基于感知延伸与锚框最适匹配的遥感图像目标检测算法 -- 西北工业大学学报,2023,41(4):820-830

	论文:2023,Vol:41,Issue(4):820-830
	引用本文：
	洪伟, 赵祥模, 王鹏, 李晓艳, 邸若海, 吕志刚, 王储. 基于感知延伸与锚框最适匹配的遥感图像目标检测算法[J]. 西北工业大学学报
	HONG Wei, ZHAO Xiangmo, WANG Peng, LI Xiaoyan, DI Ruohai, LYU Zhigang, WANG Chu. Remote sensing target detection algorithm based on perceptual extension and anchor frame best-fit matching[J]. Journal of Northwestern Polytechnical University

基于感知延伸与锚框最适匹配的遥感图像目标检测算法

洪伟¹, 赵祥模², 王鹏³, 李晓艳², 邸若海², 吕志刚², 王储²

1. 西安工业大学兵器科学与技术学院, 陕西西安 710021;
2. 西安工业大学电子信息工程学院, 陕西西安 710021;
3. 西安工业大学发展规划处, 陕西西安 710021

摘要:

针对遥感图像目标成像小、背景复杂、分布拥挤的问题,将旋转目标检测方法S²ANet作为基线网络,提出一种基于感知延伸与锚框最适匹配的遥感图像目标检测算法(HQ-S²ANet)。构建协同注意力模块(SEA),捕获特征像素间关系的同时扩展模型感知区域,实现目标与全局的关系建模;针对遥感图像背景复杂问题,改进特征金字塔(FPN)特征融合过程,在特征融合下采样过程中将感知延伸卷积模块与常规卷积交替堆叠形成感知延伸特征金字塔模块(HQFPN),保证低层细节位置信息的同时,延伸感知范围以增强模型信息捕捉能力;为解决遥感目标图像分布拥挤的问题,利用高质量锚框匹配方法(MaxIoUAssigner_HQ),通过常数因子控制锚框真值分配,在保证召回率的同时,防止低质量锚框匹配产生。实验结果表明,在DOTA数据集下,与S²ANet算法相比,HQ-S²ANet平均精度(mAP)提高3.1%,召回率(Recall)均值提高1.6%,而参数量仅增加2.61M,所提算法有效增强了遥感图像目标检测能力。

关键词: 遥感图像特征融合锚框匹配旋转检测

Remote sensing target detection algorithm based on perceptual extension and anchor frame best-fit matching

HONG Wei¹, ZHAO Xiangmo², WANG Peng³, LI Xiaoyan², DI Ruohai², LYU Zhigang², WANG Chu²

1. School of Ordnance Science and Technological, Xi'an Technological University, Xi'an 710021, China;
2. School of Electronic Information Engineering, Xi'an Technological University, Xi'an 710021, China;
3. Development Planning Service, Xi'an Technological University, Xi'an 710021, China

Abstract:

Aiming at the small imaging, complex background and crowded distribution of remote sensing image targets, a remote sensing image target detection algorithm (HQ-S²ANet) based on perceptual extension and anchor frame optimal matching is proposed by using the rotating target detection method S²ANet as a baseline network. Firstly, a cooperative attention(SEA) module is built to capture the relationship among the feature pixels when extending the model perception area to realize the relationship modeling between the target and the global. Secondly, the feature pyramid (FPN) feature fusion process is improved to form a perceptual extension feature pyramid module (HQFPN), which guarantees the low-level detail position information in the down sampling process when extending the perception area to enhance the model information capturing capability. Finally, a high-quality anchor frame is used to detect the target by using the high quality anchor frame as the baseline network. The high-quality anchor frame matching method (MaxIoUAssigner_HQ) is used to control the anchor frame truth value assignment by using a constant factor to ensure the recall rate while preventing the generation of low-quality anchor frame matching. The experimental results show that, under the DOTA dataset, the average accuracy(mAP) of HQ-S²ANet is improved by 3.1%, the parameters number increased by only 2.61M and the average recall(recall) is improved by 1.6% compared with the S²ANet algorithm, and the present algorithm effectively enhances the detection capability of the remote sensing image target.

