| Anchor-free轻量级红外目标检测方法（特邀） |
| |
| 引用本文： | 高凡,杨小冈,卢瑞涛,王思宇,高久安,夏海.Anchor-free轻量级红外目标检测方法（特邀）[J].红外与激光工程,2022,51(4):20220193-1-20220193-9. |
| |
| 作者姓名： | 高凡 杨小冈 卢瑞涛 王思宇 高久安 夏海 |
| |
| 作者单位： | 1.北京华航无线电测量研究所，北京 100013 |
| |
| 基金项目： | 国家自然科学基金（61806209）；陕西省自然科学基金（2020 JQ-490）；航空科学基金（201851 U8012） |
| |
| 摘 要： | 针对红外目标的特点，提出了一种anchor-free轻量级红外目标检测方法，提高了嵌入式平台对红外目标的检测能力。针对计算资源有限的平台，提出了一种新的轻量级卷积结构，引入非对称卷积增强标准卷积的特征表达能力，同时有效减少参数和计算量。设计并行多路特征通道，经过通道拼接生成丰富的特征，结合注意力模块和Channel Shuffle构建轻量级特征提取单元。增加SkipBranch促进浅层信息向高层传递，进一步丰富高层特征。在FLIR数据集进行实验验证，设计的轻量级网络结构精度为81.7% ，超过了 YOLOv4-tiny，但模型参数量减少了75.0%、计算量下降了71.1%，并且推理时间压缩了91.3%，能够满足嵌入式平台红外目标的实时检测需求。
|
| 关 键 词： | 红外目标 轻量化 目标检测 神经网络 非对称卷积 |
| 收稿时间： | 2022-03-17 |
Anchor-free lightweight infrared object detection method (Invited) |
| |
| Affiliation: | 1.Beijing Huahang Radio Measurement Institute, Beijing 100013, China2.Missile Engineering Institute, Rocket Force University of Engineering, Xi’an 710025, China |
| |
| Abstract: | According to the characteristics of infrared targets, an anchor-free lightweight infrared target detection method was proposed, which improved the detection ability of embedded platform. For the platform with limited computing resources, a new lightweight convolution structure was proposed. Asymmetric convolution was introduced to enhance the feature expression ability of standard convolution, reducing the amount of parameters and computation effectively. A lightweight feature extraction unit was constructed by designing parallel multi-feature path, which generated rich features through channel concatation, then combining with attention module and channel shuffle. SkipBranch was added to promote the transmission of shallow information to the high level and further enrich the characteristics of the high level. Experiments on FLIR dataset showed that the accuracy of the designed lightweight network structure was 81.7%, which exceeded YOLOv4-tiny. However, the model parameters and calculation amount were reduced by 75.0% and 71.1% respectively, and the reasoning time was compressed by 91.3%, which could meet the real-time detection requirements of infrared object on embedded platform. |
| |
| Keywords: | |
|
| 点击此处可从《红外与激光工程》浏览原始摘要信息 |
|
点击此处可从《红外与激光工程》下载全文 |
|
