首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献，搜索用时 62 毫秒
1.
ABSTRACT

Trajectory data mining is a lively research field in the domain of spatio-temporal data mining. Trajectory pattern mining comprises a set of specific pattern mining methods, which are applied as consecutive steps on a trajectory with the goal to extract and classify re-occurring spatio-temporal patterns. Despite the common nature and frequent usage of such methods by the GIScience community, a methodological approach is missing so far, especially when it comes to the use of machine learning-based classification methods. The current work closes this gap by proposing and evaluating a machine learning-based 3-steps trajectory data mining methodology using the detection and classification of stop points in vehicle trajectories as example. The work describes in detail the applied methodologies with respect to the three mining steps ‘stop detection’, ‘feature extraction’ and ‘classification in traffic-relevant and non-traffic-relevant stops’ and evaluates six machine learning-based classification algorithms using a real-world dataset of 15,498 vehicle trajectories with 5,899 detected stops (thereof 2,032 manually classified). Due to its exemplary nature, the presented methodology is suited to act as blueprint for similar trajectory data mining problems.  相似文献   

2.
ABSTRACT

Point cloud classification, which provides meaningful semantic labels to the points in a point cloud, is essential for generating three-dimensional (3D) models. Its automation, however, remains challenging due to varying point densities and irregular point distributions. Adapting existing deep-learning approaches for two-dimensional (2D) image classification to point cloud classification is inefficient and results in the loss of information valuable for point cloud classification. In this article, a new approach that classifies point cloud directly in 3D is proposed. The approach uses multi-scale features generated by deep learning. It comprises three steps: (1) extract single-scale deep features using 3D convolutional neural network (CNN); (2) subsample the input point cloud at multiple scales, with the point cloud at each scale being an input to the 3D CNN, and combine deep features at multiple scales to form multi-scale and hierarchical features; and (3) retrieve the probabilities that each point belongs to the intended semantic category using a softmax regression classifier. The proposed approach was tested against two publicly available point cloud datasets to demonstrate its performance and compared to the results produced by other existing approaches. The experiment results achieved 96.89% overall accuracy on the Oakland dataset and 91.89% overall accuracy on the Europe dataset, which are the highest among the considered methods.  相似文献   

3.
Understanding the bias of call detail records in human mobility research   总被引：1，自引：0，他引：1  
ABSTRACT

In recent years, call detail records (CDRs) have been widely used in human mobility research. Although CDRs are originally collected for billing purposes, the vast amount of digital footprints generated by calling and texting activities provide useful insights into population movement. However, can we fully trust CDRs given the uneven distribution of people’s phone communication activities in space and time? In this article, we investigate this issue using a mobile phone location dataset collected from over one million subscribers in Shanghai, China. It includes CDRs (~27%) plus other cellphone-related logs (e.g., tower pings, cellular handovers) generated in a workday. We extract all CDRs into a separate dataset in order to compare human mobility patterns derived from CDRs vs. from the complete dataset. From an individual perspective, the effectiveness of CDRs in estimating three frequently used mobility indicators is evaluated. We find that CDRs tend to underestimate the total travel distance and the movement entropy, while they can provide a good estimate to the radius of gyration. In addition, we observe that the level of deviation is related to the ratio of CDRs in an individual’s trajectory. From a collective perspective, we compare the outcomes of these two datasets in terms of the distance decay effect and urban community detection. The major differences are closely related to the habit of mobile phone usage in space and time. We believe that the event-triggered nature of CDRs does introduce a certain degree of bias in human mobility research and we suggest that researchers use caution to interpret results derived from CDR data.  相似文献   

