改进轻量化的CenterNet的小目标检测

doi:10.3778/j.issn.1002-8331.2211-0298

摘要/Abstract

摘要： 小目标检测一直是目标检测中的难点，其特点为感受视野小，无法获取足够的语义特征，相比传统目标检测算法在工业应用中难以实现轻量化。为提升小目标检测精度、计算速度以及减少计算量和成本开销，提出一种基于CenterNet的轻量化的检测器SFPN-CenterNet。采用轻量级的深度可分离卷积网络来替代原始的普通卷积；简化FPN网络，减少下采样层数对特征进行提取融合，舍弃对于小目标检测无显著作用的高层特征；改进损失函数，对原来的公式以及超参数进行优化。在自制的数据集上进行对比实验。结果表明：利用深度可分离卷积作为提取特征的卷积块，可以使网络参数量减少到原来的1/480；改进损失函数降低了小目标的误检率和漏检率。相对于原始算法，改进算法的AP提升了3.5个百分点，检测速度提高了2.74 ms。

关键词: CenterNet, 小目标检测, 深度可分离, 轻量化

Abstract: Small object detection has always been a difficult point in object detection, which is characterized by small perception field and insufficient semantic features. Compared with traditional object detection algorithms, it is difficult to achieve lightweight in industrial applications. In order to improve the accuracy and speed of small object detection and reduce the amount of computation and cost, a lightweight detector SFPN-CenterNet based on CenterNet is proposed. The lightweight deep separable convolution network is used to replace the original ordinary convolution. It simplifies the FPN network, reduces the number of down-sampling layers for feature extraction and fusion, and discards high-level features that have no significant effect on small object detection. The loss function is improved to optimize the original formula and super parameters. Comparing experiments on self-made data sets, and the results show that the amount of network parameters is reduced to 1/480 by using deep separable convolution as the convolution block for feature extraction. The improved loss function reduces the false detection rate and missed detection rate of small targets. Comparing with the original algorithm, the AP of the improved algorithm increased by 3.5 percentage points and the detection speed increased by 2.74 ms.

Key words: CenterNet, small object detection, depth separable convolution, lightweight

王英博, 刘荣霞. 改进轻量化的CenterNet的小目标检测[J]. 计算机工程与应用, 2023, 59(17): 205-211.

WANG Yingbo, LIU Rongxia. Small Object Detection of Improved Lightweight CenterNet[J]. Computer Engineering and Applications, 2023, 59(17): 205-211.

参考文献

[1] FARUK M J H，SHAHRIAR H，VALERO M，et al.Malware detection and prevention using artificial intelligence techniques[C]//2021 IEEE International Conference on Big Data，2021：5369-5377.
[2] 方钧婷，谭晓阳.注意力级联网络的金属表面缺陷检测算法[J].计算机科学与探索，2021，15（7）：1245-1254.
FANG J T，TAN X Y.Defect detection of metal surface based on attention cascade R-CNN[J].Journal of Frontiers of Computer Science and Technology，2021，15（7）：1245-1254.
[3] 黄致君，桑庆兵.改进R-FCN的船舶识别方法[J].计算机科学与探索，2020，14（6）：1045-1053.
HUANG Z J，SANG Q B.Ship detection based on improved R-FCN[J].Journal of Frontiers of Computer Science and Technology，2020，14（6）：1045-1053.
[4] 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，2014：580-587.
[5] GIRSHICK R.Fast RCNN[C]//Proceedings of the IEEE International Conference on Computer Vision，2015：1440-1448.
[6] REN S Q，HE K M，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.
[7] HE K，GKIOXARI G，DOLLáR P，et al.Mask R-CNN[C]//Proceedings of the IEEE International Conference on Computer Vision，2017：2961-2969.
[8] REDMON J，DIVVALA S，GIRSHICK R，et al.You only look once：unified，real-time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2016：779-788.
[9] LIU W，ANGUELOV D，ERHAN D，et al.SSD：single shot multibox detector[C]//European Conference on Computer Vision.Cham：Springer，2016：21-37.
[10] LIN T Y，DOLLAR P，GIRSHICK R，et al.Feature pyramid networks for object detection[C]//IEEE Conference on Computer Vision and Pattern Recognition，Honolulu，USA，2017：936-944.
[11] IANDOLA F N，HAN S，MOSKEWICZ M W，et al.Sque-ezeNet：AlexNet-level accuracy with 50x fewer parameters and <0.5?MB model size[J].arXiv：1602.07360，2016.
[12] HOWARD A G，ZHU M，CHEN B，et al.MobileNets：efficient convolutional neural networks for mobile vision applications[J].arXiv：1704.04861，2017.
[13] SANDLER M，HOWARD A，ZHU M，et al.MobieNetv2：inverted residuals and linear bottlenecks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2018：4510-4520.
[14] 魏玮，杨茹，朱叶.改进CenterNet的遥感图像目标检测[J].计算机工程与应用，2021，57（6）：191-199.
WEI W，YANG R，ZHU Y.Target detection of improved CenterNet to remote sensing images[J].Computer Engineering and Applications，2021，57（6）：191-199.
[15] 刘鑫，黄进，杨涛，等.改进CenterNet的无人机小目标捕获检测方法[J].计算机工程与应用，2022，58（14）：96-104.
LIU X，HUANG J，YANG T，et al.Improved small object detection for UAV acquisition based on CenterNet[J].Computer Engineering and Applications，2022，58（14）：96-104.
[16] LAW H，DENG J.CornerNet：detecting objects as paired keypoints[C]//Proceedings of the European Conference on Computer Vision（ECCV），2018：734-750.
[17] DUAN K，BAI S，XIE L，et al.Centernet：keypoint triplets for object detection[C]//IEEE International Conference on Computer Vision，Seoul，Korea，2019：6569-6577.
[18] LIN T Y，GOYAL P，GIRSHICK R，et al.Focal loss for dense object detection[C]//Proceedings of the IEEE International Conference on Computer Vision，2017：2980-2988.