SSD Object Detection Algorithm with Cross-layer Fusion and Receptive Field Amplification
In view of the lack of information interaction between different layers of single shot multibox detector(SSD) and the limitation of the model's receptive field,an improved SSD object detection algorithm,named ESSD(enhanced SSD),is proposed to improve the accuracy of object detection.First of al...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial office of Computer Science
2023-03-01
|
| Series: | Jisuanji kexue |
| Subjects: | |
| Online Access: | https://www.jsjkx.com/fileup/1002-137X/PDF/1002-137X-2023-50-3-231.pdf |
| _version_ | 1797845097050537984 |
|---|---|
| author | ZHANG Weiliang, CHEN Xiuhong |
| author_facet | ZHANG Weiliang, CHEN Xiuhong |
| author_sort | ZHANG Weiliang, CHEN Xiuhong |
| collection | DOAJ |
| description | In view of the lack of information interaction between different layers of single shot multibox detector(SSD) and the limitation of the model's receptive field,an improved SSD object detection algorithm,named ESSD(enhanced SSD),is proposed to improve the accuracy of object detection.First of all,using the original multi-scale feature map in the SSD model and using the idea of feature pyramid networks(FPN),a cross-layer information interaction module is designed,which enhances the semantic information capabilities of different layers and reduces the information difference of different layers.Then,in order to improve the receptive field and multi-scale detection capabilities of the model,a receptive field amplification module is designed.Finally,the batch normalization layer is used to reduce the training time and improve the convergence speed of the model.In order to evaluate the effectiveness of ESSD,experiments are conducted on the PASCAL VOC2007 and PASCAL VOC2012 test sets.Experimental results show that on the PASCAL VOC2007 data set,its <i>mAP </i>is 82.1% and the detection speed is 15.7FPS.Compared with the original SSD512,its <i>mAP </i>increases by 2.3%;on the PASCAL VOC2012 test set,its <i>mAP </i>reaches 80.6%,which is also 2.1% higher than SSD512.Experiments have proved that the ESSD detector can still meet the real-time performance under the condition of high detection accuracy. |
| first_indexed | 2024-04-09T17:33:02Z |
| format | Article |
| id | doaj.art-e6237fd3a89e46c8955e66f4dfed31aa |
| institution | Directory Open Access Journal |
| issn | 1002-137X |
| language | zho |
| last_indexed | 2024-04-09T17:33:02Z |
| publishDate | 2023-03-01 |
| publisher | Editorial office of Computer Science |
| record_format | Article |
| series | Jisuanji kexue |
| spelling | doaj.art-e6237fd3a89e46c8955e66f4dfed31aa2023-04-18T02:33:25ZzhoEditorial office of Computer ScienceJisuanji kexue1002-137X2023-03-0150323123710.11896/jsjkx.211100281SSD Object Detection Algorithm with Cross-layer Fusion and Receptive Field AmplificationZHANG Weiliang, CHEN Xiuhong0School of Artificial Intelligence and Computer Science,Jiangnan University,Wuxi,Jiangsu 214122,China;Jiangsu Key Laboratory of Media Design and Software Technology,Wuxi,Jiangsu 214122,ChinaIn view of the lack of information interaction between different layers of single shot multibox detector(SSD) and the limitation of the model's receptive field,an improved SSD object detection algorithm,named ESSD(enhanced SSD),is proposed to improve the accuracy of object detection.First of all,using the original multi-scale feature map in the SSD model and using the idea of feature pyramid networks(FPN),a cross-layer information interaction module is designed,which enhances the semantic information capabilities of different layers and reduces the information difference of different layers.Then,in order to improve the receptive field and multi-scale detection capabilities of the model,a receptive field amplification module is designed.Finally,the batch normalization layer is used to reduce the training time and improve the convergence speed of the model.In order to evaluate the effectiveness of ESSD,experiments are conducted on the PASCAL VOC2007 and PASCAL VOC2012 test sets.Experimental results show that on the PASCAL VOC2007 data set,its <i>mAP </i>is 82.1% and the detection speed is 15.7FPS.Compared with the original SSD512,its <i>mAP </i>increases by 2.3%;on the PASCAL VOC2012 test set,its <i>mAP </i>reaches 80.6%,which is also 2.1% higher than SSD512.Experiments have proved that the ESSD detector can still meet the real-time performance under the condition of high detection accuracy.https://www.jsjkx.com/fileup/1002-137X/PDF/1002-137X-2023-50-3-231.pdfobject detection|information fusion|receptive field|multi-scale|ssd |
| spellingShingle | ZHANG Weiliang, CHEN Xiuhong SSD Object Detection Algorithm with Cross-layer Fusion and Receptive Field Amplification Jisuanji kexue object detection|information fusion|receptive field|multi-scale|ssd |
| title | SSD Object Detection Algorithm with Cross-layer Fusion and Receptive Field Amplification |
| title_full | SSD Object Detection Algorithm with Cross-layer Fusion and Receptive Field Amplification |
| title_fullStr | SSD Object Detection Algorithm with Cross-layer Fusion and Receptive Field Amplification |
| title_full_unstemmed | SSD Object Detection Algorithm with Cross-layer Fusion and Receptive Field Amplification |
| title_short | SSD Object Detection Algorithm with Cross-layer Fusion and Receptive Field Amplification |
| title_sort | ssd object detection algorithm with cross layer fusion and receptive field amplification |
| topic | object detection|information fusion|receptive field|multi-scale|ssd |
| url | https://www.jsjkx.com/fileup/1002-137X/PDF/1002-137X-2023-50-3-231.pdf |
| work_keys_str_mv | AT zhangweiliangchenxiuhong ssdobjectdetectionalgorithmwithcrosslayerfusionandreceptivefieldamplification |