MFFAMM: A Small Object Detection with Multi-Scale Feature Fusion and Attention Mechanism Module
Aiming at the low detection accuracy and poor positioning for small objects of single-stage object detection algorithms, we improve the backbone network of SSD (Single Shot MultiBox Detector) and present an improved SSD model based on multi-scale feature fusion and attention mechanism module in this...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-09-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/18/8940 |
| _version_ | 1827663910139854848 |
|---|---|
| author | Zhong Qu Tongqiang Han Tuming Yi |
| author_facet | Zhong Qu Tongqiang Han Tuming Yi |
| author_sort | Zhong Qu |
| collection | DOAJ |
| description | Aiming at the low detection accuracy and poor positioning for small objects of single-stage object detection algorithms, we improve the backbone network of SSD (Single Shot MultiBox Detector) and present an improved SSD model based on multi-scale feature fusion and attention mechanism module in this paper. Firstly, we enhance the feature extraction ability of the shallow network through the feature fusion method that is beneficial to small object recognition. Secondly, the RFB (Receptive Field block) is used to expand the object’s receptive field and extract richer semantic information. After feature fusion, the attention mechanism module is added to enhance the feature information of important objects and suppress irrelevant other information. The experimental results show that our algorithm achieves 80.7% and 51.8% <i>mAP</i> on the PASCAL VOC 2007 classic dataset and MS COCO 2017 dataset, which are 3.2% and 10.6% higher than the original SSD algorithm. Our algorithm greatly improves the accuracy of object detection and meets the requirements of real-time. |
| first_indexed | 2024-03-10T00:51:52Z |
| format | Article |
| id | doaj.art-6d284107b2d348dcad478727d498c740 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T00:51:52Z |
| publishDate | 2022-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-6d284107b2d348dcad478727d498c7402023-11-23T14:50:30ZengMDPI AGApplied Sciences2076-34172022-09-011218894010.3390/app12188940MFFAMM: A Small Object Detection with Multi-Scale Feature Fusion and Attention Mechanism ModuleZhong Qu0Tongqiang Han1Tuming Yi2College of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, ChinaCollege of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, ChinaInstitute of Information Technology Southwest Computer Co., Ltd., Chongqing 400065, ChinaAiming at the low detection accuracy and poor positioning for small objects of single-stage object detection algorithms, we improve the backbone network of SSD (Single Shot MultiBox Detector) and present an improved SSD model based on multi-scale feature fusion and attention mechanism module in this paper. Firstly, we enhance the feature extraction ability of the shallow network through the feature fusion method that is beneficial to small object recognition. Secondly, the RFB (Receptive Field block) is used to expand the object’s receptive field and extract richer semantic information. After feature fusion, the attention mechanism module is added to enhance the feature information of important objects and suppress irrelevant other information. The experimental results show that our algorithm achieves 80.7% and 51.8% <i>mAP</i> on the PASCAL VOC 2007 classic dataset and MS COCO 2017 dataset, which are 3.2% and 10.6% higher than the original SSD algorithm. Our algorithm greatly improves the accuracy of object detection and meets the requirements of real-time.https://www.mdpi.com/2076-3417/12/18/8940small object detectionmulti-scalefeature fusionattention mechanismreceptive field |
| spellingShingle | Zhong Qu Tongqiang Han Tuming Yi MFFAMM: A Small Object Detection with Multi-Scale Feature Fusion and Attention Mechanism Module Applied Sciences small object detection multi-scale feature fusion attention mechanism receptive field |
| title | MFFAMM: A Small Object Detection with Multi-Scale Feature Fusion and Attention Mechanism Module |
| title_full | MFFAMM: A Small Object Detection with Multi-Scale Feature Fusion and Attention Mechanism Module |
| title_fullStr | MFFAMM: A Small Object Detection with Multi-Scale Feature Fusion and Attention Mechanism Module |
| title_full_unstemmed | MFFAMM: A Small Object Detection with Multi-Scale Feature Fusion and Attention Mechanism Module |
| title_short | MFFAMM: A Small Object Detection with Multi-Scale Feature Fusion and Attention Mechanism Module |
| title_sort | mffamm a small object detection with multi scale feature fusion and attention mechanism module |
| topic | small object detection multi-scale feature fusion attention mechanism receptive field |
| url | https://www.mdpi.com/2076-3417/12/18/8940 |
| work_keys_str_mv | AT zhongqu mffammasmallobjectdetectionwithmultiscalefeaturefusionandattentionmechanismmodule AT tongqianghan mffammasmallobjectdetectionwithmultiscalefeaturefusionandattentionmechanismmodule AT tumingyi mffammasmallobjectdetectionwithmultiscalefeaturefusionandattentionmechanismmodule |