Low-altitude small-sized object detection using lightweight feature-enhanced convolutional neural network

• 作者: YE Tao ZHAO Zongyang ZHANG Jun CHAI Xinghua ZHOU Fuqiang
• 来源期刊: 系统工程与电子技术（英文版）
• 年，卷（期）: 2021,（4）
• 所属期刊栏目: ELECTRONICS TECHNOLOGY
• 页码范围: 841-853
• 页数: 13页
• 语种: 英文
• DOI: 10.23919/JSEE.2021.000073

摘要

Unauthorized operations referred to as "black flights" of unmanned aerial vehicles (UAVs) pose a significant danger to public safety, and existing low-attitude object detection al-gorithms encounter difficulties in balancing detection precision and speed. Additionally, their accuracy is insufficient, particu-larly for small objects in complex environments. To solve these problems, we propose a lightweight feature-enhanced convolu-tional neural network able to perform detection with high preci-sion detection for low-attitude flying objects in real time to provide guidance information to suppress black-flying UAVs. The proposed network consists of three modules. A lightweight and stable feature extraction module is used to reduce the com-putational load and stably extract more low-level feature, an en-hanced feature processing module significantly improves the feature extraction ability of the model, and an accurate detec-tion module integrates low-level and advanced features to im-prove the multiscale detection accuracy in complex environ-ments, particularly for small objects. The proposed method achieves a detection speed of 147 frames per second (FPS) and a mean average precision (mAP) of 90.97％ for a dataset com-posed of flying objects, indicating its potential for low-altitude object detection. Furthermore, evaluation results based on mi-crosoft common objects in context (MS COCO) indicate that the proposed method is also applicable to object detection in gene-ral.