學術出版;;Publications

Permanent URI for this communityhdl:11536/1

收集所有交通大學研究人員與學生個人或團體所發表之各項學術著作，涵蓋期刊論文、教師專書、學位論文等項。;;Collections of Publications of NCTU Researchers and Students Like Journal Papers and Thesis.

Browse

🏛️國立陽明交通大學機構典藏

Search Results

Now showing 1 - 10 of 93

TrackNet: A Deep Learning Network for Tracking High-speed and Tiny Objects in Sports Applications
(2019-01-01)
Huang, Yu-Chuan
;
Liao, I-No
;
Chen, Ching-Hsuan
;
Ik, Ts I-Ui
;
Peng, Wen-Chih
Ball trajectory data are one of the most fundamental and useful information in the evaluation of players' performance and analysis of game strategies. It is still challenging to recognize and position a high-speed and tiny ball accurately from an ordinary video. In this paper, we develop a deep learning network, called TrackNet, to track the tennis ball from broadcast videos in which the ball images are small, blurry, and sometimes with afterimage tracks or even invisible. The proposed heatmap-based deep learning network is trained to not only recognize the ball image from a single frame but also learn flying patterns from consecutive frames. The network is evaluated on the video of the men's singles final at the 2017 Summer Universiade, which is available on YouTube. The precision, recall, and F1-measure reach 99:7%, 97:3%, and 98:5%, respectively. To prevent overfitting, 9 additional videos are partially labeled together with a subset from the previous dataset to implement 10-fold cross-validation, and the precision, recall, and F1-measure are 95:3%, 75:7%, and 84:3%, respectively.
Inferring Distant-Time Location in Low-Sampling-Rate Trajectories
(2013-01-01)
Chiang, Meng-Fen
;
Lin, Yung-Hsiang
;
Peng, Wen-Chih
;
Yu, Philip S.
;
資訊工程學系
With the growth of location-based services and social services, low sampling-rate trajectories from check-in data or photos with geotag information becomes ubiquitous. In general, most detailed moving information in low-sampling-rate trajectories are lost. Prior works have elaborated on distant-time location prediction in highsampling-rate trajectories. However, existing prediction models are pattern-based and thus not applicable due to the sparsity of data points in low-sampling-rate trajectories. To address the sparsity in low-sampling-rate trajectories, we develop a Reachability-based prediction model on Time-constrained Mobility Graph (RTMG) to predict locations for distant-time queries. Specifically, we design an adaptive temporal exploration approach to extract effective supporting trajectories that are temporally close to the query time. Based on'the supporting trajectories, a Time-constrained mobility Graph (TG) is constructed to capture mobility information at the given query time. In light of TG, we further derive the reachability probabilities among locations in TG. Thus, a location with maximum reachability from the current location among all possi ble locations in supporting trajectories is considered as the prediction result. To efficiently process queries, we proposed the index structure Sorted Interval-Tree (SOIT) to organize location records. Extensive experiments with real data demonstrated the effectiveness and efficiency of RTMG. First, RTMG with adapthie temporal exploration significantly outperforms the existing pattern-based prediction model HPM [2] over varying data sparsity in terms of higher accuracy and higher coverage. Also, the proposed index structure SOIT can efficiently speedup RTMG in large-scale trajectory dataset. In the future, we could extend RTMG by considering more factors (e.g., staying durations in locations, application usages in smart phones) to further improve the prediction accuracy.
PTGF: Public Transport General Framework for Identifying Transport Modes Based on Cellular Data
(2019-01-01)
Gou, Xiaochuan
;
Hung, Chih-Chieh
;
Li, Guanyao
;
Peng, Wen-Chih
;
交大名義發表
Public transportation is beating heart of a city. Understanding how citizens utilize public transportation can be used to optimize many applications such as traffic planning, crowd flow prediction and location-based marketing. However, obtaining how citizens used transportation is not a trivial task. It is almost not possible to ask citizens to report their exact location and their transportation mode; moreover, there are usually various public transportation that move along the similar paths. These increase challenges to identify people's transport modes. To address these issues, this paper proposes Public Transport General Framework (PTGF) to identify people's transport modes by their cellular data in both offline and online manners. Regarding the offline phase, given historical cellular data of people and urban transportation networks, PTGF derives cellular data into trajectories, to match each trajectory to public transportation networks to find the most possible transport modes for sub-trajectories of a trajectory. In the online phase, given streaming trajectories, PTGF identifies the transport modes of each location by a LSTM which are trained by historical trajectories with transport modes annotated in the offline phase. Extensive experiments are conducted by using both synthetic and real datasets. The experimental results show that the accuracy of PTGF in offline phase around 80% and that in online phase F1-score around 0.7, which could prove that the effectiveness of the proposed framework PTGF.
