CenterNet-based遥感图像目标检测方法研究

黄佳琦; 范军芳; 李蓓蓓

doi:10.15892/j.cnki.djzdxb.2023.01.004

弹箭与制导学报 >

2023 , Vol. 43 >Issue 1: 24 - 31

DOI: https://doi.org/10.15892/j.cnki.djzdxb.2023.01.004

CenterNet-based遥感图像目标检测方法研究

黄佳琦 ¹^,² ,
范军芳 ¹^,² ,
李蓓蓓 ³

展开

¹ 北京信息科技大学高动态导航技术北京市重点实验室,北京 100192
² 北京信息科技大学自动化学院,北京 100192
³ 32381部队,北京 100072

李蓓蓓(1986—),女,助理工程师,硕士,研究方向:自动化。

黄佳琦(1997—),女,硕士研究生,研究方向:智能图像识别。

收稿日期: 2022-09-11

网络出版日期: 2025-02-01

基金资助

国家重点研发计划(2020YFC1511705)

国家自然科学基金(61801032)

收起

CenterNet-based Target Detection Method for Remote Sensing Images

HUANG Jiaqi ¹^,² ,
FAN Junfang ¹^,² ,
LI Beibei ³

Expand

¹ Beijing Key Laboratory of High Dynamic Navigation Technology, Beijing Information Science and Technology University, Beijing 100192, China
² School of Automation, Beijing Information Science and Technology University, Beijing 100192, China
³ No.32381 Unit, Beijing 100072, China

Received date: 2022-09-11

Online published: 2025-02-01

Fold

摘要

快速准确地识别遥感图像中小而密集、成像质量不高的目标成为基于深度学习技术的遥感目标检测研究的重点和难点。为解决这类问题,提出结合注意力机制的CenterNet-based目标检测方法,该方法利用ResNet-50网络进行基础特征提取;在主干输出处引入改进的通道注意力模块(ECA-NET),削弱非关注点表达的同时,增强关注点的信息通道;阶段性调整学习策略,加快网络收敛速度。在遥感数据集上进行实验,与原CenterNet算法相比精度提高了13%,检测速度达52.61 帧/s。实验结果表明,改进的CenterNet-based目标检测方法在保证检测速度的前提下极大保留小目标的特征,从而降低漏检率,提升识别率,有效平衡了遥感目标检测的精度和速度,在实际环境中具有重要的应用价值。

关键词： 深度学习; 目标检测; 遥感图像; anchor-free; CenterNet; 注意力机制

本文引用格式

黄佳琦 , 范军芳 , 李蓓蓓 . CenterNet-based遥感图像目标检测方法研究[J]. 弹箭与制导学报, 2023 , 43(1) : 24 -31 . DOI: 10.15892/j.cnki.djzdxb.2023.01.004

Abstract

Fast and accurate identification of small and dense targets in low quality remote sensing images has become a focus of research community. A CenterNet-based target detection method that incorporates the attention mechanism is proposed. It utilizes ResNet-50 for basic feature extraction. An improved channel attention module (ECA-NET) is introduced at the backbone output to weaken the expression of non-concerned points, while enhancing the information channel of concerned points. The proposed method adjusts the learning strategy in different stages to accelerate the convergence speed of the model. Experiments are conducted using the remote sensing dataset. The improved CenterNet algorithm increases the accuracy by 13% as compared with the original method, and the detection speed reaches 52.61 frames per second. The experimental results show that CenterNet-based target detection method maximizes the remote sensing target representation capability of CenterNet under the condition of maintaining computational efficiency. It effectively balances the accuracy and computation speed of remote sensing target detection and is of great significance in practical applications.

Key words： deep learning; target detection; anchor-free; CenterNet; attention mechanism

[an error occurred while processing this directive]

参考文献

原文顺序 | 文献年度倒序 | 文中引用次数倒序

[1]	WANG L Z, YAN J N, MU L, et al. Knowledge discovery from remote sensing images: a review[J]. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 2020, 10(5): 1131-1137.

[2]	薛松, 钱立志, 张航, 等. 多源末制导弹载融合图像目标检测研究进展[J]. 弹箭与制导学报, 2021, 41(3): 67-75. XUE S, QIAN L Z, ZHANG H, et al. Review of key technologies for target detection of multi-source terminal missile-based fusion image[J]. Journal of Projectiles, Rockets, Missiles and Guidance, 2021, 41(3): 67-75.

[3]	李章维, 胡安顺, 王晓飞. 基于视觉的目标检测方法综述[J]. 计算机工程与应用, 2020, 56(8): 1-9. DOI LI Z W, HU A S, WANG X F. Survey of vision based object detection methods[J]. Computer Engineering and Applications, 2020, 56(8): 1-9. DOI

[4]	李婕, 周顺, 朱鑫潮, 等. 结合多通道注意力的遥感图像飞机目标检测[J]. 计算机工程与应用, 2022, 58(1): 209-217. DOI LI J, ZHOU S, ZHU X C, et al. Remote sensing image aircraft target detection combined with multiple channel attention[J]. Computer Engineering and Applications, 2022, 58(1): 209-217. DOI

[5]	XIAO Y, TIAN Z, YU J, et al. A review of object detection based on deep learning[J]. Multimedia Tools and Applications, 2020, 79: 23729-23791.

