The Lightweight Anchor Dynamic Assignment Algorithm for Object Detection
Smart security based on object detection is one of the important applications of edge computing in IoT. Anchors in object detection refer to points on the feature map, which can be used to generate anchor boxes and serve as training samples. Current object detection models do not consider the aspect...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-07-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/14/6306 |
| _version_ | 1797587521897496576 |
|---|---|
| author | Ping Han Xujun Zhuang Huahong Zuo Ping Lou Xiao Chen |
| author_facet | Ping Han Xujun Zhuang Huahong Zuo Ping Lou Xiao Chen |
| author_sort | Ping Han |
| collection | DOAJ |
| description | Smart security based on object detection is one of the important applications of edge computing in IoT. Anchors in object detection refer to points on the feature map, which can be used to generate anchor boxes and serve as training samples. Current object detection models do not consider the aspect ratio of the ground-truth boxes in anchor assignment and are not well-adapted to objects with very different shapes. Therefore, this paper proposes the Lightweight Anchor Dynamic Assignment algorithm (LADA) for object detection. LADA does not change the structure of the original detection model; first, it selects an equal proportional center region based on the aspect ratio of the ground-truth box, then calculates the combined loss of anchors, and finally divides the positive and negative samples more efficiently by dynamic loss threshold without additional models. The algorithm solves the problems of poor adaptability and difficulty in the selection of the best positive samples based on IoU assignment, and the sample assignment for eccentric objects and objects with different aspect ratios was more reasonable. Compared with existing sample assignment algorithms, the LADA algorithm outperforms the MS COCO dataset by 1.66% over the AP of the baseline FCOS, and 0.76% and 0.24% over the AP of the ATSS algorithm and the PAA algorithm, respectively, with the same model structure, which demonstrates the effectiveness of the LADA algorithm. |
| first_indexed | 2024-03-11T00:41:05Z |
| format | Article |
| id | doaj.art-74ed2dd8432d49efafdcc316a3c713f2 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T00:41:05Z |
| publishDate | 2023-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-74ed2dd8432d49efafdcc316a3c713f22023-11-18T21:15:53ZengMDPI AGSensors1424-82202023-07-012314630610.3390/s23146306The Lightweight Anchor Dynamic Assignment Algorithm for Object DetectionPing Han0Xujun Zhuang1Huahong Zuo2Ping Lou3Xiao Chen4School of Information Engineering, Wuhan University of Technology, Wuhan 430070, ChinaSchool of Information Engineering, Wuhan University of Technology, Wuhan 430070, ChinaWuhan Chuyan Information Technology Co., Ltd., Wuhan 430030, ChinaSchool of Information Engineering, Wuhan University of Technology, Wuhan 430070, ChinaSchool of Information Engineering, Wuhan University of Technology, Wuhan 430070, ChinaSmart security based on object detection is one of the important applications of edge computing in IoT. Anchors in object detection refer to points on the feature map, which can be used to generate anchor boxes and serve as training samples. Current object detection models do not consider the aspect ratio of the ground-truth boxes in anchor assignment and are not well-adapted to objects with very different shapes. Therefore, this paper proposes the Lightweight Anchor Dynamic Assignment algorithm (LADA) for object detection. LADA does not change the structure of the original detection model; first, it selects an equal proportional center region based on the aspect ratio of the ground-truth box, then calculates the combined loss of anchors, and finally divides the positive and negative samples more efficiently by dynamic loss threshold without additional models. The algorithm solves the problems of poor adaptability and difficulty in the selection of the best positive samples based on IoU assignment, and the sample assignment for eccentric objects and objects with different aspect ratios was more reasonable. Compared with existing sample assignment algorithms, the LADA algorithm outperforms the MS COCO dataset by 1.66% over the AP of the baseline FCOS, and 0.76% and 0.24% over the AP of the ATSS algorithm and the PAA algorithm, respectively, with the same model structure, which demonstrates the effectiveness of the LADA algorithm.https://www.mdpi.com/1424-8220/23/14/6306object detectionpositive and negative samplesanchor assignmentaspect ratioloss awareself adaptive |
| spellingShingle | Ping Han Xujun Zhuang Huahong Zuo Ping Lou Xiao Chen The Lightweight Anchor Dynamic Assignment Algorithm for Object Detection Sensors object detection positive and negative samples anchor assignment aspect ratio loss aware self adaptive |
| title | The Lightweight Anchor Dynamic Assignment Algorithm for Object Detection |
| title_full | The Lightweight Anchor Dynamic Assignment Algorithm for Object Detection |
| title_fullStr | The Lightweight Anchor Dynamic Assignment Algorithm for Object Detection |
| title_full_unstemmed | The Lightweight Anchor Dynamic Assignment Algorithm for Object Detection |
| title_short | The Lightweight Anchor Dynamic Assignment Algorithm for Object Detection |
| title_sort | lightweight anchor dynamic assignment algorithm for object detection |
| topic | object detection positive and negative samples anchor assignment aspect ratio loss aware self adaptive |
| url | https://www.mdpi.com/1424-8220/23/14/6306 |
| work_keys_str_mv | AT pinghan thelightweightanchordynamicassignmentalgorithmforobjectdetection AT xujunzhuang thelightweightanchordynamicassignmentalgorithmforobjectdetection AT huahongzuo thelightweightanchordynamicassignmentalgorithmforobjectdetection AT pinglou thelightweightanchordynamicassignmentalgorithmforobjectdetection AT xiaochen thelightweightanchordynamicassignmentalgorithmforobjectdetection AT pinghan lightweightanchordynamicassignmentalgorithmforobjectdetection AT xujunzhuang lightweightanchordynamicassignmentalgorithmforobjectdetection AT huahongzuo lightweightanchordynamicassignmentalgorithmforobjectdetection AT pinglou lightweightanchordynamicassignmentalgorithmforobjectdetection AT xiaochen lightweightanchordynamicassignmentalgorithmforobjectdetection |