首页 | 本学科首页   官方微博 | 高级检索  
     

小样本红外图像的样本扩增与目标检测算法
引用本文:吴晗,张志龙,李楚为,李航宇. 小样本红外图像的样本扩增与目标检测算法[J]. 控制理论与应用, 2021, 38(9): 1477-1485
作者姓名:吴晗  张志龙  李楚为  李航宇
作者单位:国防科技大学电子科学学院自动目标识别重点实验室,湖南长沙410073
基金项目:国家自然科学基金(61101185),湖南省研究生科研创新项目(CX20200044)
摘    要:深度卷积神经网络模型在很多公开的可见光目标检测数据集上表现优异,但是在红外目标检测领域,目标样本稀缺一直是制约检测识别精度的难题.针对该问题,本文提出了一种小样本红外图像的样本扩增与目标检测算法.采用基于注意力机制的生成对抗网络进行红外样本扩增,生成一系列保留原始可见光图像关键区域的红外连续图像,并且使用空间注意力机制...

关 键 词:红外图像  目标检测  稀缺样本  生成对抗网络  注意力机制  YOLOv3算法
收稿时间:2021-01-16
修稿时间:2021-07-29

Infrared image sample amplification and object detection method with small samples
WU Han,ZHANG Zhi-long,LI Chu-wei and LI Hang-yu. Infrared image sample amplification and object detection method with small samples[J]. Control Theory & Applications, 2021, 38(9): 1477-1485
Authors:WU Han  ZHANG Zhi-long  LI Chu-wei  LI Hang-yu
Affiliation:National University of Defense Technology,National University of Defense Technology,National University of Defense Technology,National University of Defense Technology
Abstract:The deep convolutional neural network model performs well on many public visible-lighted object detectiondatasets, but in the field of infrared object detection, the scarcity of object samples has always been a problem that plaguesthe increase in detection and recognition accuracy. In response to this problem, this paper proposes an algorithm forsample amplification and object detection of infrared images with small samples. The attention-based generative adversarialnetwork was adopted to amplify the infrared samples to generate a series of infrared continuous images retaining the keyareas of the original visible light image, and the spatial attention model and other methods were used to further improvethe recognition accuracy of the YOLOv3 object detection algorithm. The experimental results on the OSU Color-Thermalinfrared-visible light dataset and Grayscale-Thermal dataset show that the infrared data amplification technology of theproposed algorithm effectively improves the accuracy of the deep convolutional neural network for infrared object detection,and the mAP (mean average precision) of the proposed method is 20% higher than the original YOLOv3 algorithm.
Keywords:infrared images   object detection   scarce samples   generative adversarial network   attention model   YOLOv3 algorithm
本文献已被 万方数据 等数据库收录!
点击此处可从《控制理论与应用》浏览原始摘要信息
点击此处可从《控制理论与应用》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号