4.
ABSTRACT

Datasets collecting the ever-changing position of moving individuals are usually big and possess high spatial and temporal resolution to reveal activity patterns of individuals in greater detail. Information about human mobility, such as ‘when, where and why people travel’, is contained in these datasets and is necessary for urban planning and public policy making. Nevertheless, how to segregate the users into groups with different movement and behaviours and generalise the patterns of groups are still challenging. To address this, this article develops a theoretical framework for uncovering space-time activity patterns from individual’s movement trajectory data and segregating users into subgroups according to these patterns. In this framework, individuals’ activities are modelled as their visits to spatio-temporal region of interests (ST-ROIs) by incorporating both the time and places the activities take place. An individual’s behaviour is defined as his/her profile of time allocation on the ST-ROIs she/he visited. A hierarchical approach is adopted to segregate individuals into subgroups based upon the similarity of these individuals’ profiles. The proposed framework is tested in the analysis of the behaviours of London foot patrol police officers based on their GPS trajectories provided by the Metropolitan Police.  相似文献   

5.
ABSTRACT

Effective public transit planning needs to address realistic travel demands, which can be illustrated by corridors across major residential areas and activity centers. It is vital to identify public transit corridors that contain the most significant transit travel demand patterns. We propose a two-stage approach to discover primary public transit corridors at high spatio-temporal resolutions using massive real-world smart card and bus trajectory data, which manifest rich transit demand patterns over space and time. The first stage was to reconstruct chained trips for individual passengers using multi-source massive public transit data. In the second stage, a shared-flow clustering algorithm was developed to identify public transit corridors based on reconstructed individual transit trips. The proposed approach was evaluated using transit data collected in Shenzhen, China. Experimental results demonstrated that the proposed approach is a practical tool for extracting time-varying corridors for many potential applications, such as transit planning and management.  相似文献   

6.
王菲  杨秋菊 《极地研究》2018,30(2):123-131
极光是由带电粒子经磁层—电离层碰撞大气而产生的。面对形态各异、演变过程复杂的极光图像,对其合理分类为进一步探究日地电磁活动和能量耦合等空间物理问题奠定了基础。针对该问题,引入深度学习的方法,通过卷积神经网络模型自主表征极光特征并实现极光图像分类。该方法对2003年北极黄河站越冬观测的38 044幅和8 001幅典型极光图像分类正确率达93.17%和91.5%;自动识别2004—2009年观测数据的极光形态,4类极光时间分布规律与三波段激发谱能量分布基本一致。实验结果表明,基于卷积神经网络的极光表征方法,能有效实现极光图像的自动分类。  相似文献   

7.
ABSTRACT

Terrain feature detection is a fundamental task in terrain analysis and landscape scene interpretation. Discovering where a specific feature (i.e. sand dune, crater, etc.) is located and how it evolves over time is essential for understanding landform processes and their impacts on the environment, ecosystem, and human population. Traditional induction-based approaches are challenged by their inefficiency for generalizing diverse and complex terrain features as well as their performance for scalable processing of the massive geospatial data available. This paper presents a new deep learning (DL) approach to support automatic detection of terrain features from remotely sensed images. The novelty of this work lies in: (1) a terrain feature database containing 12,000 remotely sensed images (1,000 original images and 11,000 derived images from data augmentation) that supports data-driven model training and new discovery; (2) a DL-based object detection network empowered by ensemble learning and deep and deeper convolutional neural networks to achieve high-accuracy object detection; and (3) fine-tuning the model’s characteristics and behaviors to identify the best combination of hyperparameters and other network factors. The introduction of DL into geospatial applications is expected to contribute significantly to intelligent terrain analysis, landscape scene interpretation, and the maturation of spatial data science.  相似文献   

8.
ABSTRACT

Agricultural terraces are important for agricultural production and soil-and-water conservation. They comprise treads and risers that require manual construction and maintenance. If managed improperly, risers will collapse, causing soil loss, gully erosion, and cultivation threats. However, mapping terrace risers remains a challenge. This study presents a novel approach to automatically map terrace risers by combining remote sensing images and digital elevation models (DEMs). First, a terraced hillslope was extracted via a hill-shading method and edges in the image were detected using a Canny edge detector. Next, the DEM was used to generate the contour direction, and edges along this direction were searched and coded as candidate terrace risers via directional detection. Finally, the results of directional detection and the edge image obtained from the Canny detector were overlaid to backtrack complete terrace risers. The approach was validated using four study areas with different topographic characteristics in the Loess Plateau, China. The results verify that the approach achieves outstanding performance and robustness in mapping terrace risers. The precision, recall, and F-measure were 90.81%–97.57%, 88.53%–94.10%, and 90.13%–95.80%, respectively. This approach is flexible and applicable with freely available images and DEM sources.  相似文献   

