• Journal of Information Processing Systems
• /
• 제11권1호
• /
• pp.39-56
• /
• 2015

Continuous multi-interval prediction (CMIP) is used to continuously predict the trend of a data stream based on various intervals simultaneously. The continuous integrated hierarchical temporal memory (CIHTM) network performs well in CMIP. However, it is not suitable for CMIP in real-time mode, especially when the number of prediction intervals is increased. In this paper, we propose a real-time integrated hierarchical temporal memory (RIHTM) network by introducing a new type of node, which is called a Zeta1FirstSpecializedQueueNode (ZFSQNode), for the real-time continuous multi-interval prediction (RCMIP) of data streams. The ZFSQNode is constructed by using a specialized circular queue (sQUEUE) together with the modules of original hierarchical temporal memory (HTM) nodes. By using a simple structure and the easy operation characteristics of the sQUEUE, entire prediction operations are integrated in the ZFSQNode. In particular, we employed only one ZFSQNode in each level of the RIHTM network during the prediction stage to generate different intervals of prediction results. The RIHTM network efficiently reduces the response time. Our performance evaluation showed that the RIHTM was satisfied to continuously predict the trend of data streams with multi-intervals in the real-time mode.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제36권6호
• /
• pp.479-490
• /
• 2009

여러 가지 실세계 문제들은 마르코프 결정 문제(Markov decision problem) 들로 형식화하여 풀 수 있으나, 풀이 과정의 높은 계산 복잡도 때문에 실세계 문제들을 직접적으로 다루는 데 많은 어려움이 있다. 이를 해결하기 위해 많은 시간적 추상화(Temporal abstraction) 방법들이 제안되어 왔고 이를 자동화하기 위한 여러 방법들 또한 연구되어 왔으나, 이들 방법들은 명시적인 효율성 척도를 갖고 있지 않아 이론적인 성능 보장을 하지 못하는 문제가 있었다. 본 연구에서는 문제의 크기가 커지더라도 좋은 성능이 보장되는 자동적인 시간적 추상화 구현 방법에 대해 제안한다. 이를 위하여 네트워크 척도(Network measurements)를 이용하여 마르코프 결정 문제의 풀이 효율과 상태 궤적 그래프(State trajectory graph)의 위상 특성간의 관계를 분석하고, 네트워크 척도들 중 평균 측지 거리(Mean geodesic distance)가 마르코프 결정 문제의 풀이 성능과 밀접한 관계가 있다는 사실을 알아내었다. 이 사실을 기반으로 하여, 낮은 평균 측지 거리를 보장하는 복잡계망 모델(Complex network model)을 사용하여 시간적 추상화를 만들어 나가는 알고리즘을 제안한다. 제안된 알고리즘은 사실적인 3차원 게임 환경을 비롯한 여러 문제에 대해 테스트되었고, 문제 크기의 증가에도 불구하고 효율적인 풀이 성능을 보여 주었다.

• ETRI Journal
• /
• 제42권3호
• /
• pp.411-419
• /
• 2020

In real-world intelligent transportation systems, accuracy in vehicle license plate detection and recognition is considered quite critical. Many algorithms have been proposed for still images, but their accuracy on actual videos is not satisfactory. This stems from several problematic conditions in videos, such as vehicle motion blur, variety in viewpoints, outliers, and the lack of publicly available video datasets. In this study, we focus on these challenges and propose a license plate detection and recognition scheme for videos based on a temporal matching prior network. Specifically, to improve the robustness of detection and recognition accuracy in the presence of motion blur and outliers, forward and bidirectional matching priors between consecutive frames are properly combined with layer structures specifically designed for plate detection. We also built our own video dataset for the deep training of the proposed network. During network training, we perform data augmentation based on image rotation to increase robustness regarding the various viewpoints in videos.

• 대한원격탐사학회지
• /
• 제37권4호
• /
• pp.719-731
• /
• 2021

This study proposes a two-stage hybrid classification model for crop classification using multi-temporal remote sensing images; the model combines feature embedding by using an autoencoder (AE) with a convolutional neural network (CNN) classifier to fully utilize features including informative temporal and spatial signatures. Long short-term memory (LSTM)-based AE (LAE) is fine-tuned using class label information to extract latent features that contain less noise and useful temporal signatures. The CNN classifier is then applied to effectively account for the spatial characteristics of the extracted latent features. A crop classification experiment with multi-temporal unmanned aerial vehicle images is conducted to illustrate the potential application of the proposed hybrid model. The classification performance of the proposed model is compared with various combinations of conventional deep learning models (CNN, LSTM, and convolutional LSTM) and different inputs (original multi-temporal images and features from stacked AE). From the crop classification experiment, the best classification accuracy was achieved by the proposed model that utilized the latent features by fine-tuned LAE as input for the CNN classifier. The latent features that contain useful temporal signatures and are less noisy could increase the class separability between crops with similar spectral signatures, thereby leading to superior classification accuracy. The experimental results demonstrate the importance of effective feature extraction and the potential of the proposed classification model for crop classification using multi-temporal remote sensing images.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 1997년도 Proceedings International Workshop on New Video Media Technology
• /
• pp.43-48
• /
• 1997