Modeling User Mobility for Location Promotion in Location-based Social Networks
(2015-01-01)
Zhu, Wen-Yuan
;
Peng, Wen-Chih
;
Chen, Ling-Jyh
;
Zheng, Kai
;
Zhou, Xiaofang
With the explosion of smartphones and social network services, location-based social networks (LBSNs) are increasingly seen as tools for businesses (e.g., restaurants, hotels) to promote their products and services. In this paper, we investigate the key techniques that can help businesses promote their locations by advertising wisely through the underlying LBSNs. In order to maximize the benefit of location promotion, we formalize it as an influence maximization problem in an LBSN, i.e., given a target location and an LBSN, which a set of k users (called seeds) should be advertised initially such that they can successfully propagate and attract most other users to visit the target location. Existing studies have proposed different ways to calculate the information propagation probability, that is how likely a user may influence another, in the settings of static social network. However, it is more challenging to derive the propagation probability in an LBSN since it is heavily affected by the target location and the user mobility, both of which are dynamic and query dependent. This paper proposes two user mobility models, namely Gaussian-based and distance based mobility models, to capture the check-in behavior of individual LBSN user, based on which location-aware propagation probabilities can be derived respectively. Extensive experiments based on two real LBSN datasets have demonstrated the superior effectiveness of our proposals than existing static models of propagation probabilities to truly reflect the information propagation in LBSNs.
Interpretable Multi-Task Learning for Product Quality Prediction with Attention Mechanism
(2019-01-01)
Yeh, Cheng-Han
;
Fan, Yao-Chung
;
Peng, Wen-Chih
;
資訊工程學系
;
Department of Computer Science
In this paper, we investigate the problem of mining multivariate time series data generated from sensors mounted on manufacturing stations for early product quality prediction. In addition to accurate quality prediction, another crucial requirement for industrial production scenarios is model interpretability, i.e., to understand the significance of an individual time series with respect to the final quality. Aiming at the goals, this paper proposes a multi-task learning model with an encoder-decoder architecture augmented by the matrix factorization technique and the attention mechanism. Our model design brings two major advantages. First, by jointly considering the input multivariate time series reconstruction task and the quality prediction in a multi-task learning model, the performance of the quality prediction task is boosted. Second, by incorporating the matrix factorization technique, we enable the proposed model to pay/learn attentions on the component of the multivariate time series rather than on the time axis. With the attention on components, the correlation between a sensor reading and a final quality measure can be quantized to improve the model interpretability. Comprehensive performance evaluation on real data sets is conducted. The experimental results validate that strengths of the proposed model on quality prediction and model interpretability.
Exploiting Centrality Information with Graph Convolutions for Network Representation Learning
(2019-01-01)
Chen, Hongxu
;
Yin, Hongzhi
;
Chen, Tong
;
Quoc Viet Hung Nguyen
;
Peng, Wen-Chih
Network embedding has been proven effective to learn low-dimensional vector representations for network vertices, and recently received a tremendous amount of research attention. However, most of existing methods for network embedding merely focus on preserving the first and second order proximities between nodes, and the important properties of node centrality are neglected. Various centrality measures such as Degree, Closeness, Betweenness, Eigenvector and PageRank centralities have been designed to measure the importance of individual nodes. In this paper, we focus on a novel yet unsolved problem that aims to learn low-dimensional continuous nodes representations that not only preserve the network structure, but also keep the centrality information. We propose a generalizable model, namely GraphCSC, that utilizes both linkage information and centrality information to learn low-dimensional vector representations for network vertices. The learned embeddings by GraphCSC are able to preserve different centrality information of nodes. In addition, we further propose GraphCSC-M, a more comprehensive model that can preserve different centrality information simultaneously through learning multiple centrality specific embeddings, and a novel attentive multi-view learning approach is developed to compress multiple embeddings of one node into a compact vector representation. Extensive experiments have been conducted to demonstrate that our model is able to preserve different centrality information of nodes, and achieves better performance on several benchmark tasks compared with recent state-of-the-art network embedding methods.
Activity Sense Organs: Energy-efficient Activity Sensing with Adaptive Duty Cycle Control
(2016-01-01)
Wu, Fang-Jing
;
Tseng, Yu-Chee
;
Peng, Wen-Chih
;
資訊工程學系
;
Department of Computer Science
