SSD Object Detection Model Based on Multi-Frequency Feature Theory
In order to further improve the accuracy and real-time performance of the traditional Single Shot Multibox Detector (SSD) object detection model, an improved SSD multi-object detection model is proposed. Firstly, aiming at the defect of weak correlation between prediction object score and positionin...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9078755/ |
| _version_ | 1818857014821912576 |
|---|---|
| author | Jinling Li Qingshan Hou Jinsheng Xing Jianguo Ju |
| author_facet | Jinling Li Qingshan Hou Jinsheng Xing Jianguo Ju |
| author_sort | Jinling Li |
| collection | DOAJ |
| description | In order to further improve the accuracy and real-time performance of the traditional Single Shot Multibox Detector (SSD) object detection model, an improved SSD multi-object detection model is proposed. Firstly, aiming at the defect of weak correlation between prediction object score and positioning accuracy in the traditional SSD model, the improved model enhanced the correlation between the two by adding Intersection Over Union(IoU) prediction loss branch. Secondly, in order to reduce the spatial redundancy of traditional SSD model, a multi-frequency feature component convolution module is designed, which greatly reduces the calculation overhead and hardware overhead of the traditional model. Finally, in order to accelerate the convergence speed of the improved model, the Adaptive and Momental Bound (AdaMod) optimizer is introduced to modify the adaptive learning rate of the improved model which is too large in the training process. Experimental results show that the improved model has stronger detection capabilities, better overall detection results, and improved detection accuracy and real-time detection. |
| first_indexed | 2024-12-19T08:33:40Z |
| format | Article |
| id | doaj.art-4f266cc97aa64a57b8993f2ff9278e2b |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-19T08:33:40Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-4f266cc97aa64a57b8993f2ff9278e2b2022-12-21T20:29:07ZengIEEEIEEE Access2169-35362020-01-018822948230510.1109/ACCESS.2020.29904779078755SSD Object Detection Model Based on Multi-Frequency Feature TheoryJinling Li0Qingshan Hou1https://orcid.org/0000-0002-3839-1399Jinsheng Xing2Jianguo Ju3https://orcid.org/0000-0002-9682-2679School of Economics and Management, Shanxi Normal University, Linfen, ChinaSchool of Mathematics and Computer Science, Shanxi Normal University, Linfen, ChinaSchool of Mathematics and Computer Science, Shanxi Normal University, Linfen, ChinaSchool of Information Science and Technology, Northwestern University, Xi’an, ChinaIn order to further improve the accuracy and real-time performance of the traditional Single Shot Multibox Detector (SSD) object detection model, an improved SSD multi-object detection model is proposed. Firstly, aiming at the defect of weak correlation between prediction object score and positioning accuracy in the traditional SSD model, the improved model enhanced the correlation between the two by adding Intersection Over Union(IoU) prediction loss branch. Secondly, in order to reduce the spatial redundancy of traditional SSD model, a multi-frequency feature component convolution module is designed, which greatly reduces the calculation overhead and hardware overhead of the traditional model. Finally, in order to accelerate the convergence speed of the improved model, the Adaptive and Momental Bound (AdaMod) optimizer is introduced to modify the adaptive learning rate of the improved model which is too large in the training process. Experimental results show that the improved model has stronger detection capabilities, better overall detection results, and improved detection accuracy and real-time detection.https://ieeexplore.ieee.org/document/9078755/Object detectionIoU predicting lossSSDmulti-frequency feature component convolutionAdaMod optimizer |
| spellingShingle | Jinling Li Qingshan Hou Jinsheng Xing Jianguo Ju SSD Object Detection Model Based on Multi-Frequency Feature Theory IEEE Access Object detection IoU predicting loss SSD multi-frequency feature component convolution AdaMod optimizer |
| title | SSD Object Detection Model Based on Multi-Frequency Feature Theory |
| title_full | SSD Object Detection Model Based on Multi-Frequency Feature Theory |
| title_fullStr | SSD Object Detection Model Based on Multi-Frequency Feature Theory |
| title_full_unstemmed | SSD Object Detection Model Based on Multi-Frequency Feature Theory |
| title_short | SSD Object Detection Model Based on Multi-Frequency Feature Theory |
| title_sort | ssd object detection model based on multi frequency feature theory |
| topic | Object detection IoU predicting loss SSD multi-frequency feature component convolution AdaMod optimizer |
| url | https://ieeexplore.ieee.org/document/9078755/ |
| work_keys_str_mv | AT jinlingli ssdobjectdetectionmodelbasedonmultifrequencyfeaturetheory AT qingshanhou ssdobjectdetectionmodelbasedonmultifrequencyfeaturetheory AT jinshengxing ssdobjectdetectionmodelbasedonmultifrequencyfeaturetheory AT jianguoju ssdobjectdetectionmodelbasedonmultifrequencyfeaturetheory |