The video transmission for real-time viewing over the Internet is a core operation for the multimedia services. However, its realization is very difficult because the Internet has two major problems, namely, very narrow endpoint-bandwidth and the network jitter. We already proposed a scalable video transmission method in [8] which used MPEG-4 video VM(Verification Model) 2.0[3] for very low bit rate coding and an adaptive temporal rate control of video objects to overcome the network jitter problem. In this paper, we present the improved adaptive temporal rate control scheme for the scalable transmission. Experimental results for three test video sequences show that the adaptive temporal rate control can transfer the video bitstream at source frame rate under variable network condition.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제3권5호
• /
• pp.527-547
• /
• 2009

Moving object management is widely used in traffic, logistic and data mining applications in ubiquitous environments. It is required to analyze spatio-temporal data and trajectories for moving object management. In this paper, we proposed a novel index structure for spatio-temporal aggregation of trajectory in a constrained network, named aCN-RB-tree. It manages aggregation values of trajectories using a constraint network-based index and it also supports direction of trajectory. An aCN-RB-tree consists of an aR-tree in its center and an extended B-tree. In this structure, an aR-tree is similar to a Min/Max R-tree, which stores the child nodes' max aggregation value in the parent node. Also, the proposed index structure is based on a constrained network structure such as a FNR-tree, so that it can decrease the dead space of index nodes. Each leaf node of an aR-tree has an extended B-tree which can store timestamp-based aggregation values. As it considers the direction of trajectory, the extended B-tree has a structure with direction. So this kind of aCN-RB-tree index can support efficient search for trajectory and traffic zone. The aCN-RB-tree can find a moving object trajectory in a given time interval efficiently. It can support traffic management systems and mining systems in ubiquitous environments.

• 응용통계연구
• /
• 제36권1호
• /
• pp.1-19
• /
• 2023

본 연구는 학생 간의 연결에 어떤 속성이 유효한지 종단 네트워크 분석을 통해 알아보고자 하였으며, 종단 네트워크 모형인 TERGM (temporal exponential random graph model)과 SAOM (stochastic actor-oriented model) 통계적 모형을 사용하고 결과를 비교하였다. TERGM 모형은 네트워크 전체의 연결 형성을 바탕으로, SAOM 모형은 특정 행위자가 형성하는 주변 네트워크를 대상으로 연구 결과를 해석하였다. TERGM 모형은 시간 항을 통해 이전 시점의 영향을 표현하였으며, SAOM 모형은 비율 함수로 행위자의 기회에 의해 진화하는 네트워크를 구현해 시간적 종속성을 고려하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권9호
• /
• pp.2361-2376
• /
• 2023

The electrocardiogram (ECG) signal is commonly used to screen and diagnose cardiovascular diseases. In recent years, deep neural networks have been regarded as an effective way for automatic ECG disease diagnosis. The convolutional neural network is widely used for ECG signal extraction because it can obtain different levels of information. However, most previous studies adopt single scale convolution filters to extract ECG signal features, ignoring the complementarity between ECG signal features of different scales. In the paper, we propose a dual-scale network with convolution filters of different sizes for 12-lead ECG classification. Our model can extract and fuse ECG signal features of different scales. In addition, different spatial and time periods of the feature map obtained from the 12-lead ECG may have different contributions to ECG classification. Therefore, we add a spatial-temporal attention to each scale sub-network to emphasize the representative local spatial and temporal features. Our approach is evaluated on PTB-XL dataset and achieves 0.9307, 0.8152, and 89.11 on macro-averaged ROC-AUC score, a maximum F1 score, and mean accuracy, respectively. The experiment results have proven that our approach outperforms the baselines.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1993년도 학술논문발표회 논문집 제12권 1호
• /
• pp.121-124
• /
• 1993

This paper is a study on the word recognition using neural network. A limited vocabulary, speaker independent, isolated word recognition system has been built. This system recognizes isolated word without performing segmentation, phoneme identification, or dynamic time wrapping. It needs a static pattern approach to recognize a spatio-temporal pattern. The preprocessing only includes preceding and tailing silence removal, and word length determination. A LPC analysis is performed on each of 24 equally spaced frames. The PARCOR coefficients plus 3 other features from each frame is extracted. In order to simplify a structure of neural network, we composed binary code form to decrease output nodes.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권5호
• /
• pp.1909-1928
• /
• 2020

Predicting human mobility has always been an important task in Location-based Social Network. Previous efforts fail to capture spatial dependence effectively, mainly reflected in weakening the location topology information. In this paper, we propose a neural network-based method which can capture spatial-temporal dependence to predict the next location of a person. Specifically, we involve a graph convolutional network (GCN) based on a seq2seq framework to capture the location topology information and temporal dependence, respectively. The encoder of the seq2seq framework first generates the hidden state and cell state of the historical trajectories. The GCN is then used to generate graph embeddings of the location topology graph. Finally, we predict future trajectories by aggregated temporal dependence and graph embeddings in the decoder. For evaluation, we leverage two real-world datasets, Foursquare and Gowalla. The experimental results demonstrate that our model has a better performance than the compared models.