[6]	段仲静, 李少波, 胡建军, 等. 深度学习目标检测方法及其主流框架综述[J]. 激光与光电子学进展, 2020, 57(12): 51-66. DUAN Z J, LI S B, HU J J, et al. Review of deep learning based object detection methods and their mainstream frameworks[J]. Laser & Optoelectronics Progress, 2020, 57(12): 51-66.

[7]	ZHAO Z Q, ZHENG P, XU S T, et al. Object detection with deep learning: a review[J]. IEEE Transactions on Neural Networks and Learning Systems, 2019: 3212-3232.

[8]	CHEN Y, GAO J, ZHANG K. R-CNN-Based satellite components detection in optical images[J]. International Journal of Aerospace Engineering, 2020, 2020: 6187-6197.

[9]	GIRSHICK R. Fast R-CNN[J]. Computer Science, 2015, 2015: 1440-1448.

[10]	REN S, HE K, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017, 39(6): 1137-1149.

[11]	HE K, GKIOXARI G, PIOTR D, et al. Mask R-CNN[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017, 2017: 2117-2123.

[12]	REDOM J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]// IEEE. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition, New York: IEEE, 2016: 779-788.

[13]	REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]// IEEE. Proceedings of the Conference on Computer Vision & Pattern Recognition, 2017: 6517-6525.

[14]	REDMON J, FARHADI A. YOLOv3: an incremental improvement[EB/OL]. (2018-04-18)[2022-07-11]. https://arxiv.org/abs/1804.02767.

[15]	杨传栋, 刘桢, 马翰宇, 等. 一种基于改进Yolov3的弹载图像多目标检测方法[J]. 弹箭与制导学报, 2020, 40(4): 149-153. YANG C D, LIU Z, MA H Y, et al. A multi-target detection method for missle-borne images based on improved YOLOv3[J]. Journal of Projectiles, Rockets, Missiles and Guidance, 2020, 40(4): 149-153.

[16]	BOCHKOVSKIY A, WANG C Y, LIAO H. YOLOv4: optimal speed and accuracy of object detection[EB/OL]. (2020-04-23)[2022-07-11]. https://arxiv.org/abs/2004.10934.

[17]	郭昊昌, 于力, 刘镇涛. 一种基于YOLOv4改进的军事目标检测方法[J]. 弹箭与制导学报, 2021, 41(6): 53-58. GUO H C, YU L, LIU Z T. An object detection method for military target based on improved YOLOv4[J]. Journal of Projectiles, Rockets, Missiles and Guidance, 2021, 41(6): 53-58.

[18]	WEI L, DRAGOMIR A, DUMITRU E, et al. SSD: single shot multibox detector[C]// EVVC. Proceedings of the 2016 European Conference on Computer Vision. Berlin:Springer, 2016: 2644-2651.

[19]	李科岑, 王晓强, 林浩, 等. 深度学习中的单阶段小目标检测方法综述[J]. 计算机科学与探索, 2022, 16(1):41-58. DOI LI K C, WANG X Q, LIN H, et al. Survey of one-stage small object detection methods in deep learning[J]. Journal of Frontiers of Computer Science and Technology, 2022, 16(1): 41-58.

[20]	LAW H, DENG J. CornerNet: detecting objects as paired keypoints[J]. International Journal of Computer Vision, 2020, 128(3): 642-656.

[21]	TIAN Z, SHEN C, CHEN H, et al. Fcos: fully convolutional one-stage object detection[C]// IEEE. Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. New York: IEEE, 2020: 2132-2140.

[22]	ZHOU X, ZHUO J, KRHENBü H L, et al. Bottom-up object detection by grouping extreme and center points[EB/OL]. (2019-02-03)[2022-07-13]. https://arxiv.org/abs/1901.08043v2.

[23]	ZHOU X, WANG D, KRHENBüHL P. Objects as points[EB/OL]. (2019-04-16)[2022-07-13]. https://arxiv.org/abs/1904.07850v1.

[24]	WANG Q, WU B, ZHU P, et al. ECA-Net: efficient channel attention for deep convolutional neural networks[C]// IEEE. Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). New York: IEEE, 2020: 1131-1139.

[25]	HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]// IEEE. Proceedings of the 2016 Computer Vision and Pattern Recognition Conference. NewYork: IEEE, 2016: 770-778.

[26]	NEWELL A, YANG K, JIA D. Stacked hourglass networks for human pose estimation[C]// ECCV. Proceedings of the 2016 European Conference on Computer Vision. Berlin:Springer, 2016: 981-990.

[27]	ZHANG T, TAO D, YANG J. Discriminative locality alignment[C]// ECCV. Proceedings of the 2008 European Conference on Computer Vision. Berlin:Springer, 2008: 725-738.

[28]	HU J, SHEN L, ALBANIE S, et al. Squeeze-and-excitation networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(8): 2011-2023. DOI PMID

[29]	LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[EB/OL]. (2017-08-07)[2022-07-14]. https://arxiv.org/abs/1708.02002.

[30]	LONG Y, GONG Y, XIAO Z, et al. Accurate object localization in remote sensing images based on convolutional neural networks[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(5): 2486-2498.

Options

文章导航