9.
ABSTRACT

Spatial interpolation is a traditional geostatistical operation that aims at predicting the attribute values of unobserved locations given a sample of data defined on point supports. However, the continuity and heterogeneity underlying spatial data are too complex to be approximated by classic statistical models. Deep learning models, especially the idea of conditional generative adversarial networks (CGANs), provide us with a perspective for formalizing spatial interpolation as a conditional generative task. In this article, we design a novel deep learning architecture named conditional encoder-decoder generative adversarial neural networks (CEDGANs) for spatial interpolation, therein combining the encoder-decoder structure with adversarial learning to capture deep representations of sampled spatial data and their interactions with local structural patterns. A case study on elevations in China demonstrates the ability of our model to achieve outstanding interpolation results compared to benchmark methods. Further experiments uncover the learned spatial knowledge in the model’s hidden layers and test the potential to generalize our adversarial interpolation idea across domains. This work is an endeavor to investigate deep spatial knowledge using artificial intelligence. The proposed model can benefit practical scenarios and enlighten future research in various geographical applications related to spatial prediction.  相似文献   

10.
This article describes a novel approach for finding similar trajectories, using trajectory segmentation based on movement parameters (MPs) such as speed, acceleration, or direction. First, a segmentation technique is applied to decompose trajectories into a set of segments with homogeneous characteristics with respect to a particular MP. Each segment is assigned to a movement parameter class (MPC), representing the behavior of the MP. Accordingly, the segmentation procedure transforms a trajectory to a sequence of class labels, that is, a symbolic representation. A modified version of edit distance called normalized weighted edit distance (NWED) is introduced as a similarity measure between different sequences. As an application, we demonstrate how the method can be employed to cluster trajectories. The performance of the approach is assessed in two case studies using real movement datasets from two different application domains, namely, North Atlantic Hurricane trajectories and GPS tracks of couriers in London. Three different experiments have been conducted that respond to different facets of the proposed techniques and that compare our NWED measure to a related method.  相似文献   

11.
ABSTRACT

The increasing popularity of Location-Based Social Networks (LBSNs) and the semantic enrichment of mobility data in several contexts in the last years has led to the generation of large volumes of trajectory data. In contrast to GPS-based trajectories, LBSN and context-aware trajectories are more complex data, having several semantic textual dimensions besides space and time, which may reveal interesting mobility patterns. For instance, people may visit different places or perform different activities depending on the weather conditions. These new semantically rich data, known as multiple-aspect trajectories, pose new challenges in trajectory classification, which is the problem that we address in this paper. Existing methods for trajectory classification cannot deal with the complexity of heterogeneous data dimensions or the sequential aspect that characterizes movement. In this paper we propose MARC, an approach based on attribute embedding and Recurrent Neural Networks (RNNs) for classifying multiple-aspect trajectories, that tackles all trajectory properties: space, time, semantics, and sequence. We highlight that MARC exhibits good performance especially when trajectories are described by several textual/categorical attributes. Experiments performed over four publicly available datasets considering the Trajectory-User Linking (TUL) problem show that MARC outperformed all competitors, with respect to accuracy, precision, recall, and F1-score.  相似文献   

