Robust visual tracking via CAMShift and structural local sparse appearance model

Zhao, Houqiang; Ke, Xiang; Cao, Shuai; Wang, Xuanyin

doi:10.1016/j.jvcir.2015.11.008

Cited by 15 publications

(7 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The input ship training patterns are mapped into feature space φ(p m ), which is represented by the kernel h. The minimal solution for (2) is considered as a linear combination of the input ship training data, and we determine the maximum response of the KCF model by the (4). The optimal solution for the KCF tracker in both the linear and nonlinear situations is re-formulated as (5). Motivated by the studies in [24], [25], we can obtain a closed form solution for the KCF tracking algorithm which is shown in (6).…”

Section: A Ship Tracking With Kcf Modelmentioning

confidence: 99%

“…Chen et al proposed a robust ship tracking framework by combining multi-view and sparse representation algorithms [4]. Zhao et al developed a robust ship tracking model by adaptively fusing the Camshift algorithm and structural sparse appearance features [5]. Correlation filtering based tracking frameworks, the newly emerging branch for the discriminant models, show satisfied performance (i.e., track ship at high accuracy and speed) in the manner of determining the maximum responses between the image candidate region and the training ship samples.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Augmented Ship Tracking Under Occlusion Conditions From Maritime Surveillance Videos

Chen

Yang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Ship tracking provides crucial on-site microscopic kinematic traffic information which benefits maritime traffic flow analysis, ship safety enhancement, traffic control, etc., and thus has attracted considerable research attentions in the maritime surveillance community. Conventional ship tracking methods yield satisfied results by exploring distinct visual ship features in maritime images, which may fail when the target ship is partially or fully sheltered by obstacles (e.g., ships, waves, etc.) in maritime videos. To overcome the difficulty, we propose an augmented ship tracking framework via the kernelized correlation filter (KCF) and curve fitting algorithm. First, the KCF model is introduced to track ships in the consecutive maritime images and obtain raw ship trajectory dataset. Second, the data anomaly detection and rectification procedure are implemented to rectify the contaminated ship positions. For the purpose of performance evaluation, we implement the proposed framework and another three popular ship tracking models on the four typical ship occlusion videos. The experimental results show that our proposed framework successfully tracks ships in maritime video clips with high accuracy (i.e., the average root mean square error (RMSE), root mean square percentage error (RMSPE), mean absolute deviation (MAD) and mean absolute percentage error (MAPE) are less than 10), which significantly outperforms the other popular ship trackers.INDEX TERMS Smart ship, curve fitting, kernelized correlation filter, visual ship tracking, ship occlusion.

show abstract

Section: A Ship Tracking With Kcf Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Augmented Ship Tracking Under Occlusion Conditions From Maritime Surveillance Videos

Chen

Yang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Target model and tracking algorithm are two key factors directly affecting the tracking performance. The hue histogram model which has the advantages of a simple expression, less computation, and so on, has been used in many tracking methods, such as Meanshift and Camshift [7–10]. However, because the hue histogram model describes the target characteristics by the statistical information, it would cause losing some detailed information of the target.…”

Section: Introductionmentioning

confidence: 99%

Target tracking based on the cognitive associative network

Xiu

Chai

2019

IET Image Processing

View full text Add to dashboard Cite

“…the adaptive target scale and orientation measurement method (Zhao et al, 2016) is used to adapt to the severe deformation of the target outline; and a new model update strategy is put forward based on similarity measurement to achieve effective and accurate model update.…”

Section: Introductionmentioning

confidence: 99%

A target tracking and location robot system based on omnistereo vision

Cai

Fan

Weng

et al. 2017

View full text Add to dashboard Cite

Purpose Because of their large field of view, omnistereo vision systems have been widely used as primary vision sensors in autonomous mobile robot tasks. The purpose of this article is to achieve real-time and accurate tracking by the omnidirectional vision robot system. Design/methodology/approach The authors provide in this study the key techniques required to obtain an accurate omnistereo target tracking and location robot system, including stereo rectification and target tracking in complex environment. A simple rectification model is proposed, and a local image processing method is used to reduce the computation time in the localization process. A target tracking method is improved to make it suitable for omnidirectional vision system. Using the proposed methods and some existing methods, an omnistereo target tracking and location system is established. Findings The experiments are conducted with all the necessary stages involved in obtaining a high-performance omnistereo vision system. The proposed correction algorithm can process the image in real time. The experimental results of the improved tracking algorithm are better than the original algorithm. The statistical analysis of the experimental results demonstrates the effectiveness of the system. Originality/value A simple rectification model is proposed, and a local image processing method is used to reduce the computation time in the localization process. A target tracking method is improved to make it suitable for omnidirectional vision system. Using the proposed methods and some existing methods, an omnistereo target tracking and location system is established.

show abstract

Robust visual tracking via CAMShift and structural local sparse appearance model

Cited by 15 publications

References 24 publications

Augmented Ship Tracking Under Occlusion Conditions From Maritime Surveillance Videos

Augmented Ship Tracking Under Occlusion Conditions From Maritime Surveillance Videos

Target tracking based on the cognitive associative network

A target tracking and location robot system based on omnistereo vision

Contact Info

Product

Resources

About