Accurate Detection Method for Small Bird Targets in Low-altitude Scenarios

ZHANG Bo; CHEN Mingli; ZHOU Xiaobo; CHENG Peihao; XIA Linlin

doi:10.12382/bgxb.2025.1080

您当前的位置：

首页 >

文章列表页 >

Accurate Detection Method for Small Bird Targets in Low-altitude Scenarios

更新时间：2026-04-21

- Accurate Detection Method for Small Bird Targets in Low-altitude Scenarios
- Acta Armamentarii (2026)
- 作者机构：
  
  东北电力大学自动化工程学院,吉林,吉林,132012
- 作者简介：
- 基金信息：
- DOI：10.12382/bgxb.2025.1080
  CLC： V351.3
- Received：08 December 2025，
  
  Online First：21 April 2026，
- 稿件说明：
移动端阅览
张波，陈明理，周小波，等. 面向低空场景下的鸟类小目标精准检测方法[J/OL]. 兵工学报, 2026(2026-04-21). https://doi.org/10.12382/bgxb.2025.1080.

ZHANG B, CHEN M L, ZHOU X B, et al. Accurate detection method for small bird targets in low-altitude scenarios[J/OL]. Acta Armamentarii, 2026(2026-04-21). https://doi.org/10.12382/bgxb.2025.1080. (in Chinese)
张波，陈明理，周小波，等. 面向低空场景下的鸟类小目标精准检测方法[J/OL]. 兵工学报, 2026(2026-04-21). https://doi.org/10.12382/bgxb.2025.1080. DOI：

ZHANG B, CHEN M L, ZHOU X B, et al. Accurate detection method for small bird targets in low-altitude scenarios[J/OL]. Acta Armamentarii, 2026(2026-04-21). https://doi.org/10.12382/bgxb.2025.1080. (in Chinese) DOI：

摘要

针对航天发射场及机场共有的低空复杂背景下鸟类小目标识别难、易发生鸟击事件、威胁航空安全与人员生命等问题，提出一种基于YOLOv8n的深度增强尺度感知网络DESN-YOLOv8n(Deeply Enhanced Scale-aware Networkbasedon YOLOv8n)，专注于鸟类小目标检测。该模型通过构建主干网络特征融合模块C2F_DCA，利用深度可分离卷积与高效通道注意力机制增强空间特征提取能力，放大超小目标微弱特征。同时考虑颈部网络Concat模块直接拼接浅、深层特征，关键特征易随上下采样丢失问题，引入自适应空间特征融合模块，以聚焦小目标有效区域。在损失计算方面，使用归一化Wasserstein距离损失函数，将边界框建模为2D高斯分布，有效防止漏检和定位偏移。公共数据集AirBirds上的实验结果表明，新模型在小目标检测精度、特征增强能力和回归优化效果方面均取得显著提升。

Abstract

To address the challenge of detecting small bird targets in low-altitude complex backgrounds—common in aerospace launch sites and airports-where such targets are difficult to identify and bird strike incidents pose significant threats to aviation safety and human lives

this paper proposesa deeply enhanced scale-aware network based on YOLOv8n(DESN-YOLOv8n)

specifically designedfor small bird target detection.Themodelconstructs a backbone feature fusion moduletermedC2F_DCA

which integrates depthwise separable convolutionsand an efficient channel attentionmechanism to enhance spatial feature extraction and amplify subtle characteristics of ultra-small targets. Additionally

to address the issuethatdirect concatenation of shallow and deep features in the neck’s Concat module may lead to the loss of critical information during upsampling and downsampling

an adaptive spatial feature fusionmodule is introduced to focus on salient regions of small targets.In terms of loss computation

a normalized Wasserstein distanceloss function is employed

which models bounding boxes as 2D Gaussian distributions to effectively mitigatemissed detectionsandlocalization deviation.Experimental results on the public dataset AirBirds demonstrate that the proposed model achieves significant improvements in small target detection accuracy

feature enhancement capability

and regression optimization performance.

