Local Compact Binary Count Based Nonparametric Background Modeling for Foreground Detection in Dynamic Scenes
Background subtraction is one of the most fundamental and challenging tasks in computer vision. Many background subtraction algorithms work well under the assumption that the backgrounds are static over short time periods but degrade dramatically in dynamic scenes, such as swaying trees, rippling wa...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2019-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8758405/ |
| _version_ | 1818911309757939712 |
|---|---|
| author | Wei He Yong K-Wan Kim Hak-Lim Ko Jianhui Wu Wujing Li Bing Tu |
| author_facet | Wei He Yong K-Wan Kim Hak-Lim Ko Jianhui Wu Wujing Li Bing Tu |
| author_sort | Wei He |
| collection | DOAJ |
| description | Background subtraction is one of the most fundamental and challenging tasks in computer vision. Many background subtraction algorithms work well under the assumption that the backgrounds are static over short time periods but degrade dramatically in dynamic scenes, such as swaying trees, rippling water, and waving curtains. In this paper, we propose an effective background subtraction method to address these difficulties by combining color features with texture features in the ViBe framework. Specifically, we present a novel local compact binary count (LCBC) feature that can capture local binary gray-scale difference information and totally discard the local binary structural information. The effective fusion of color and LCBC information significantly improves the performance of the ViBe model, making it very robust to background variations while still highlighting the moving objects. We further embed the total variation (TV) norm regularization technique into the proposed method, which can enhance the spatial smoothness of foreground objects, thereby further improving the accuracy of the method. We evaluate the proposed method against ten sequences containing dynamic backgrounds and show that our method outperforms many state-of-the-art methods in reducing the false positives without compromising the reasonable foreground definitions. The experimental results on challenging well-known data sets demonstrate that the proposed method works effectively on a wide range of dynamic background scenes. |
| first_indexed | 2024-12-19T22:56:39Z |
| format | Article |
| id | doaj.art-2b80a935c54641cea2299658afcb956a |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-19T22:56:39Z |
| publishDate | 2019-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-2b80a935c54641cea2299658afcb956a2022-12-21T20:02:38ZengIEEEIEEE Access2169-35362019-01-017923299234010.1109/ACCESS.2019.29277458758405Local Compact Binary Count Based Nonparametric Background Modeling for Foreground Detection in Dynamic ScenesWei He0https://orcid.org/0000-0003-4603-5615Yong K-Wan Kim1Hak-Lim Ko2Jianhui Wu3Wujing Li4Bing Tu5https://orcid.org/0000-0001-5802-9496School of Information Science and Engineering, Hunan Institute of Science and Technology, Yueyang, ChinaDepartment of Information and Communication Engineering, Hoseo University, Asan, South KoreaDepartment of Information and Communication Engineering, Hoseo University, Asan, South KoreaSchool of Information Science and Engineering, Hunan Institute of Science and Technology, Yueyang, ChinaSchool of Information Science and Engineering, Hunan Institute of Science and Technology, Yueyang, ChinaSchool of Information Science and Engineering, Hunan Institute of Science and Technology, Yueyang, ChinaBackground subtraction is one of the most fundamental and challenging tasks in computer vision. Many background subtraction algorithms work well under the assumption that the backgrounds are static over short time periods but degrade dramatically in dynamic scenes, such as swaying trees, rippling water, and waving curtains. In this paper, we propose an effective background subtraction method to address these difficulties by combining color features with texture features in the ViBe framework. Specifically, we present a novel local compact binary count (LCBC) feature that can capture local binary gray-scale difference information and totally discard the local binary structural information. The effective fusion of color and LCBC information significantly improves the performance of the ViBe model, making it very robust to background variations while still highlighting the moving objects. We further embed the total variation (TV) norm regularization technique into the proposed method, which can enhance the spatial smoothness of foreground objects, thereby further improving the accuracy of the method. We evaluate the proposed method against ten sequences containing dynamic backgrounds and show that our method outperforms many state-of-the-art methods in reducing the false positives without compromising the reasonable foreground definitions. The experimental results on challenging well-known data sets demonstrate that the proposed method works effectively on a wide range of dynamic background scenes.https://ieeexplore.ieee.org/document/8758405/Foreground detectionnonparametric background modelinglocal compact binary countdynamic backgroundvideo signal processing |
| spellingShingle | Wei He Yong K-Wan Kim Hak-Lim Ko Jianhui Wu Wujing Li Bing Tu Local Compact Binary Count Based Nonparametric Background Modeling for Foreground Detection in Dynamic Scenes IEEE Access Foreground detection nonparametric background modeling local compact binary count dynamic background video signal processing |
| title | Local Compact Binary Count Based Nonparametric Background Modeling for Foreground Detection in Dynamic Scenes |
| title_full | Local Compact Binary Count Based Nonparametric Background Modeling for Foreground Detection in Dynamic Scenes |
| title_fullStr | Local Compact Binary Count Based Nonparametric Background Modeling for Foreground Detection in Dynamic Scenes |
| title_full_unstemmed | Local Compact Binary Count Based Nonparametric Background Modeling for Foreground Detection in Dynamic Scenes |
| title_short | Local Compact Binary Count Based Nonparametric Background Modeling for Foreground Detection in Dynamic Scenes |
| title_sort | local compact binary count based nonparametric background modeling for foreground detection in dynamic scenes |
| topic | Foreground detection nonparametric background modeling local compact binary count dynamic background video signal processing |
| url | https://ieeexplore.ieee.org/document/8758405/ |
| work_keys_str_mv | AT weihe localcompactbinarycountbasednonparametricbackgroundmodelingforforegrounddetectionindynamicscenes AT yongkwankim localcompactbinarycountbasednonparametricbackgroundmodelingforforegrounddetectionindynamicscenes AT haklimko localcompactbinarycountbasednonparametricbackgroundmodelingforforegrounddetectionindynamicscenes AT jianhuiwu localcompactbinarycountbasednonparametricbackgroundmodelingforforegrounddetectionindynamicscenes AT wujingli localcompactbinarycountbasednonparametricbackgroundmodelingforforegrounddetectionindynamicscenes AT bingtu localcompactbinarycountbasednonparametricbackgroundmodelingforforegrounddetectionindynamicscenes |