• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.1
• /
• pp.216-238
• /
• 2023

In intelligent transportation systems, traffic management is an important task. The accurate forecasting of traffic characteristics like flow, congestion, and density is still active research because of the non-linear nature and uncertainty of the spatiotemporal data. Inclement weather, such as rain and snow, and other special events such as holidays, accidents, and road closures have a significant impact on driving and the average speed of vehicles on the road, which lowers traffic capacity and causes congestion in a widespread manner. This work designs a model for multivariate short-term traffic congestion prediction using SLSTM_AE-BiLSTM. The proposed design consists of a Bidirectional Long Short Term Memory(BiLSTM) network to predict traffic flow value and a Convolutional Neural network (CNN) model for detecting the congestion status. This model uses spatial static temporal dynamic data. The stacked Long Short Term Memory Autoencoder (SLSTM AE) is used to encode the weather features into a reduced and more informative feature space. BiLSTM model is used to capture the features from the past and present traffic data simultaneously and also to identify the long-term dependencies. It uses the traffic data and encoded weather data to perform the traffic flow prediction. The CNN model is used to predict the recurring congestion status based on the predicted traffic flow value at a particular urban traffic network. In this work, a publicly available Caltrans PEMS dataset with traffic parameters is used. The proposed model generates the congestion prediction with an accuracy rate of 92.74% which is slightly better when compared with other deep learning models for congestion prediction.

• Journal of KIISE:Computing Practices and Letters
• /
• v.11 no.3
• /
• pp.284-291
• /
• 2005

I propose a framework for detecting, recognizing and classifying facial features based on learned expression patterns. The framework recognizes facial expressions by using PCA and expression HMM(EHMM) which is Hidden Markov Model (HMM) approach to represent the spatial information and the temporal dynamics of the time varying visual expression patterns. Because the low level spatial feature extraction is fused with the temporal analysis, a unified spatio-temporal approach of HMM to common detection, tracking and classification problems is effective. The proposed recognition framework is accomplished by applying posterior probability between current visual observations and previous visual evidences. Consequently, the framework shows accurate and robust results of recognition on as well simple expressions as basic 6 facial feature patterns. The method allows us to perform a set of important tasks such as facial-expression recognition, HCI and key-frame extraction.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.18 no.3
• /
• pp.39-50
• /
• 2015

This research selected Daegu Metropolitan City representing a combination of natural space and urban space for this case study. To achieve this, a prerequisite was to set up an optimal block size to evaluate landscape diversity of the research site by using a RPR-Area Curve. Further, landscape diversity evaluation was conducted based on land cover map by using FRAGSTATS to analyze spatio-temporal changes. Notably, this research regarded it as the most significant to set forth criteria in order to apply landscape diversity to the development plans of the newtown and outskirt of a city under high pressure development. Results derived from this research are summarized as follows. According to the results derived from establishing the optimal block size, a size about $2km^2$ was analyzed to measure landscape diversity of the research site. Also, according to the results derived from land diversity evaluation based on land cover map, land diversity was highly measured around urban stream such as Nakdong River and Geumho River, and in particular, the value of landscape diversity was measured considerably high around the urban parks. Results derived from analysis on spatio-temporal changes of land diversity demonstrated that a certain level of urban development exerted a positive effect on an increase in land diversity, but consistent urban development lowered a value of landscape diversity. Results derived from regression analysis to set forth the optimal urban space showed that an urban area of a space about $2km^2$ exerted a positive effect at a rate of about 0~43.3% and a negative effect at a rate about 43.3~100%. In conclusion, the results of this research are considered to provide important basic data for future urban and landscape planning. Nonetheless, as only the layout on the 2D plane was analyzed in this research, further research in future is required to complexly consider diverse factors such as height of structure and change in visible real area arising from geographical features.

