| dc.contributor.author | XU Fang, MIAO Duoqian, ZHANG Hongyun | |
| dc.contributor.other | College of Electronic and Information Engineering,Tongji University,Shanghai 201804,China | |
| dc.date.accessioned | 2025-08-27T02:35:33Z | |
| dc.date.accessioned | 2025-10-08T08:22:49Z | |
| dc.date.available | 2025-10-08T08:22:49Z | |
| dc.date.issued | 01-11-2023 | |
| dc.identifier.uri | http://digilib.fisipol.ugm.ac.id/repo/handle/15717717/35669 | |
| dc.description.abstract | Different from other scale objects,small objects have the characteristics of carrying less semantic information and a small number of training samples.Therefore,the current object detection algorithm has the problem of low detection accuracy for small objects.Aiming at this problem,a Transformer object detection algorithm based on multi-granularity is proposed.Firstly,adopting the multi-granularity idea,a new Transformer serialization method is designed to predict the object position granularly from coarse to fine,thereby improving the object location effect of the model.Then,based on the three-way decision idea,fine-grained mining of small object samples and regular-scale object samples increases the number of small object samples and hardnegative samples.Finally,experimental results on the COCO dataset show that,the small object detection average accuracy(APs) of the algorithm reaches 31.5%,and the mean average accuracy(mAP) reaches 49.1%.Compared with the baseline model,the APs is improved by 1.4% and the mAP is improved by 2.2%.The algorithm effectively improves the detection effect of small objects and significantly improves the overall accuracy of object detection. | |
| dc.language.iso | ZH | |
| dc.publisher | Editorial office of Computer Science | |
| dc.subject.lcc | Computer software | |
| dc.title | Transformer Object Detection Algorithm Based on Multi-granularity | |
| dc.type | Article | |
| dc.description.keywords | small object detection|multi-granularity|three-way decision|transformer|deep learning | |
| dc.description.pages | 143-150 | |
| dc.description.doi | 10.11896/jsjkx.230600028 | |
| dc.title.journal | Jisuanji kexue | |
| dc.identifier.oai | 54af0038a82a4736b29977994f3c10c3 | |
| dc.journal.info | Volume 50, Issue 11 | |
