[1]赵文清,杨盼盼.双向特征融合与注意力机制结合的目标检测[J].智能系统学报,2021,16(6):1098-1105.[doi:10.11992/tis.202012029]
ZHAO Wenqing,YANG Panpan.Target detection based on bidirectional feature fusion and an attention mechanism[J].CAAI Transactions on Intelligent Systems,2021,16(6):1098-1105.[doi:10.11992/tis.202012029]
点击复制
ZHAO Wenqing,YANG Panpan.Target detection based on bidirectional feature fusion and an attention mechanism[J].CAAI Transactions on Intelligent Systems,2021,16(6):1098-1105.[doi:10.11992/tis.202012029]
双向特征融合与注意力机制结合的目标检测
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
16
期数:
2021年第6期
页码:
1098-1105
栏目:
学术论文—知识工程
出版日期:
2021-11-05
- Title:
- Target detection based on bidirectional feature fusion and an attention mechanism
- Keywords:
- feature pyramid; bidirectional fusion; feature extraction; SeNet attention mechanism; sample; semantic information; target detection; deep learning
- 分类号:
- TP391
- 摘要:
- 目标检测使用特征金字塔检测不同尺度的物体时，忽略了高层信息和低层信息之间的关系，导致检测效果差；此外，针对某些尺度的目标，检测中容易出现漏检。本文提出双向特征融合与注意力机制结合的方法进行目标检测。首先，对SSD(single shot multibox detector)模型深层特征层与浅层特征层进行特征融合，然后将得到的特征与深层特征层进行融合。其次，在双向融合中加入了通道注意力机制，增强了语义信息。最后，提出了一种改进的正负样本判定策略，降低目标的漏检率。将本文提出的算法与当前主流算法在VOC数据集上进行了比较，结果表明，本文提出的算法在对目标进行检测时，目标平均准确率有较大提高。
- Abstract:
- When using a feature pyramid to detect objects of different dimensions, the relationship between high- and low-level information is ignored, resulting in a poor detection effect; in addition, for targets of a certain scale, detection is easily missed. In this paper, a method combining bidirectional feature fusion and an attention mechanism is proposed for target detection. First, the deep and shallow feature layers of the single-shot multibox detector (SSD) model are fused, then the obtained features are fused with the deep feature layer. Second, the channel attention mechanism is added to the two-way fusion to enhance semantic information. Finally, an improved positive and negative sample decision strategy is proposed to reduce the target misdetection rate. The algorithm proposed in this paper is compared with the current mainstream algorithms in the VOC dataset. The results show that the average accuracy of the proposed algorithm is greatly improved when detecting targets.
- 参考文献/References:
-
[1] GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Columbus, USA, 2014: 580-587.
[2] UIJLINGS J R R, VAN DE SANDE K E A, GEVERS T, et al. Selective search for object recognition[J]. International journal of computer vision, 2013, 104(2): 154-171.
[3] HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[J]. IEEE transactions on pattern analysis and machine intelligence, 2015, 37(9): 1904-1916.
[4] GIRSHICK R. Fast R-CNN[C]//Proceedings of the IEEE International Conference on Computer Vision. Santiago, Chile, 2015: 1440-1448.
[5] REN Shaoqing, HE Kaiming, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE transactions on pattern analysis and machine intelligence, 2017, 39(6): 1137-1149.
[6] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA, 2016: 779-788.
[7] REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA, 2017: 6517-6525.
[8] LIU Wei, ANGUELOV D, ERHAN D, et al. SSD: single shot multibox detector[C]//Proceedings of the 14th European Conference on Computer Vision. Amsterdam, The Netherland, 2016: 21-37.
[9] FU Chengyang, LIU Wei, RANGA A, et al. DSSD: deconvolutional single shot detector[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Hawaii, USA, 2017: 2881-2890.
[10] SINGH B, DAVIS L S. An analysis of scale invariance in object detection-SNIP[C]//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA, 2018: 3578-3587.
[11] 温静, 李雨萌. 基于多尺度反卷积深度学习的显著性检测[J]. 计算机科学, 2020, 47(11): 179-185
WEN Jing, LI Yumeng. Salient object detection based on multi-scale deconvolution deep learning[J]. Computer science, 2020, 47(11): 179-185
[12] LI Jianan, LIANG Xiaodan, WEI Yunchao, et al. Perceptual generative adversarial networks for small object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA, 2017: 1951-1959.
[13] GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial nets[C]//Proceedings of the 27th International Conference on Neural Information Processing Systems. Montreal, Canada, 2014: 2672-2680.
[14] LIN T Y, DOLLáR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA, 2017: 936-944.