12.
Human mobility patterns can provide valuable information in understanding the impact of human behavioral regularities in urban systems, usually with a specific focus on traffic prediction, public health or urban planning. While existing studies on human movement have placed huge emphasis on spatial location to predict where people go next, the time dimension component is usually being treated with oversimplification or even being neglected. Time dimension is crucial to understanding and detecting human activity changes, which play a negative role in prediction and thus may affect the predictive accuracy. This study aims to predict human movement from a spatio-temporal perspective by taking into account the impact of activity changes. We analyze and define changes of human activity and propose an algorithm to detect such changes, based on which a Markov chain model is used to predict human movement. The Microsoft GeoLife dataset is used to test our methodology, and the data of two selected users is used to evaluate the performance of the prediction. We compare the predictive accuracy (R2) derived from the data with and without implementing the activity change detection. The results show that the R2 is improved from 0.295 to 0.762 for the user with obvious activity changes and from 0.965 to 0.971 for the user without obvious activity changes. The method proposed by this study improves the accuracy in analyzing and predicting human movement and lays the foundation for related urban studies.  相似文献   

13.
ABSTRACT

Missing data is a common problem in the analysis of geospatial information. Existing methods introduce spatiotemporal dependencies to reduce imputing errors yet ignore ease of use in practice. Classical interpolation models are easy to build and apply; however, their imputation accuracy is limited due to their inability to capture spatiotemporal characteristics of geospatial data. Consequently, a lightweight ensemble model was constructed by modelling the spatiotemporal dependencies in a classical interpolation model. Temporally, the average correlation coefficients were introduced into a simple exponential smoothing model to automatically select the time window which ensured that the sample data had the strongest correlation to missing data. Spatially, the Gaussian equivalent and correlation distances were introduced in an inverse distance-weighting model, to assign weights to each spatial neighbor and sufficiently reflect changes in the spatiotemporal pattern. Finally, estimations of the missing values from temporal and spatial were aggregated into the final results with an extreme learning machine. Compared to existing models, the proposed model achieves higher imputation accuracy by lowering the mean absolute error by 10.93 to 52.48% in the road network dataset and by 23.35 to 72.18% in the air quality station dataset and exhibits robust performance in spatiotemporal mutations.  相似文献   

14.
With a huge volume of trajectories being collected and stored in databases, more and more researchers try to discover outlying trajectories from trajectory databases. In this article, we propose a novel framework called relative distance-based trajectory outliers detection (RTOD). In RTOD, we first employed relative distances to measure the dissimilarity between trajectory segments, and then formally defined the outlying trajectories based on distance measures. In order to improve the time performance, we proposed an optimization method that employs R-tree and local feature correlation matrix to eliminate unrelated trajectory segments. Finally, we conducted extensive experiments to estimate the advantages of the proposed approach. The experimental results show that our proposed approach is more efficient and effective at identifying outlying trajectories than existing algorithms. Particularly, we analyzed the effect of each parameter in theory.  相似文献   

15.
Spatial clustering can be used to discover hotspots in trajectory data. A trajectory clustering approach based on decision graph and data field is proposed as an effective method to select parameters for clustering, to determine the number of clusters, and to identify cluster centers. Synthetic data and real-world taxi trajectory data are utilized to demonstrate the effectiveness of the proposed approach. Results show that the proposed method can automatically determine the parameters for clustering as well as perform efficiently in trajectory clustering. Hotspots are identified and visualized during different times of a single day and at the same times on different days. The dynamic patterns of hotspots can be used to identify crowded areas and events, which are crucial for urban transportation planning and management.  相似文献   

16.
ABSTRACT

Animal movement is a dynamic spatio-temporal process. While trajectory data reflect the instantaneous animal position in space and time, other factors influence movement decisions between these observed positions. While some methods incorporate environmental (habitat) context into their understanding of the animal movement process, it is often captured in terms of simple parameters or weights influencing model results; primary behavioral data are not used directly to inform these models. Here, a new space-time constrained agent-based model is introduced, capable of producing ordered, behaviorally informed animal potential paths between observed space-time anchors. Potential paths generated by this approach incorporate both observed animal behavior and classical space-time constraints, and are used to construct associated visit probability distributions. Additionally, the notion of a behavioral space-time path is introduced, a variant of the space-time path based on the results of behaviorally aware animal movement simulation. The results of this approach demonstrate a means to better understand the varied movement opportunities within space-time prisms from an animal behavior perspective. From a spatial ecology perspective, not only is the environmental context considered, but the animal’s choice of transition and movement magnitude between contexts is modeled. This approach provides insight into the complex sequence of behaviorally informed actions driving animal movement decision-making.  相似文献   

