Multi-Saliency Map and Machine Learning Based Human Detection for the Embedded Top-View Imaging System
Compared to the side view, a top-view is robust against occlusion generated by objects located indoors. It offers a better wide view angle and much visibility of a scene. However, there are still problems to be handled. The top-view image shows asymmetrical features and radially distorted scenes aro...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9426910/ |
| _version_ | 1818594784603799552 |
|---|---|
| author | Seung Jun Lee Byeong Hak Kim Min Young Kim |
| author_facet | Seung Jun Lee Byeong Hak Kim Min Young Kim |
| author_sort | Seung Jun Lee |
| collection | DOAJ |
| description | Compared to the side view, a top-view is robust against occlusion generated by objects located indoors. It offers a better wide view angle and much visibility of a scene. However, there are still problems to be handled. The top-view image shows asymmetrical features and radially distorted scenes around the corners, such as omnidirectional view images and self-occlusion. Conventional human detection methods are suitable for finding moving objects in front view imaging systems. And there are some limitations, such as slow execution speed due to computational complexity. In this paper, we propose an efficient method. A static saliency map with low activity and a dynamic saliency map with a lot of movement are respectively detected. These two models were fused to create a multi-saliency map, and both characteristics were used simultaneously to improve detection rates. To handle problems such as asymmetry, a rotation matrix was calculated around the center, and Histogram of Oriented Gradient (HOG) features descriptor were extracted from the multi-saliency map to create an image patch (a small image region of interest containing human candidates). For the classification of image patches, we used machine learning-based supervised learning models support-vector machine (SVM) algorithm to improve performance. As a result of the proposed algorithm, it showed low resource occupancy and achieved Average Precision of 92.3% and 96.12% when Intersection over Union were 50% and 45% respectively. |
| first_indexed | 2024-12-16T11:05:38Z |
| format | Article |
| id | doaj.art-9bbe9eb655344773846238d8f458f2b2 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-16T11:05:38Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-9bbe9eb655344773846238d8f458f2b22022-12-21T22:33:52ZengIEEEIEEE Access2169-35362021-01-019706717068210.1109/ACCESS.2021.30786239426910Multi-Saliency Map and Machine Learning Based Human Detection for the Embedded Top-View Imaging SystemSeung Jun Lee0https://orcid.org/0000-0001-5876-1572Byeong Hak Kim1Min Young Kim2https://orcid.org/0000-0001-7263-3403School of Electronic and Electrical Engineering, Kyungpook National University, Daegu, South KoreaKorea Institute of Industrial Technology, Daegu, South KoreaSchool of Electronic and Electrical Engineering, Kyungpook National University, Daegu, South KoreaCompared to the side view, a top-view is robust against occlusion generated by objects located indoors. It offers a better wide view angle and much visibility of a scene. However, there are still problems to be handled. The top-view image shows asymmetrical features and radially distorted scenes around the corners, such as omnidirectional view images and self-occlusion. Conventional human detection methods are suitable for finding moving objects in front view imaging systems. And there are some limitations, such as slow execution speed due to computational complexity. In this paper, we propose an efficient method. A static saliency map with low activity and a dynamic saliency map with a lot of movement are respectively detected. These two models were fused to create a multi-saliency map, and both characteristics were used simultaneously to improve detection rates. To handle problems such as asymmetry, a rotation matrix was calculated around the center, and Histogram of Oriented Gradient (HOG) features descriptor were extracted from the multi-saliency map to create an image patch (a small image region of interest containing human candidates). For the classification of image patches, we used machine learning-based supervised learning models support-vector machine (SVM) algorithm to improve performance. As a result of the proposed algorithm, it showed low resource occupancy and achieved Average Precision of 92.3% and 96.12% when Intersection over Union were 50% and 45% respectively.https://ieeexplore.ieee.org/document/9426910/Top-viewhuman detectionimage subtractionsaliency mapclusteringclassification |
| spellingShingle | Seung Jun Lee Byeong Hak Kim Min Young Kim Multi-Saliency Map and Machine Learning Based Human Detection for the Embedded Top-View Imaging System IEEE Access Top-view human detection image subtraction saliency map clustering classification |
| title | Multi-Saliency Map and Machine Learning Based Human Detection for the Embedded Top-View Imaging System |
| title_full | Multi-Saliency Map and Machine Learning Based Human Detection for the Embedded Top-View Imaging System |
| title_fullStr | Multi-Saliency Map and Machine Learning Based Human Detection for the Embedded Top-View Imaging System |
| title_full_unstemmed | Multi-Saliency Map and Machine Learning Based Human Detection for the Embedded Top-View Imaging System |
| title_short | Multi-Saliency Map and Machine Learning Based Human Detection for the Embedded Top-View Imaging System |
| title_sort | multi saliency map and machine learning based human detection for the embedded top view imaging system |
| topic | Top-view human detection image subtraction saliency map clustering classification |
| url | https://ieeexplore.ieee.org/document/9426910/ |
| work_keys_str_mv | AT seungjunlee multisaliencymapandmachinelearningbasedhumandetectionfortheembeddedtopviewimagingsystem AT byeonghakkim multisaliencymapandmachinelearningbasedhumandetectionfortheembeddedtopviewimagingsystem AT minyoungkim multisaliencymapandmachinelearningbasedhumandetectionfortheembeddedtopviewimagingsystem |