复杂背景目标探测识别技术综述

doi:10.3981/j.issn.2097-0781.2022.04.005

前瞻科技 ›› 2022, Vol. 1 ›› Issue (4): 69-80.DOI: 10.3981/j.issn.2097-0781.2022.04.005

复杂背景目标探测识别技术综述

崔得东¹^,²(), 冯涛¹^,², 白昆¹^,², 杨挺², 马龙¹^,², 尚毅²

1.西安现代控制技术研究所，西安 710065
2.西安导引科技有限责任公司，西安 710065

收稿日期:2022-10-24 修回日期:2022-11-03 出版日期:2022-12-20 发布日期:2023-01-17
作者简介:崔得东，研究员，中国兵器科技带头人，国务院政府特殊津贴专家。主要研究方向为智能化弹药导引技术。担任国家重点型号项目副总设计师。获国家科学技术进步奖一等奖1项，国防科学技术进步奖二等奖、省部级科学技术进步奖二等奖4项。出版专著1部，发表论文10余篇，获授权国防发明专利11项、新型实用专利1项。电子信箱：dedocui@126.com。

Overview of Target Detection and Recognition Technology in Complex Background

CUI Dedong¹^,²(), FENG Tao¹^,², BAI Kun¹^,², YANG Ting², MA Long¹^,², SHANG Yi²

1. Xi’an Modern Control Technology Research Institute, Xi’an 710065, China
2. Xi’an Guidance Technology Co., Ltd., Xi’an 710065, China

Received:2022-10-24 Revised:2022-11-03 Online:2022-12-20 Published:2023-01-17

摘要/Abstract

摘要：

目标探测识别技术是图像信息处理技术的热门研究领域，面向陆战场复杂背景条件的目标探测识别技术如何成熟应用更是该领域亟待解决的痛点问题。文章介绍了陆战场复杂背景条件下影响目标探测识别性能的主要因素；分析了陆战场目标背景的复杂性、多样性和不确定性；回顾了国内外目标识别技术的发展历程；阐述了陆战场复杂背景目标探测识别的关键技术，包括多波段复合探测技术、目标自主识别技术、弱小目标检测识别技术、多通道异构信息融合技术；展望了陆战场目标探测识别技术的发展趋势。

关键词: 陆战场, 复杂背景, 目标探测识别

Abstract:

Target detection and recognition technology is a popular research field in image information processing technology. How to maturely apply this technology to land battlefields with complex backgrounds is an urgent pain point to be resolved. This paper first introduces the main factors affecting the performance of target detection and recognition under complex background conditions of land battlefields and analyzes the complexity, diversity, and uncertainty of the target background of land battlefields. Then, it reviews the research status of target recognition algorithms. Furthermore, it expounds on the key technologies of target detection and recognition under complex backgrounds, including multispectral complex target detection, automatic target recognition, dim and small target detection, and multi-channel heterogeneous information fusion. Finally, this paper predicts the tendencies of target detection and recognition technology in land battlefields.

Key words: land battlefield, complex background, target detection and recognition

崔得东, 冯涛, 白昆, 杨挺, 马龙, 尚毅. 复杂背景目标探测识别技术综述[J]. 前瞻科技, 2022, 1(4): 69-80.

CUI Dedong, FENG Tao, BAI Kun, YANG Ting, MA Long, SHANG Yi. Overview of Target Detection and Recognition Technology in Complex Background[J]. Science and Technology Foresight, 2022, 1(4): 69-80.

图/表 7

图1 主动毫米波/红外成像/半主动激光三模复合导引头及其复合探测系统

图2 目标表征技术的发展里程碑 LBP：Local Binary Pattern，局部二值模式特征；Haar：Haar-like Features，哈尔特征；HOG：Histogram of Oriented Gradient，方向梯度直方图特征；DPM：Deformable Parts Model，部件可变模型。

图3 传统目标表征技术流程

表1 典型的传统目标表征技术的优缺点

特征	优点	缺点	适用范围
LBP	对光照不敏感；运算速度快	计算量大	模板匹配
HOG	忽略光照颜色影响；表征数据维数降低	描述子生成过程长，实时性差；难以处理遮挡问题；噪点敏感	目标探测
SIFT	旋转、尺度缩放、亮度不变性；抗遮挡	方向信息敏感	目标识别目标分类

