Scalable Pattern Recognition and Real Time Tracking of Moving Objects

Pudasaini, Dipak; Abhari, Abdolreza

doi:10.23919/springsim.2019.8732855

Cited by 9 publications

(5 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then, two multicriteria decision‐making methods were employed to obtain an aggregated score that was utilized to estimate whether the trajectory had unusual patterns. In [49], the trajectories were divided into two modes, normal and abnormal, depending on whether the vehicles were moving in a straight line. In addition, the results of two trajectory pattern recognition learning methods, the spectral clustering algorithm and naive Bayes algorithm, were compared.…”

Section: Related Workmentioning

confidence: 99%

Trajectory dimensionality reduction and hyperparameter settings of DBSCAN for trajectory clustering

Long

et al. 2022

IET Intelligent Trans Sys

View full text Add to dashboard Cite

The density-based spatial clustering of application with noise (DBSCAN) algorithm has good robustness and is widely employed to cluster vehicle trajectories for vehicle movement pattern recognition. However, the distance or similarity between two trajectories varies from tens to hundreds of thousands, and there is no effective method for determining the values of the hyperparameters eps and MinPts of DBSCAN. In addition, with increasing sizes of trajectory datasets, some trajectory clustering methods that directly analyse points and line segments incur large computational costs and time overhead. To solve these two dilemmas, the authors propose an effective trajectory dimensionality reduction method and a DBSCAN hyperparameter initial value setting method. The trajectory dimensionality reduction algorithm processes trajectories with different lengths into the same dimensionality (the same number of feature points). The reserved points preserve the spatial and temporal information of these trajectories as much as possible. The DBSCAN hyperparameter initial value setting algorithm obtains the effective initial values of eps and MinPts for facilitating subsequent adjustments. Finally, we validate these proposed methods on two trajectory datasets collected from two real-world scenes, and the experimental results are promising and effective.

show abstract

Section: Related Workmentioning

confidence: 99%

Trajectory dimensionality reduction and hyperparameter settings of DBSCAN for trajectory clustering

Long

et al. 2022

IET Intelligent Trans Sys

View full text Add to dashboard Cite

show abstract

“…Some of these methods are based on CNN, and some are not. The tracking methods without using CNN are faster but have low accuracy [5]. The CNN based tracking methods are more accurate, but the execution time is high [6].…”

Section: Problem Statements and Contributionsmentioning

confidence: 99%

“…The large numbers of trajectories collected from tracking model are passed to machine learning algorithms for pattern recognition. The scalable pattern recognition of moving objects has been proposed in [5]. The supervised and unsupervised machine learning algorithms have been used for the pattern recognition of the trajectories.…”

Section: ) Pattern Recognitionmentioning

confidence: 99%

Edge-based Video Analytic for Smart Cities

Pudasaini¹,

Abhari²

2021

IJACSA

Self Cite

View full text Add to dashboard Cite

Video analytic is the important tool for smart city development. The video analytic application requires more memories and high processing devices. The problems of cloudbased approach for video analytic are high latency and more network bandwidth to transfer data into the cloud. To overcome these problems, we propose a model based on dividing the jobs into smaller sub-tasks with less processing requirements in a typical video analytics application for the development of smart city. The object detection, tracking and pattern recognition method to reduce the size of videos based on edge network will be proposed. We will design a video analytic model, and simulation is performed using iFogSim simulator. We will also propose Convolutional Neural Network (CNN) based object tracking model. The experimental verification shows that our tracking model is more than 96% accurate, and the proposed edge and cloud-based model is more than 80% effective than only cloudbased approach for video analytic applications.

show abstract

“…Spatial similarity focuses on finding trajectories with similar geometric shapes and ignores the temporal dimension, such as the Euclidean distance (ED) [14], the closest-pair distance (CPD) [15], the one-way distance (OWD) [16], the Hausdorff distance [17], and the angular metric for shape similarity [18]. Spatiotemporal similarity takes into account both the spatial and the temporal dimensions of trajectories, such as the dynamic time warping (DTW) [19][20][21], the edit distance on real sequences (EDR) [22], and the longest common subsequence (LCSS) [23]. Trajectory matching algorithms have been widely studied in recent years [24][25][26].…”

Section: Literature Reviewmentioning

confidence: 99%

Phone‐vehicle trajectory matching framework based on ALPR and cellular signalling data

Wan

Cai

2020

IET Intelligent Trans Sys

View full text Add to dashboard Cite

With the advancement of positioning techniques, a large amount of trajectory data has been produced. Matching vehicles with mobile phones using different trajectories can benefit many applications, such as driving behaviour analysis and travel mode split. Moreover, as a privacy attack method, it can provide theoretical inspiration for privacy protection theory. To address this problem, a new trajectory matching framework for processing massive Automatic License Plate Recognition (ALPR) and cellular signalling data is proposed. Information entropy was adopted to address the movement frequency of trajectories and then the infrequent vehicles and phones that did not meet the threshold were pruned. Next, an effective matching algorithm was devised to match the trajectories of vehicles and mobile phones. Moreover, to solve the problem of obtaining a small number of matching results, a data augmentation algorithm was proposed to add new, matching records. Last, a classification model was constructed with LightGBM to determine whether the vehicle matches the phone. Experimental results on real datasets show that the framework outperforms typical techniques in terms of effectiveness and efficiency. The data obtained by data augmentation have distribution characteristics similar to those of the original data. The proposed classification model achieves an accuracy of 93.6%.

show abstract

Scalable Pattern Recognition and Real Time Tracking of Moving Objects

Cited by 9 publications

References 16 publications

Trajectory dimensionality reduction and hyperparameter settings of DBSCAN for trajectory clustering

Trajectory dimensionality reduction and hyperparameter settings of DBSCAN for trajectory clustering

Edge-based Video Analytic for Smart Cities

Phone‐vehicle trajectory matching framework based on ALPR and cellular signalling data

Contact Info

Product

Resources

About