Recently, smartphones with built-in sensors are exploited to detect human activities. Since sensing human activities may exhaust energy of a smartphone, this work brings up the concept of "activity sense organs" to the uprising smartphone sensing paradigm. As the human sense organs are the most important sensors to perceive real-world information, this work identifies the most important sensors to perceive a particular human activity based on the "quality of the features" contributed by sensors. To optimize the energy usage of activity sensing, we exploit the importance of sensors to profile the duty-cycles of sensors for activity detection. Specifically, a smartphone will operate in a duty-cycled sensing mode, where sensors in the activity sense organs will collect data at the higher sampling rate than those sensors not in the activity sense organs. We design a knowledge lifecycle to find out the activity sense organs, profile the sensing modes of activity detection, and adjust the sensing mode for real-time activity detection. The experimental results indicate that the activity sense organs provide sufficient information for activity detection.
Stitching Aerial Images for Vehicle Positioning and Tracking
(2018-01-01)
Tsao, Paul
;
Ik, Tsi-Ui
;
Chen, Guan-Wen
;
Peng, Wen-Chih
;
資訊工程學系
In recent years, applications such as reproduction of traffic accident scene, calculation of traffic flow, etc. use UAVs to collect data by aerial images. If a UAV does not move, the coordinates of objects in images can be considered as reference for predicting parameters such as distance and direction. However, this does not apply if the UAV is moving. In this research, methods of image processing are used to obtain the pinhole model which is similar to perpendicularly taken images, which should be distance-preserving. Then, the method of image stitching is used to find out the correlation of relative position between each image in order to construct a global coordinates system over every images. The common used methods of image stitching are not considered about distance-preserving, and cause the cumulative error usually. In this case, GPS information is used to estimate the starting position of images, then stitching those images to a panorama through applying SIFT feature pair and fine-tuning by the gradient method. At last, this paper integrates methods as mentioned into implementation and presents visualized data, then both the trajectory of UAV and the relative position of objects between images and the panorama can be observed by user in order to verify feasibility and precision of methods. Also the results of this paper can be applied with traffic flows detection in the future.
A Study on Multiple Wearable Sensors for Activity Recognition
(2017-01-01)
Huang, Yu-Chuan
;
Yi, Chih-Wei
;
Peng, Wen-Chih
;
Lin, Hsing-Chen
;
Huang, Ching-Yu
In the past few years, human activity recognition is an active area of machine learning. The possible applications include daily activity monitoring for elders, exercise and fitness workout assistant systems, life style analysis, etc. In this work, tri-axial accelerometers were worn at the right wrist, left wrist and waist to collect motion data for activity recognition. Three supervised machine learning algorithms including random forests, decision trees and support vector machines were implemented to classify daily activities into running, walking, standing, sitting and dining from inertial data. The purposes of this study are to understand how good the machine learning algorithms can achieve and how the wearing location and number of sensors impact the recognition accuracy. Our results showed that the multi-sensors achieve the accuracy of 81%, and dominant hand sensor achieves the accuracy of 80%, which is 7% higher than non-dominant hand sensor.
Public Transportation Mode Detection from Cellular Data
(2017-01-01)
Li, Guanyao
;
Chen, Chun-Jie
;
Huang, Sheng-Yun
;
Chou, Ai-Jou
;
Gou, Xiaochuan
Public transportation is essential in people's daily life and it is crucial to understand how people move around the city. Some prior works have exploited GPS, Wi-Fi or bluetooth to collect data, in which extra sensors or devices were needed. Other works utilized data from smart card systems. However, some public transportation systems have their own smart card system and the smart card data cannot include all kinds of transportation modes, which makes it unsuitable for our study.Nowadays, each user has his/her own mobile phones and from the cellular data of mobile phone service providers, it is possible to know the uses' transportation mode and the fine-grained crowd flows. As such, given a set of cellular data, we propose a system for public transportation mode detection, crowd density estimation, and crowd flow estimation. Note that we only have cellular data, no extra sensor data collected from users' mobile phones. In this paper, we refer to some external data sources (e.g., the bus routing networks) to identify transportation modes. Users' cellular data sometimes have uncertainty about user location information. Thus, we propose two approaches for different transportation mode detection considering the cell tower properties, spatial and temporal factors. We demonstrate our system using the data from Chunghwa Telecom, which is the largest telecommunication company in Taiwan, to show the usefulness of our system.

學術出版;;Publications

Browse

Filters

Settings

Sort By

Results per page

🏛️國立陽明交通大學機構典藏

Search Results