17.
Abstract

This paper presents an approach to qualitative and spatial assessment of slum population numbers in Hyderabad, India using circle-based population data from the Census of India and results of the analysis of high resolution QuickBird satellite image data (2003) derived from automatic line detection and lacunarity algorithm. This approach provides plausible and spatially explicit aggregate statistics of slum population numbers within the city. This work suggests that both over- and underreporting of slum population numbers does occur in Hyderabad, and provides an improved view on the slum distribution patterns within this urban agglomeration.  相似文献   

18.
The advanced technologies in location-based services and telecom have yield large volumes of trajectory data. Understanding these data effectively requires intuitive yet accurate visual analysis. The visual analysis of massive trajectory data is challenged by the numerous interactions among different locations, which cause massive clutter. This paper presents a new methodology for visual analysis by integrating algebraic multigrid (AMG) method in data aggregation. The non-parametric method helps to build a multi-layer node representation from a graph which is extracted from trajectory data. The comparison with AMG and other methods shows that AMG method is more advanced in both the spatial representation and the importance of nodes. The new method is tested with real-world dataset of cell-phone signalling records in Beijing. The results show that our method is suitable for processing and creating abstraction of massive trajectory dataset, revealing inherent patterns and creating intuitive and vivid flow maps.  相似文献   

19.
Spatiotemporal proximity analysis to determine spatiotemporal proximal paths is a critical step for many movement analysis methods. However, few effective methods have been developed in the literature for spatiotemporal proximity analysis of movement data. Therefore, this study proposes a space-time-integrated approach for spatiotemporal proximal analysis considering space and time dimensions simultaneously. The proposed approach is based on space-time buffering, which is a natural extension of conventional spatial buffering operation to space and time dimensions. Given a space-time path and spatial tolerance, space-time buffering constructs a space-time region by continuously generating spatial buffers for any location along the space-time path. The constructed space-time region can delimit all space-time locations whose spatial distances to the target trajectory are less than a given tolerance. Five space-time overlapping operations based on this space-time buffering are proposed to retrieve all spatiotemporal proximal trajectories to the target space-time path, in terms of different spatiotemporal proximity metrics of space-time paths, such as Fréchet distance and longest common subsequence. The proposed approach is extended to analyze space-time paths constrained in road networks. The compressed linear reference technique is adopted to implement the proposed approach for spatiotemporal proximity analysis in large movement datasets. A case study using real-world movement data verifies that the proposed approach can efficiently retrieve spatiotemporal proximal paths constrained in road networks from a large movement database, and has significant computational advantage over conventional space-time separated approaches.  相似文献   

20.
ABSTRACT

Building footprints are among the most predominant features in urban areas, and provide valuable information for urban planning, solar energy suitability analysis, etc. We aim to automatically and rapidly identify building footprints by leveraging deep learning techniques and the increased availability of remote sensing datasets at high spatial resolution. The task is computationally challenging due to the use of large training datasets and large number of parameters. In related work, You-Only-Look-Once (YOLO) is a state-of-the-art deep learning framework for object detection. However, YOLO is limited in its capacity to identify small objects that appear in groups, which is the case for building footprints. We propose a LOcally-COnstrained (LOCO) You-Only-Look-Once framework to detect small and densely-distributed building footprints. LOCO is a variant of YOLO. Its layer architecture is determined by the spatial characteristics of building footprints and it uses a constrained regression modeling to improve the robustness of building size predictions. We also present an invariant augmentation based voting scheme to further improve the precision in the prediction phase. Experiments show that LOCO can greatly improve the solution quality of building detection compared to related work.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号