• Journal of Electrical Engineering and Technology
• /
• v.9 no.6
• /
• pp.2118-2125
• /
• 2014

Gait Recognition is a new technique to identify the people by the way they walk. Human gait is a spatio-temporal phenomenon that typifies the motion characteristics of an individual. The proposed method makes a simple but efficient attempt to gait recognition. For each video file, spatial silhouettes of a walker are extracted by an improved background subtraction procedure using Gaussian Mixture Model (GMM). Here GMM is used as a parametric probability density function represented as a weighted sum of Gaussian component densities. Then, the relevant features are extracted from the silhouette tracked from the given video file using the Principal Component Analysis (PCA) method. The Fisher Linear Discriminant Analysis (FLDA) classifier is used in the classification of dimensional reduced image derived by the PCA method for gait recognition. Although gait images can be easily acquired, the gait recognition is affected by clothes, shoes, carrying status and specific physical condition of an individual. To overcome this problem, it is combined with footprint as a multimodal biometric system. The minutiae is extracted from the footprint and then fused with silhouette image using the Discrete Stationary Wavelet Transform (DSWT). The experimental result shows that the efficiency of proposed fusion algorithm works well and attains better result while comparing with other fusion schemes.

• Proceedings of the KSRS Conference
• /
• 2007.10a
• /
• pp.361-364
• /
• 2007

A small sized device with computing, communicating, sensing capability is changing our life. It will be deployed in the world and acquire a lot of data from the world. It is used for various applications such as military surveillance, environmental monitoring, structure health monitoring, building management, asset tracking, etc. In this paper we focus on USN middleware for asset tracking. A mobile asset is moving here and there within a specific area. The USN middleware tracks the mobile assets in real-time by using sensor nodes and notify their current positions to a user. To achieve the goal, the USN middleware provides some features related to the positions of mobile assets.. They are storing location data by using 3D indexing method, retrieving them by using spatio-temporal query, making trace of an asset, and retrieving the history data of an asset. In the paper, we developed USN middleware to adapt the requirements of asset tracking. It can help users increase the efficiency of their business related to mobile assets and make a valuable decision.

• Wind and Structures
• /
• v.25 no.3
• /
• pp.301-328
• /
• 2017

Multivariate fluctuating pressures acting on a 2:1 rectangular section (2-D) with dimensions of 9 cm by 4.5 cm has been studied using wind tunnel experiments under uniform and smooth flow condition for various angles of wind incidence. Based on the variation of mean pressure coefficient distributions along the circumference of the rectangular section with angle of wind incidence, and with the aid of skin friction coefficients, three distinct flow regimes with two transition regimes have been identified. Further, variations of mean drag and lift coefficients, Strouhal number with angles of wind incidence have been studied. The applicability of Universal Strouhal number based on vortex street similarity of wakes in bluff bodies to the 2:1 rectangular section has been studied for different angles of wind incidence. The spatio-temporal correlation features of the measured pressure data have been studied using Proper Orthogonal Decomposition (POD) technique. The contribution of individual POD modes to the aerodynamic force components, viz, drag and lift, have been studied. It has been demonstrated that individual POD modes can be associated to different physical phenomena, which contribute to the overall aerodynamic forces.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.9
• /
• pp.2782-2804
• /
• 2024

The integration of artificial intelligence technology with medicine has rapidly evolved, with increasing demands for quality of life. However, falls remain a significant risk leading to severe injuries and fatalities, especially among the elderly. Therefore, the development and application of computer vision-based fall detection technologies have become increasingly important. In this paper, firstly, the keypoint detection algorithm ViTPose++ is used to obtain the coordinates of human body keypoints from the camera images. Human skeletal feature maps are generated from this keypoint coordinate information. Meanwhile, human dense feature maps are produced based on the DensePose algorithm. Then, these two types of feature maps are confused as dual-channel inputs for the model. The convolutional gated recurrent unit is introduced to extract the frame-to-frame relevance in the process of falling. To further integrate features across three dimensions (spatio-temporal-channel), a dual-channel fall detection algorithm based on video streams is proposed by combining the Convolutional Block Attention Module (CBAM) with the ConvGRU. Finally, experiments on the public UR Fall Detection Dataset demonstrate that the improved ConvGRU-CBAM achieves an F1 score of 92.86% and an AUC of 95.34%.