关键词

Keywords

references

HUSSAIN S, PEI Y. Crashworthiness assessment of NASA CRM wing in single and multiple bird strikes scenarios[J]. Meccanica, 2025, 60(5):1217-1240.

RAJPUT M, KHEDGIKAR S, TAMANE S, et al. Multi-modal bird species identification: integrating deep neural networks, transfer learning, and spectrogram analysis[C]//Proceedings of the IEEE International Conference on Information Technology. Bangalore, India: IEEE, 2024: 1-7.

JIANG J, CHAI L, LI B C. Bird detection based on audio-video fusion[C]//Proceedings of the 2025 37th Chinese Control and Decision Conference. Xiamen, China: IEEE, 2025: 4104-4109.

CHEN Y C, CHU J F, HSIEH K W, et al. Automatic wild bird repellent system that is based on deep-learning-based wild bird detection and integrated with a laser rotation mechanism[J]. Scientific Reports, 2024, 14(1): 1-15.

岳胜哲, 王正杰. 基于实例分割与光流的动态环境SLAM[J]. 兵工学报, 2024, 45(1): 156-165.

YUE S Z, WANG Z J. A SLAM in dynamic environments based on instance segmentation and optical flow[J]. Acta Armamentarii, 2024, 45(1): 156-165. (in Chinese)

郭永红, 牛海涛, 史超, 等. 基于卷积和注意力机制的小样本目标检测[J]. 兵工学报, 2023, 44(11): 3508-3515.

GUO Y H, NIU H T, SHI C, et al. Few-shot object detection based on convolution network and attention mechanism[J]. Acta Armamentarii, 2023, 44(11): 3508-3515. (in Chinese)

张良安, 陈洋, 谢胜龙, 等. 基于机器视觉与深度学习的飞机防护栅裂纹检测系统[J]. 兵工学报, 2023, 44(2): 507-516.

ZHANG L A, CHEN Y, XIE S L, et al. Crack detection system for aircraft protective grill based on machine vision and deep learning [J]. Acta Armamentarii, 2023, 44(2): 507-516. (in Chinese)

丁习文, 程宏昌, 袁渊, 等. 基于改进SSD算法的紫外像增强器视场瑕疵检测方法[J]. 兵工学报, 2024, 45(12): 4350-4363.

DING X W, CHENG H C, YUAN Y, et al. Detection method for field-of-view defect of ultraviolet image intensifier based on improved SSD algorithm [J]. Acta Armamentarii, 2024, 45(12): 4350-4363. (in Chinese)

ZHAO Y A, LṺ W Y, XU S LM, et al. DETRs beat YOLOs on real-time object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Seattle, WA, US: IEEE, 2024: 16965-16974.

钟帅, 王丽萍. MCS-RETR:改进RT-DETR的无人机航拍图像目标检测方法[J].航空学报, 2025, 46(3): 331987.

ZHONG S, WANG L P. MCS-RETR: an improved RT-DETR object detection method for UAV aerial images [J]. Acta Aeronautica et Astronautica Sinica, 2025, 46(3):331987. (in Chinese)

姚雨, 宋春林, 邵江琦.无人机航拍军事车辆实时检测及定位算法[J].兵工学报, 2024, 45(增刊1): 354-360.

YAO Y, SONG C L, SHAO J Q. Real-time detection and localization algorithm for military vehicles in drone aerial images[J].Acta Armamentarii, 2024, 45(S1): 354-360. (in Chinese)

徐图南, 高昂, 陈昱成, 等. 面向战场低空目标的轻量化识别方法[J]. 兵工学报, 2025, 46(2): 240170.

XU T N, GAO A, CHEN Y C, et al. A lightweight recognition method for low altitude targets in the battlefield[J]. Acta Armamentarii, 2025, 46(2): 240170. (in Chinese)

冯迎宾, 郭枭尊, 晏佳华. 基于多尺度注意力机制的无人机小目标检测算法[J].兵工学报, 2025, 46(1): 231124.

