were similar within nonassimilating contexts (/k#s/=/k#t/=/k#p/). This might mean that spatiotemporal attributes during constriction duration were stable and consistent across different contexts, which might be subsequently associated with the nontarget status of the velar in place assimilation. Second, the gestural overlap could be represented as the order of /k#s/ (less) < /k#p/ (intermediate) < /k#t/ (more) as we measured the onset-to-onset lag (a longer lag indicated shorter gestural overlap.). This indicates a gestural overlap within nonassimilating contexts may not be constrained by any of the several constraints including the perceptual recoverability constraint (e.g., more overlap in Front-to-Back sequences compared to the reverse order (Back-to-Front) since perceptual cues in C1 can be recovered anytime during C2 articulation), the low-level speech motor constraint (e.g., more overlap in lingual-nonlingual sequences as compared to the lingual-lingual sequences), or phonological contexts effects (e.g., similarity in gestural overlap within nonassimilating contexts). As one possible account for more overlap in /k#t/ sequences as compared to /k#p/, we suspect speakers' knowledge may be receptive to extreme encroachment on C1 by the gestural overlap of the coronal in C2 since it does not obscure the perceptual cue of C1 as much as the labial in C2. Third, actual jaw position during C2 was higher in coronals (/s/, /t/) than in the labial (/p/). However, within the coronals, there was no manner-dependent jaw height difference in C2 (/s/=/t/). Vertical jaw position of C1 and C2 was seen as inter-dependent as higher jaw position in C1 was closely associated with C2. Lastly, a greater gap in jaw height was associated with longer intergestural timing (e.g., less overlap), but was confined to the cluster type (/kp/) with the lingual-nonlingual sequence. This study showed that Korean jaw articulation was independent from coordinating primary articulators in gestural overlap in some cluster types (/k#s/, /k#t/) while not in others (e.g., /k#p/). Overall, the results coherently indicate the velar stop (/k/) in C1 was robust in articulation, which may have subsequently contributed to the nontarget status of the velar (/k/) in place assimilation processes.

• 수면정신생리
• /
• 제5권2호
• /
• pp.119-133
• /
• 1998

Attentional processes facilitate cognitive and behavioral performance in several ways. Attention serves to reduce the amount of information to receive. Attention enables humans to direct themselves to appropriate aspects of external environmental events and internal operations. Attention facilitates the selection of salient information and the allocation of cognitive processing appropriate to that information. Attention is not a unitary process that can be localized to a single neuroanatomical region. Before the cortical registration of sensory information, activation of important subcortical structures occurs, which is called as an orienting response. Once sensory information reaches the sensory cortex, a large number of perceptual processes occur, which provide various levels of perceptual resolution of the critical features of the stimuli. After this preattentional processing, information is integrated within higher cortical(heteromodal) systems in inferior parietal and temporal lobes. At this stage, the processing characteristics can be modified, and the biases of the system have a direct impact on attentional selection. Information flow has been traced through sensory analysis to a processing stage that enables the new information to be focused and modified in relation to preexisting biases. The limbic and paralimbic system play significant roles in modulating attentional response. It is labeled with affective salience and is integrated according to ongoing pressures from the motivational drive system of the hypothalamus. The salience of information greatly influences the allocation of attention. The frontal lobe operate response selection system with a reciprocal interaction with both the attention system of the parietal lobe and the limbic system. In this attentional process, the search with the spatial field is organized and a sequence of attentional responses is generated. Affective, motivational and appectitive impulses from limbic system and hypothalamus trigger response intention, preparation, planning, initiation and control of frontal lobe on this process. The reticular system, which produces ascending activation, catalyzes the overall system and increases attentional capacity. Also additional energetic pressures are created by the hypothalamus. As psychophysiological measurement, skin conductance, pupil diameter, muscle tension, heart rate, alpha wave of EEG can be used. Event related potentials also provide physiological evidence of attention during information process. NI component appears to be an electrophysiological index of selective attention. P3 response is developed during the attention related to stimulus discrimination, evaluation and response.