浏览全部资源
扫码关注微信
1.安徽理工大学 计算机科学与工程学院,安徽 淮南 232001
2.合肥综合性国家科学中心能源研究院(安徽省能源实验室),安徽 合肥 230031
3.安徽理工大学 机械工程学院,安徽 淮南232001
[ "苏树智(1987-),男,山东泰安人,博士,副教授,研究生导师,2017年于江南大学获得博士学位,主要从事模式识别及计算机视觉方面的研究。E-mail: sushuzhi@foxmail.com" ]
[ "陈润斌(1996-),男,安徽黄山人,硕士研究生,主要从事计算机视觉及深度学习方面的研究。E-mail: rbchen163@163.com" ]
收稿日期:2021-12-24,
修回日期:2022-02-19,
纸质出版日期:2022-07-10
移动端阅览
苏树智,陈润斌,朱彦敏等.重定位非极大值抑制算法[J].光学精密工程,2022,30(13):1620-1630.
SU Shuzhi,CHEN Runbin,ZHU Yanmin,et al.Relocation non-maximum suppression algorithm[J].Optics and Precision Engineering,2022,30(13):1620-1630.
苏树智,陈润斌,朱彦敏等.重定位非极大值抑制算法[J].光学精密工程,2022,30(13):1620-1630. DOI: 10.37188/OPE.20223013.1620.
SU Shuzhi,CHEN Runbin,ZHU Yanmin,et al.Relocation non-maximum suppression algorithm[J].Optics and Precision Engineering,2022,30(13):1620-1630. DOI: 10.37188/OPE.20223013.1620.
非极大值抑制(Non-Maximum Suppression, NMS)算法作为目标检测任务的后处理算法,其作用是从候选框集合中选出最优边界框并抑制其他候选框。传统NMS算法选取类别置信度最高的候选框作为最优边界框,忽略了类别置信度与定位精度之间的相关性,类别置信度高并不意味着该框的定位精度高。为了解决以上问题,提出一种新的重定位非极大值抑制(Relocation Non-Maximum Suppression, R-NMS)算法。选择类别置信度得分最高的候选框作为最优边界框,利用R-NMS算法提出的一种边界框距离度量方法替代交并比衡量边界框之间的距离。然后,获取最优边界框周围候选框的位置信息,利用位置信息对最优边界框执行重定位操作从而得到新的最优边界框。采用PASCAL VOC2012数据集进行测试,实验结果表明,与传统算法NMS和Soft-NMS相比,R-NMS算法在目标检测器YOLOv3上的mAP分别提高0.7%、0.5%,R-NMS算法在Faster-RCNN上的mAP达到80.83%。该算法能够有效提高目标检测器的检测精度。
Non-Maximum Suppression (NMS) is a post-processing algorithm used for object detection. It selects optimal bounding boxes from the bounding boxes set and suppresses other bounding boxes. NMS selects the bounding box with the highest score of classification confidence as the optimal bounding box. However, it ignores the correlation between localization accuracy and the classification confidence score. The classification confidence score cannot effectively represent the localization accuracy. This paper proposes a novel Relocation Non-Maximum Suppression (R-NMS) algorithm to solve the above-mentioned problem. First, the bounding box with the highest score of classification confidence in the bounding boxes set is selected as the optimal bounding box. Second, a new box distance measurement method is proposed based on R-NMS instead of using Intersection over Union (IoU) to measure the distance between the bounding boxes. Then, the location information of the bounding boxes around the optimal bounding box is obtained. Finally, the location information is used to relocate the optimal bounding box to obtain the new optimal bounding box. Compared with NMS and Soft-NMS, the mAP of R-NMS on YOLOv3 increased by 0.7 % and 0.5 %, respectively. The mAP of R-NMS on Faster-RCNN is 80.83 %, and the effectiveness of the proposed algorithm in the improvement of the mAP of various object detectors is confirmed.
WU X W , SAHOO D , HOI S C H . Recent advances in deep learning for object detection [J]. Neurocomputing , 2020 , 396 : 39 - 64 . doi: 10.1016/j.neucom.2020.01.085 http://dx.doi.org/10.1016/j.neucom.2020.01.085
SUN Y N , XUE B , ZHANG M J , et al . Automatically designing CNN architectures using the genetic algorithm for image classification [J]. IEEE Transactions on Cybernetics , 2020 , 50 ( 9 ): 3840 - 3854 . doi: 10.1109/tcyb.2020.2983860 http://dx.doi.org/10.1109/tcyb.2020.2983860
LIU Y , WU Y H , WEN P S , et al . Leveraging instance-, image- and dataset-level information for weakly supervised instance segmentation [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2022 , 44 ( 3 ): 1415 - 1428 . doi: 10.1109/tpami.2020.3023152 http://dx.doi.org/10.1109/tpami.2020.3023152
CHENG Q M , ZHANG Q , FU P , et al . A survey and analysis on automatic image annotation [J]. Pattern Recognition , 2018 , 79 : 242 - 259 . doi: 10.1016/j.patcog.2018.02.017 http://dx.doi.org/10.1016/j.patcog.2018.02.017
CIAPARRONE G , LUQUE SÁNCHEZ F , TABIK S , et al . Deep learning in video multi-object tracking: a survey [J]. Neurocomputing , 2020 , 381 : 61 - 88 . doi: 10.1016/j.neucom.2019.11.023 http://dx.doi.org/10.1016/j.neucom.2019.11.023
GIRSHICK R , DONAHUE J , DARRELL T , et al . Rich feature hierarchies for accurate object detection and semantic segmentation [C]. 2014 IEEE Conference on Computer Vision and Pattern Recognition . 2328,2014 , Columbus , OH , USA . IEEE , 2014 : 580 - 587 . doi: 10.1109/cvpr.2014.81 http://dx.doi.org/10.1109/cvpr.2014.81
