Robust Visual Ship Tracking with an Ensemble Framework via Multi-View Learning and Wavelet Filter
Maritime surveillance videos provide crucial on-spot kinematic traffic information (traffic volume, ship speeds, headings, etc.) for varied traffic participants (maritime regulation departments, ship crew, ship owners, etc.) which greatly benefits automated maritime situational awareness and maritim...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-02-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/20/3/932 |
_version_ | 1798040525398343680 |
---|---|
author | Xinqiang Chen Huixing Chen Huafeng Wu Yanguo Huang Yongsheng Yang Wenhui Zhang Pengwen Xiong |
author_facet | Xinqiang Chen Huixing Chen Huafeng Wu Yanguo Huang Yongsheng Yang Wenhui Zhang Pengwen Xiong |
author_sort | Xinqiang Chen |
collection | DOAJ |
description | Maritime surveillance videos provide crucial on-spot kinematic traffic information (traffic volume, ship speeds, headings, etc.) for varied traffic participants (maritime regulation departments, ship crew, ship owners, etc.) which greatly benefits automated maritime situational awareness and maritime safety improvement. Conventional models heavily rely on visual ship features for the purpose of tracking ships from maritime image sequences which may contain arbitrary tracking oscillations. To address this issue, we propose an ensemble ship tracking framework with a multi-view learning algorithm and wavelet filter model. First, the proposed model samples ship candidates with a particle filter following the sequential importance sampling rule. Second, we propose a multi-view learning algorithm to obtain raw ship tracking results in two steps: extracting a group of distinct ship contour relevant features (i.e., Laplacian of Gaussian, local binary pattern, Gabor filter, histogram of oriented gradient, and canny descriptors) and learning high-level intrinsic ship features by jointly exploiting underlying relationships shared by each type of ship contour features. Third, with the help of the wavelet filter, we performed a data quality control procedure to identify abnormal oscillations in the ship positions which were further corrected to generate the final ship tracking results. We demonstrate the proposed ship tracker’s performance on typical maritime traffic scenarios through four maritime surveillance videos. |
first_indexed | 2024-04-11T22:08:50Z |
format | Article |
id | doaj.art-80a058ab063c41f8ac8ef8bec82cf1c9 |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-04-11T22:08:50Z |
publishDate | 2020-02-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-80a058ab063c41f8ac8ef8bec82cf1c92022-12-22T04:00:37ZengMDPI AGSensors1424-82202020-02-0120393210.3390/s20030932s20030932Robust Visual Ship Tracking with an Ensemble Framework via Multi-View Learning and Wavelet FilterXinqiang Chen0Huixing Chen1Huafeng Wu2Yanguo Huang3Yongsheng Yang4Wenhui Zhang5Pengwen Xiong6Institute of Logistics Science and Engineering, Shanghai Maritime University, Shanghai 201306, ChinaMerchant Marine College, Shanghai Maritime University, Shanghai 201306, ChinaMerchant Marine College, Shanghai Maritime University, Shanghai 201306, ChinaSchool of Electrical Engineering and Automation, Jiangxi University of Science and Technology, Ganzhou 341000, ChinaInstitute of Logistics Science and Engineering, Shanghai Maritime University, Shanghai 201306, ChinaSchool of Traffic and Transportation, Northeast Forestry University, Harbin 150040, ChinaSchool of Information Engineering, Nanchang University, Nanchang 330031, ChinaMaritime surveillance videos provide crucial on-spot kinematic traffic information (traffic volume, ship speeds, headings, etc.) for varied traffic participants (maritime regulation departments, ship crew, ship owners, etc.) which greatly benefits automated maritime situational awareness and maritime safety improvement. Conventional models heavily rely on visual ship features for the purpose of tracking ships from maritime image sequences which may contain arbitrary tracking oscillations. To address this issue, we propose an ensemble ship tracking framework with a multi-view learning algorithm and wavelet filter model. First, the proposed model samples ship candidates with a particle filter following the sequential importance sampling rule. Second, we propose a multi-view learning algorithm to obtain raw ship tracking results in two steps: extracting a group of distinct ship contour relevant features (i.e., Laplacian of Gaussian, local binary pattern, Gabor filter, histogram of oriented gradient, and canny descriptors) and learning high-level intrinsic ship features by jointly exploiting underlying relationships shared by each type of ship contour features. Third, with the help of the wavelet filter, we performed a data quality control procedure to identify abnormal oscillations in the ship positions which were further corrected to generate the final ship tracking results. We demonstrate the proposed ship tracker’s performance on typical maritime traffic scenarios through four maritime surveillance videos.https://www.mdpi.com/1424-8220/20/3/932visual ship trackingmulti-view learningwavelet filterdata quality controlsmart ship |
spellingShingle | Xinqiang Chen Huixing Chen Huafeng Wu Yanguo Huang Yongsheng Yang Wenhui Zhang Pengwen Xiong Robust Visual Ship Tracking with an Ensemble Framework via Multi-View Learning and Wavelet Filter Sensors visual ship tracking multi-view learning wavelet filter data quality control smart ship |
title | Robust Visual Ship Tracking with an Ensemble Framework via Multi-View Learning and Wavelet Filter |
title_full | Robust Visual Ship Tracking with an Ensemble Framework via Multi-View Learning and Wavelet Filter |
title_fullStr | Robust Visual Ship Tracking with an Ensemble Framework via Multi-View Learning and Wavelet Filter |
title_full_unstemmed | Robust Visual Ship Tracking with an Ensemble Framework via Multi-View Learning and Wavelet Filter |
title_short | Robust Visual Ship Tracking with an Ensemble Framework via Multi-View Learning and Wavelet Filter |
title_sort | robust visual ship tracking with an ensemble framework via multi view learning and wavelet filter |
topic | visual ship tracking multi-view learning wavelet filter data quality control smart ship |
url | https://www.mdpi.com/1424-8220/20/3/932 |
work_keys_str_mv | AT xinqiangchen robustvisualshiptrackingwithanensembleframeworkviamultiviewlearningandwaveletfilter AT huixingchen robustvisualshiptrackingwithanensembleframeworkviamultiviewlearningandwaveletfilter AT huafengwu robustvisualshiptrackingwithanensembleframeworkviamultiviewlearningandwaveletfilter AT yanguohuang robustvisualshiptrackingwithanensembleframeworkviamultiviewlearningandwaveletfilter AT yongshengyang robustvisualshiptrackingwithanensembleframeworkviamultiviewlearningandwaveletfilter AT wenhuizhang robustvisualshiptrackingwithanensembleframeworkviamultiviewlearningandwaveletfilter AT pengwenxiong robustvisualshiptrackingwithanensembleframeworkviamultiviewlearningandwaveletfilter |