吴杰;段锦;赫立群;李英超;朱文涛;

• 1:长春理工大学电子信息工程学院
• 2:长春理工大学空间光电技术研究所基础技术实验室

摘要(Abstract)：

为了解决传统特征提取方法在遥感图像中飞机检测准确率和实时性不足的问题,基于YOLOv3-tiny网络在准确率提升方面提出两点改进。改进点一:将网络提取图像特征点的方式改进为分组卷积,即将一幅图像分成三个通道进行卷积操作,配合通道特征变换以加强各通道之间的语义关联;改进点二:将网络深层特征增加一个尺度检测,并进行上采样与浅层特征图进行融合预测。在速度提升方面引入深度可分离卷积代替传统卷积以降低参数计算量,达到模型轻量化。根据改进后的网络提出一种包含33个卷积层的改进型卷积神经网络DS-YOLO,对改进前后网络分别在自制遥感飞机图像上进行训练,选出最优的权重,用来对目标小、曝光度高、背景干扰等低质量测试集进行测试分析。实验结果表明,改进后的算法在测试集上精准度提升了14.1%,召回率提升了16.8%,检测低质量遥感飞机图像效果更佳。

关键词(KeyWords)： 深度可分离卷积;分组卷积;DS-YOLO模型;通道特征变换;多尺度预测

Abstract:

Keywords:

基金项目(Foundation): 国家重点研发计划（2017YFC0803806）;; 国家自然科学基金重大项目（61890960)

作者(Author): 吴杰;段锦;赫立群;李英超;朱文涛;

Email:

DOI:

参考文献(References)：

• [1] LOWE D G.Object recognition from local scale-invariant features[C].Computer Vision,1999,2:1150-1157.
• [2] WANG X,HAN T X,YAN S.An HOG-LBP human detector with partial occlusion handling[C]//Proceedings of IEEE International Conference on Computer Vision,2009,1:32-39.
• [3] ANSARI N.Tracing multiple attackers with Deterministic Packet Marking(DPM)[C]//Proceedings of IEEE Pacific Rim Conference on Communications,Computers&Signal Processing,2003.
• [4] GIRSHICK R.Fast R-CNN[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2015:1440-1448.
• [5] HE K.GKIOXARI G.DOLLAR P,et al.Mask R-CNN[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2018.
• [6] HOWARD A G,ZHU M,CHEN B,et al.Mobilenets:Efficient convolutional neural networks for mobile vision applications[J].arXiv:1704.04861,2017.
• [7]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.
• [8] PARK U,TONG Y,JAIN A K.Age-invariant face recognition[J].IEEE Transactions on Software Engineering,2010,32(5):947-954.
• [9]陈辰，柴志雷，夏珺.基于Zynq7000 FPGA异构平台的YOLOv2加速器设计与实现[J].计算机科学与探索，2019,13(10):1677-1693.
• [10]张富凯，杨峰，李策.基于改进YOLOv3的快速车辆检测方法[J].计算机工程与应用，2019,55(2):12-20.
• [11] REDMON J,DIVVALA S,GIRSHICK R,et al.You only look once:Unified,real-time object detection[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2016:779-788.
• [12] SHIN H C,ROTH H R,GAO M,et al.Deep convolutional neural networks for computer-aided detection:CNN architectures,dataset characteristics and transfer learning[J].IEEE Transactions on Medical Imaging,2016,35(5):1285-1298.
• [13] IANDOLA F N,HAN S,MOSKEWICZ M W,et al.SqueezeNet:AlexNet-level accuracy with 50x fewer parameters and<0.5 MB model size[EB/OL].[2019-08-15].https://arxiv.org/abs/1602.07360.
• [14] REDMON J,FARHADI A.YOLOv3:An incremental improvement[J].arXiv:1804.02767,2018.
• [15] REDMON J,FARHADI A.YOLO9000:Better,faster,stronger[J].arXiv:1612.08242,2016.
• [16] YOON J,HWANG S J.Combined group and exclusive sparsity for deep neural networks[C]//Proceedings of the 34th International Conference on Machine Learning,2017:3958-3966.
• [17] DRIAS H,CHERIF N F,KECHID A.A hybrid clustering algorithm based on k-means and k-medoids[M].Berlin,Germany:Springer International Publising,2016:22-24.