• Journal of the Korean Geographical Society
• /
• v.48 no.4
• /
• pp.509-532
• /
• 2013

Minimum Housing Standard is an instrument to cope with the problems of public health and community hygiene, deterioration of working class housing conditions appeared commonly in the process of capitalist industrialization and rapid rural-to-urban migration. This paper aims to examine the institutionalization of histories of minimum housing standard in the advanced countries, and analyze the spatio-temporal changes and characteristics of households failing to meet the New Minimum Housing Standard in Seoul Metropolitan since 1995. The analysis of this paper is based on the census data on population and housing. The results are as follows; Households failing to meet the New Minimum Housing Standard in Seoul are 501,000 households(1.368 million person, 14.4%). This means Seoul has overtaken the national average 11.8% for the first time and there are structurally marginal band of households who can not improve the housing conditions by themselves. In addition, the fact that the rate of Seoul households living in the marginal shelter including the basement and rooftop room is the highest in Korea means the housing quality issues of Seoul is serious. Spatial distribution of households failing to meet the standard is divided into the northeast area and the southwest area in Seoul. Main features of the households are female-headed families, middle and old-aged people, divorce families, lower educated people, under and graduate students, non-apartments, dweller in 15~20 year old houses.

• Journal of Korean Society of Forest Science
• /
• v.103 no.1
• /
• pp.105-112
• /
• 2014

The main purpose of this study is to measure spatio-temporal variation of forest tree volume based on the RCP(Representative Concentration Pathway) 8.5 scenario, targeting on Pinus densiflora forests which is the main tree species in South Korea. To estimate nationwide scale, $5^{th}$ forest type map and National Forest Inventory data were used. Also, to reflect the impact of change in place and climate on growth of forest trees, growth model reflecting the climate and topography features were applied. The result of the model validation, which compared the result of the model with the forest statistics of different cities and provinces, showed a high suitability. Considering the continuous climate change, volume of Pinus densiflora forest is predicted to increase from $131m^3/ha$ at present to $212.42m^3/ha$ in the year of 2050. If the climate maintains as the present, volume is predicted to increase to $221.92m^3/ha$. With the climate change, it is predicted that most of the region, except for some of the alpine region, will have a decrease in growth rate of Pinus densiflora forest. The growth rate of Pinus densiflora forest will have a greater decline, especially in the coastal area and the southern area. With the result of this study, it will be possible to quantify the effect of climate change on the growth of Pinus densiflora forest according to spatio-temporal is possible. The result of the study can be useful in establishing the forest management practices, considering the adaptation of climate change.

• Review of Culture and Economy
• /
• v.20 no.2
• /
• pp.89-117
• /
• 2017

As the industrial paradigm shifts and the manufacturing industry declines, many changes also take place in the region as well. In this regard, interest in industrial heritage as a facet of cultural heritage is on the increase. In this paper, the meaning of regional 'cultural contents' as industrial heritage is investigated within the scope of specific region. It is meant to move beyond the viewpoint of considering industrial heritage as only relating to industrial machinery and relevant landmarks from the past. The concept of industrial heritage is established more clearly through the review policy and case study analysis of existing research; the analysis is conducted to investigate the characteristics associated with it, and then to explore how best to utilize it. In particular, this paper attempts to focus on how it operates within these parameters using a spatio-temporal context as much as possible, and concentrating on the recognition and experience of the subject of industrial heritage as being traceable through human story. This research is based on the case of 'Sapporo' which focuses on modern history based on historical importance, and the 'Cheongju' case study, which contrasts the former by focusing on urban regeneration using a spatial lens. This paper identifies the possibility of regional development through the examination of past identity and diversity in the present, and highlights the features that could be linked to future usability and development. In addition, it proposes the possibility that the cycle of regional development could change in the process of the different stages of territorialization, de-territorialization and re-territorialization.