ReID-DeePNet: A Hybrid Deep Learning System for Person Re-Identification
Person re-identification has become an essential application within computer vision due to its ability to match the same person over non-overlapping cameras. However, it is a challenging task because of the broad view of cameras with a large number of pedestrians appearing with various poses. As a r...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-09-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/10/19/3530 |
| _version_ | 1827653987563732992 |
|---|---|
| author | Hussam J. Mohammed Shumoos Al-Fahdawi Alaa S. Al-Waisy Dilovan Asaad Zebari Dheyaa Ahmed Ibrahim Mazin Abed Mohammed Seifedine Kadry Jungeun Kim |
| author_facet | Hussam J. Mohammed Shumoos Al-Fahdawi Alaa S. Al-Waisy Dilovan Asaad Zebari Dheyaa Ahmed Ibrahim Mazin Abed Mohammed Seifedine Kadry Jungeun Kim |
| author_sort | Hussam J. Mohammed |
| collection | DOAJ |
| description | Person re-identification has become an essential application within computer vision due to its ability to match the same person over non-overlapping cameras. However, it is a challenging task because of the broad view of cameras with a large number of pedestrians appearing with various poses. As a result, various approaches of supervised model learning have been utilized to locate and identify a person based on the given input. Nevertheless, several of these approaches perform worse than expected in retrieving the right person in real-time over multiple CCTVs/camera views. This is due to inaccurate segmentation of the person, leading to incorrect classification. This paper proposes an efficient and real-time person re-identification system, named ReID-DeePNet system. It is based on fusing the matching scores generated by two different deep learning models, convolutional neural network and deep belief network, to extract discriminative feature representations from the pedestrian image. Initially, a segmentation procedure was developed based on merging the advantages of the Mask R-CNN and GrabCut algorithm to tackle the adverse effects caused by background clutter. Afterward, the two different deep learning models extracted discriminative feature representations from the pedestrian segmented image, and their matching scores were fused to make the final decision. Several extensive experiments were conducted, using three large-scale and challenging person re-identification datasets: Market-1501, CUHK03, and P-DESTRE. The ReID-DeePNet system achieved new state-of-the-art Rank-1 and mAP values on these three challenging ReID datasets. |
| first_indexed | 2024-03-09T21:28:18Z |
| format | Article |
| id | doaj.art-2de25d4fc58b42c696c8bf27efe89116 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-09T21:28:18Z |
| publishDate | 2022-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-2de25d4fc58b42c696c8bf27efe891162023-11-23T21:03:06ZengMDPI AGMathematics2227-73902022-09-011019353010.3390/math10193530ReID-DeePNet: A Hybrid Deep Learning System for Person Re-IdentificationHussam J. Mohammed0Shumoos Al-Fahdawi1Alaa S. Al-Waisy2Dilovan Asaad Zebari3Dheyaa Ahmed Ibrahim4Mazin Abed Mohammed5Seifedine Kadry6Jungeun Kim7Computer Center, University of Anbar, Ramadi 31001, IraqComputer Science Department, Al-Ma’aref University College, Ramadi 31001, IraqComputer Engineering Technology Department, Information Technology Collage, Imam Ja’afar Al-Sadiq University, Baghdad 10072, IraqDepartment of Computer Science, College of Science, Nawroz University, Duhok 42001, IraqComputer Engineering Technology Department, Information Technology Collage, Imam Ja’afar Al-Sadiq University, Baghdad 10072, IraqCollege of Computer Science and Information Technology, University of Anbar, Ramadi 31001, IraqDepartment of Applied Data Science, Noroff University College, 4612 Kristiansand, NorwayDepartment of Software, Kongju National University, Cheonan 31080, KoreaPerson re-identification has become an essential application within computer vision due to its ability to match the same person over non-overlapping cameras. However, it is a challenging task because of the broad view of cameras with a large number of pedestrians appearing with various poses. As a result, various approaches of supervised model learning have been utilized to locate and identify a person based on the given input. Nevertheless, several of these approaches perform worse than expected in retrieving the right person in real-time over multiple CCTVs/camera views. This is due to inaccurate segmentation of the person, leading to incorrect classification. This paper proposes an efficient and real-time person re-identification system, named ReID-DeePNet system. It is based on fusing the matching scores generated by two different deep learning models, convolutional neural network and deep belief network, to extract discriminative feature representations from the pedestrian image. Initially, a segmentation procedure was developed based on merging the advantages of the Mask R-CNN and GrabCut algorithm to tackle the adverse effects caused by background clutter. Afterward, the two different deep learning models extracted discriminative feature representations from the pedestrian segmented image, and their matching scores were fused to make the final decision. Several extensive experiments were conducted, using three large-scale and challenging person re-identification datasets: Market-1501, CUHK03, and P-DESTRE. The ReID-DeePNet system achieved new state-of-the-art Rank-1 and mAP values on these three challenging ReID datasets.https://www.mdpi.com/2227-7390/10/19/3530person re-identificationdeep learningdeep belief networkMask R-CNNGrabCut algorithmMarket-1501 dataset |
| spellingShingle | Hussam J. Mohammed Shumoos Al-Fahdawi Alaa S. Al-Waisy Dilovan Asaad Zebari Dheyaa Ahmed Ibrahim Mazin Abed Mohammed Seifedine Kadry Jungeun Kim ReID-DeePNet: A Hybrid Deep Learning System for Person Re-Identification Mathematics person re-identification deep learning deep belief network Mask R-CNN GrabCut algorithm Market-1501 dataset |
| title | ReID-DeePNet: A Hybrid Deep Learning System for Person Re-Identification |
| title_full | ReID-DeePNet: A Hybrid Deep Learning System for Person Re-Identification |
| title_fullStr | ReID-DeePNet: A Hybrid Deep Learning System for Person Re-Identification |
| title_full_unstemmed | ReID-DeePNet: A Hybrid Deep Learning System for Person Re-Identification |
| title_short | ReID-DeePNet: A Hybrid Deep Learning System for Person Re-Identification |
| title_sort | reid deepnet a hybrid deep learning system for person re identification |
| topic | person re-identification deep learning deep belief network Mask R-CNN GrabCut algorithm Market-1501 dataset |
| url | https://www.mdpi.com/2227-7390/10/19/3530 |
| work_keys_str_mv | AT hussamjmohammed reiddeepnetahybriddeeplearningsystemforpersonreidentification AT shumoosalfahdawi reiddeepnetahybriddeeplearningsystemforpersonreidentification AT alaasalwaisy reiddeepnetahybriddeeplearningsystemforpersonreidentification AT dilovanasaadzebari reiddeepnetahybriddeeplearningsystemforpersonreidentification AT dheyaaahmedibrahim reiddeepnetahybriddeeplearningsystemforpersonreidentification AT mazinabedmohammed reiddeepnetahybriddeeplearningsystemforpersonreidentification AT seifedinekadry reiddeepnetahybriddeeplearningsystemforpersonreidentification AT jungeunkim reiddeepnetahybriddeeplearningsystemforpersonreidentification |