Key words: remote sensing image feature fusion anchor frame rotation detection

收稿日期: 2022-07-26 修回日期:

DOI: 10.1051/jnwpu/20234140820

基金项目: 国家自然科学基金(62171360)、陕西省科技厅重点研发计划(2022GY-110)、国家重点研发计划(2022YFF0604900)与2022年度陕西高校青年创新团队项目资助

通讯作者: 王鹏(1978—),西安工业大学教授,主要从事图像处理与智能检测研究。e-mail:wp_xatu@163.com Email：wp_xatu@163.com

作者简介: 洪伟(1999—),西安工业大学硕士研究生,主要从事深度学习与遥感图像目标检测研究。

相关功能

PDF(4155KB) Free

打印本文

把本文推荐给朋友

作者相关文章

洪伟在本刊中的所有文章

赵祥模 在本刊中的所有文章

王鹏在本刊中的所有文章

李晓艳 在本刊中的所有文章

邸若海 在本刊中的所有文章

吕志刚 在本刊中的所有文章

王储在本刊中的所有文章


	参考文献:
	[1] 胡凡. 遥感图像中密集小目标检测算法设计与实现[D]. 南京:东南大学, 2021 HU Fan. Design and implementation of dense small target detection algorithm in remote sensing images[D]. Nanjing: Southeast University, 2021 (in Chinese) [2] YANG X, YANG J, YAN J, et al. SCRDet: towards more robust detection for small, cluttered and rotated objects[C]//2019 IEEE/CVF International Conference on Computer Vision, Seoul, 2019: 8231-8240 [3] ZHANG G, LU S, ZHANG W. CAD-Net: a context-aware detection network for objects in remote sensing imagery[J]. IEEE Trans on Geoscience and Remote Sensing, 2019(99): 1-10 [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] HAN J, DING J, XUE N, et al. Redet: a rotation-equivariant detector for aerial object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021: 2786-2795 [6] XIA G S, BAI X, DING J, et al. DOTA: a large-scale dataset for object detection in aerial images[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 3974-3983 [7] DING J, XUE N, LONG Y, et al. Learning roi transformer for oriented object detection in aerial images[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 2849-2858 [8] LI Y, LI S, DU H, et al. YOLO-ACN: focusing on small target and occluded object detection[J]. IEEE Access, 2020(99): 1 [9] HAN J, DING J, LI J, et al. Align deep features for oriented object detection[J]. IEEE Trans on Geoscience and Remote Sensing, 2020, 60: 1-11 [10] WEI L, DRAGOMIR A, DUMITRU E, et al. SSD: single shot multiBox detector[C]//European Conference on Computer Vision, 2016: 21-37 [11] 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 [12] LIN T Y, DOLLAR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//IEEE Conference on Computer Vision and Pattern Recognition, 2017 [13] LI W, CHEN Y, HU K, et al. Oriented reppoints for aerial object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022: 1829-1838 [14] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778 [15] ANG Q, WU B, ZHU P, et al. ECA-Net: Efficient channel attention for deep convolutional neural networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 11534-11542 [16] YU F, KOLTUN V. Multi-scale context aggregation by dilated convolutions[J/OL].(2016-04-16)[2022-06-15]. https://arxiv.org/abs/1511.07122 [17] MEHTA S, RASTEGARI M, CASPI A, et al. Espnet: Efficient spatial pyramid of dilated convolutions for semantic segmentation[C]//Proceedings of the European Conference on Computer Vision, 2018: 552-568 [18] HOU Liping, LU Ke, XUE Jian, et al. Shape-adaptive selection and measurement for oriented object detection[C]//the 36th AAAI Conference on Artificial Intelligence, 2022 [19] YANG X, YAN J. Arbitrary-oriented object detection with circular smooth label[C]//European Conference on Computer Vision, 2020: 677-694 [20] YANG Z, LIU S, HU H, et al. Reppoints: point set representation for object detection[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019: 9657-9666 [21] GUO Z, LIU C, ZHANG X, et al. Beyond bounding-box: convex-hull feature adaptation for oriented and densely packed object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021: 8792-8801 [22] ZHANG S, CHI C, YAO Y, et al. Bridging the gap between anchor-based and anchor-free detection via adaptive training sample selection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 9759-9768

邮编:710072 电话：029-88495455 Email：xuebao@nwpu.edu.cn

本系统由北京仁和汇智信息技术有限公司设计开发技术支持：info@rhhz.net