KRIZHEVSKY A , SUTSKEVER I , HINTON G E . ImageNet classification with deep convolutional neural networks [J]. Communications of the ACM , 2017 , 60 ( 6 ): 84 - 90 . doi: 10.1145/3065386 http://dx.doi.org/10.1145/3065386
WAN S H , GOUDOS S . Faster R-CNN for multi-class fruit detection using a robotic vision system [J]. Computer Networks , 2020 , 168 : 107036 . doi: 10.1016/j.comnet.2019.107036 http://dx.doi.org/10.1016/j.comnet.2019.107036
WANG N , GAO Y , CHEN H , et al . NAS-FCOS: efficient search for object detection architectures [J]. International Journal of Computer Vision , 2021 , 129 ( 12 ): 3299 - 3312 . doi: 10.1007/s11263-021-01523-2 http://dx.doi.org/10.1007/s11263-021-01523-2
王建林 , 付雪松 , 黄展超 , 等 . 改进YOLOv2卷积神经网络的多类型合作目标检测 [J]. 光学 精密工程 , 2020 , 28 ( 1 ): 251 - 260 . doi: 10.3788/ope.20202801.0251 http://dx.doi.org/10.3788/ope.20202801.0251
WANG J L , FU X S , HUANG ZH CH , et al . Multi-type cooperative targets detection using improved YOLOv2 convolutional neural network [J]. Optics and Precision Engineering , 2020 , 28 ( 1 ): 251 - 260 . (in Chinese) . doi: 10.3788/ope.20202801.0251 http://dx.doi.org/10.3788/ope.20202801.0251
鞠默然 , 罗海波 , 刘广琦 , 等 . 采用空间注意力机制的红外弱小目标检测网络 [J]. 光学 精密工程 , 2021 , 29 ( 4 ): 843 - 853 . doi: 10.37188/OPE.20212904.0843 http://dx.doi.org/10.37188/OPE.20212904.0843
JU M R , LUO H B , LIU G Q , et al . Infrared dim and small target detection network based on spatial attention mechanism [J]. Optics and Precision Engineering , 2021 , 29 ( 4 ): 843 - 853 . (in Chinese) . doi: 10.37188/OPE.20212904.0843 http://dx.doi.org/10.37188/OPE.20212904.0843
马立 , 巩笑天 , 欧阳航空 . Tiny YOLOV3目标检测改进 [J]. 光学 精密工程 , 2020 , 28 ( 4 ): 988 - 995 .
MA L , GONG X T , OUYANG H K . Improvement of Tiny YOLOV 3 target detection [J]. Optics and Precision Engineering , 2020 , 28 ( 4 ): 988 - 995 . (in Chinese)
NEUBECK A , GOOL LVAN . Efficient non-maximum suppression [C]. 18th International Conference on Pattern Recognition (ICPR' 06 ). 2024,2006 , Hong Kong, China. IEEE , 2006: 850 - 855 .
QIU S H , WEN G J , DENG Z P , et al . Accurate non-maximum suppression for object detection in high-resolution remote sensing images [J]. Remote Sensing Letters , 2018 , 9 ( 3 ): 237 - 246 . doi: 10.1080/2150704x.2017.1415473 http://dx.doi.org/10.1080/2150704x.2017.1415473
HE Y H , ZHU C C , WANG J R , et al . Bounding box regression with uncertainty for accurate object detection [C]. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 1520,2019 , Long Beach, CA, USA. IEEE , 2019 : 2883 - 2892 . doi: 10.1109/cvpr.2019.00300 http://dx.doi.org/10.1109/cvpr.2019.00300
MA W C , LI K D , WANG G H . Location-aware box reasoning for anchor-based single-shot object detection [J]. IEEE Access , 8 : 129300 - 129309 .
Jiang B , Luo R , Mao J , et al . Acquisition of Localization Confidence for Accurate Object Detection [C]. Proceedings of the European Conference on Computer Vision. 814,2018 , Munich, Germany. Springer , 2018 : 784 - 799 . doi: 10.1007/978-3-030-01264-9_48 http://dx.doi.org/10.1007/978-3-030-01264-9_48
ZENG J X , XIONG J L , FU X , et al . ReFPN-FCOS: one-stage object detection for feature learning and accurate localization [J]. IEEE Access , 8 : 225052 - 225063 .
LIU S T , HUANG D , WANG Y H . Adaptive NMS: refining pedestrian detection in a crowd [C]. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 1520,2019 , Long Beach, CA, USA. IEEE , 2019 : 6452 - 6461 . doi: 10.1109/cvpr.2019.00662 http://dx.doi.org/10.1109/cvpr.2019.00662
ZHENG Z H , WANG P , LIU W , et al . Distance-IoU loss: faster and better learning for bounding box regression [J]. Proceedings of the AAAI Conference on Artificial Intelligence , 2020 , 34 ( 7 ): 12993 - 13000 . doi: 10.1609/aaai.v34i07.6999 http://dx.doi.org/10.1609/aaai.v34i07.6999
BODLA N , SINGH B , CHELLAPPA R , et al . Soft-NMS-improving object detection with one line of code [C]. 2017 IEEE International Conference on Computer Vision . 2229,2017 , Venice, Italy . IEEE , 2017 : 5562 - 5570 . doi: 10.1109/iccv.2017.593 http://dx.doi.org/10.1109/iccv.2017.593
0
浏览量
885
下载量
1
CSCD
关联资源
相关文章
相关作者
相关机构