FENG Y B, GUO X Z, YAN J H. Small UVA target detection algorithm based on multi-scale attention mechanism[J].Acta Armamentarii, 2025, 46(1): 231124. (in Chinese)

KISANTAL M, WOJNA Z, MURAWSKI J, et al. Augmentation for small object detection[C]//Proceedings of the IEEE International Conference on Advances in Computing and Information Technology. Athens, Greece: IEEE, 2019: 119-133.

FAN J C, LIU X L, WANG X Z, et al. Multi-background island bird detection based on faster R-CNN[J]. Cybernetics and Systems, 2020, 52(1): 26-35.

XIANG W B, SONG Z Y, ZHANG G X, et al. Birds detection in natural scenes based on improved faster RCNN[J]. Applied Sciences, 2022, 12(12): 6094-6105.

BEJERAS B N, PUA M L C, VILLAVERDE J F. Birds and human detection system for crop field’s breach surveillance using single shot multibox detector (SSD) algorithm[C]//Proceedings of the Eurasia Conference on IOT, Communication and Engineering. Yunlin, Taiwan,China: IEEE, 2023: 376-381.

YU C S, SHIN Y. MCG-RTDETR: multi-convolution and context-guided network with cascaded group attention for object detection in unmanned aerial vehicle imagery[J].Remote Sensing, 2024, 16(17): 3169.

陈天华, 朱家煊, 印杰. 基于注意力机制的鸟类识别算法[J]. 计算机应用, 2024, 44(4): 1114-1120.

CHEN T H, ZHU J X, YIN J.Bird recognition algorithm based on attention mechanism[J]. Journal of Computer Applications, 2024, 44(4):1114-1120. (in Chinese)

王青羽, 姚国清, 方朝阳. 基于YOLOv8n的鄱阳湖轻量级鸟类目标检测与识别模型研究[J].江西师范大学学报(自然科学版), 2025, 49(1): 86-94.

WANG Q Y, YAO G Q, FANG C Y.The research on lightweight bird object detection and recognition model in poyang lake based on YOLOv8n[J].Journal of Jiangxi Normal University (Natural Science Edition), 2025, 49(1): 86-94. (in Chinese)

SUN H Y, WANG Y C, CAI X D, et al. AirBirds: a large-scale challenging dataset for bird strike prevention in real-world airports[C]//Proceedings of the Asian Conference on Computer Vision. Macao, China: Springer, 2022: 409-424.

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. Salt Lake City, UT, US: IEEE, 2018: 3974-3983.

DU D W, ZHU P F, WEN L Y, et al. VisDrone-DET2019: the vision meets drone object detection in image challenge results[C]//Proceedings of the IEEE International Conference on Computer Vision Workshop. Seoul, Korea: IEEE, 2019: 213-226.

Views

下载量

CNKI被引量

Alert me when the article has been cited

提交

Tools

Publicity Resources

Acoustic Emission Diagnosis of Bearing Fault Based on MSCNN-LSTM and Improved Attention Mechanism

A Radar Signal Modulation Recognition Method Based on Multi-scale Dual Attention Network

Marine Ship Target Segmentation Method Based on Improved DeepLabV3+

Research on Infrared Target Detection and Tracking in Dark Environments

A Method for Detecting the Camouflaged Small Target in Complex Scene Using Airborne Polarization Remote Sensing

Related Author

LIN Shanshan

YU Yang

LIANG Zheming

LIU Qiang

LIN Mingtong

YANG Zheng

Renlong LIAO

Zhongtao LUO

Related Institution

Shenyang Aircraft Design and Research Institute

School of Information Science and Engineering, Shenyang University of Technology

School of Information and Communication Engineering, University of Electronic Science and Technology of China

Wuhan Maritime Communication Research Institute

School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications

Postal code：100089
Tel：010-68963060/68962718 Fax：010-68963025 Email：bgxb@cos.org.cn
Technical support is provided by Beijing Founder electronics co., LTD 京ICP备05059581号-4 京公网安备11010802024360号
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
Total Visits：114991 Visits Today：908

⁰