[15] 刘涛, 汪西莉. 采用卷积核金字塔和空洞卷积的单阶段目标检测[J]. 中国图象图形学报, 2020, 25(1): 102-112
LIU Tao, WANG Xili. Single-stage object detection using filter pyramid and atrous convolution[J]. Journal of image and graphics, 2020, 25(1): 102-112
[16] 陈景明, 金杰, 王伟锋. 基于特征金字塔网络的改进算法[J]. 激光与光电子学进展, 2019, 56(21): 211505
CHEN Jingming, JIN Jie, WANG Weifeng. Improved algorithm based on feature pyramid networks[J]. Laser & optoelectronics progress, 2019, 56(21): 211505
[17] 张涛, 张乐. 一种基于多尺度特征融合的目标检测算法[J]. 激光与光电子学进展, 2021, 58(2): 0215003
ZHANG Tao, ZHANG Le. Multiscale feature fusion-based object detection algorithm[J]. Laser & optoelectronics progress, 2021, 58(2): 0215003
[18] 和超, 张印辉, 何自芬. 多尺度特征融合工件目标语义分割[J]. 中国图象图形学报, 2020, 25(3): 476-485
HE Chao, ZHANG Yinhui, HE Zifen. Semantic segmentation of workpiece target based on multiscale feature fusion[J]. Journal of image and graphics, 2020, 25(3): 476-485
[19] 鞠默然, 罗江宁, 王仲博, 等. 融合注意力机制的多尺度目标检测算法[J]. 光学学报, 2020, 40(13): 1315002
JU Moran, LUO Jiangning, WANG Zhongbo, et al. Multi-scale target detection algorithm based on attention mechanism[J]. Acta optica sinica, 2020, 40(13): 1315002
[20] 张筱晗, 姚力波, 吕亚飞, 等. 双向特征融合的数据自适应SAR图像舰船目标检测模型[J]. 中国图象图形学报, 2020, 25(9): 1943-1952
ZHANG Xiaohan, YAO Libo, LV Yafei, et al. Data-adaptive single-shot ship detector with a bidirectional feature fusion module for SAR images[J]. Journal of image and graphics, 2020, 25(9): 1943-1952
[21] 高杨, 肖迪. 基于多层特征融合的小目标检测算法[J]. 计算机工程与设计, 2020, 41(7): 1905-1909
GAO Yang, XIAO Di. Small object detection algorithm based on multi-feature fusion[J]. Computer engineering and design, 2020, 41(7): 1905-1909
[22] 杨锐, 张宝华, 张艳月, 等. 基于深度特征自适应融合的运动目标跟踪算法[J]. 激光与光电子学进展, 2020, 57(18): 181501
YANG Rui, ZHANG Baohua, ZHANG Yanyue, et al. Moving object tracking algorithm based on depth feature adaptive fusion[J]. Laser & optoelectronics progress, 2020, 57(18): 181501
[23] 赵文清, 程幸福, 赵振兵, 等. 注意力机制和Faster RCNN相结合的绝缘子识别[J]. 智能系统学报, 2020, 15(1): 92-98
ZHAO Wenqing, CHENG Xingfu, ZHAO Zhenbing, et al. Insulator recognition based on attention mechanism and faster RCNN[J]. CAAI transactions on intelligent systems, 2020, 15(1): 92-98
[24] 徐诚极, 王晓峰, 杨亚东. Attention-YOLO: 引入注意力机制的YOLO检测算法[J]. 计算机工程与应用, 2019, 55(6): 13-23
XU Chengji, WANG Xiaofeng, YANG yadong. Attention-YOLO: YOLO detection algorithm that introduces attention mechanism[J]. Computer engineering and applications, 2019, 55(6): 13-23
[25] 单义, 杨金福, 武随烁, 等. 基于跳跃连接金字塔模型的小目标检测[J]. 智能系统学报, 2019, 14(6): 1144-1151
SHAN Yi, YANG Jinfu, WU Suishuo, et al. Skip feature pyramid network with a global receptive field for small object detection[J]. CAAI transactions on intelligent systems, 2019, 14(6): 1144-1151
- 相似文献/References:
-
[1]赵文清,孔子旭,赵振兵.隔级融合特征金字塔与CornerNet相结合的小目标检测[J].智能系统学报,2021,16(1):108.[doi:10.11992/tis.202004033]
ZHAO Wenqing,KONG Zixu,ZHAO Zhenbing.Small target detection based on a combination of feature pyramid and CornerNet[J].CAAI Transactions on Intelligent Systems,2021,16():108.[doi:10.11992/tis.202004033]
备注/Memo
收稿日期:2020-12-17。
基金项目:河北省自然科学基金项目(F2021502013)；中央高校基本科研业务费面上项目(2020MS153，2021PT018)
作者简介:赵文清，教授，博士，主要研究方向为人工智能与图像处理。获河北省科技进步二等奖、三等奖各1项。发表学术论文50余篇;杨盼盼，硕士研究生，主要研究方向为深度学习和目标检测
通讯作者:赵文清.E-mail:zhaowenqing@ncepu.edu.cn
更新日期/Last Update:
2021-12-25