图4 典型的基于深度学习的多波段图像融合模型结构

图5 特征提取器示意图

图6 基于目标识别的战场态势感知模型

参考文献 32

[1]	陈健, 王永明, 赵革, 等. 舰船目标特性研究综述[J]. 舰船电子工程, 2008, 28(5): 17-21.
[2]	刘进, 张天序. 不变矩构造方法的研究[J]. 华中科技大学学报(自然科学版), 2003, 31(3): 1-3.
[3]	张天序, 刘进. 不变矩稳定性及基于多级特征模型的目标识别研究[J]. 计算机学报, 2004, 27(10): 1335-1340.
[4]	张天序, 曹杨, 刘进, 等. 基于不变矩的前视红外图像机场目标识别[J]. 华中科技大学学报(自然科学版), 2007, 35(1): 17-19.
[5]	孙红辉, 王红霞, 田涛. 一种基于不变矩和BP网络的目标识别方法[J]. 微电子学与计算机, 2011, 28(3): 63-65.
[6]	Tabb K, Davey N, Adams R, et al. The recognition and analysis of animate objects using neural networks and active contour models[J]. Neurocomputing, 2002, 43(1-4): 145-172. DOI URL
[7]	桑农, 唐红卫, 曹治国, 等. 基于Gabor滤波器组和特征分组神经网络目标识别[J]. 华中科技大学学报(自然科学版), 2003, 31(11): 47-49.
[8]	杨福平, 白振兴. BP神经网络和D-S证据理论的目标识别[J]. 火力与指挥控制, 2006, 31(10): 88-90.
[9]	Bobick A F, Davis J W. The recognition of human movement using temporal templates[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2001, 23(3): 257-267. DOI URL
[10]	牛照东, 许旻, 王卫华, 等. 一种利用层次化模板匹配的地面建筑红外图像自动识别方法[C]// 第十三届全国信号处理学术年会(CCSP-2007)论文集. 北京: 中国仪器仪表学会, 中国电子学会, 2007: 179-182.
[11]	张义广, 冯志高, 张天序, 等. 基于可见光图像模板匹配的目标识别算法[J]. 激光与红外, 2007, 37(12): 1322-1324.
[12]	Han F, Shan Y, Cekander R, et al. A two-stage approach to people and vehicle detection with HOG-based SVM[C]// Performance Metrics for Intelligent Systems 2006 Workshop. 2006: 133-140.
[13]	Li X, Guo X S. A HOG feature and SVM based method for forward vehicle detection with single camera[C]// 2013 5th International Conference on Intelligent Human-Machine Systems and Cybernetics. Piscataway: IEEE Press, 2013: 263-266.
[14]	Pang Y W, Yuan Y, Li X L, et al. Efficient HOG human detection[J]. Signal Processing, 2011, 91(4): 773-781. DOI URL
[15]	Viola P, Jones M. Rapid object detection using a boosted cascade of simple features[C]// Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE Press, 2001, 1: 511-518.
[16]	Yang M, Crenshaw J, Augustine B, et al. AdaBoost-based face detection for embedded systems[J]. Computer Vision and Image Understanding, 2010, 114(11): 1116-1125. DOI URL
[17]	Padilla R, Filho Costa C F F, Costa M G F. Evaluation of haar cascade classifiers designed for face detection[J]. International Journal of Computer, Electrical, Automation, Control and Information Engineering, 2012, 6(4): 466-469.
[18]	Girshick R, Donahue J, Darrell T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]// 2014 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE Press, 2014: 580-587.
[19]	Sultana F, Sufian A, Dutta P. A review of object detection models based on convolutional neural network[M]// Advances in Intelligent Systems and Computing. Singapore: Springer Singapore, 2020: 1-16.
[20]	Girshick R. Fast R-CNN[C]// 2015 IEEE International Conference on Computer Vision. Piscataway: IEEE Press, 2005: 1440-1448.
[21]	Ren S Q, He K M, Girshick R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149. DOI PMID
[22]	Purkait P, Zhao C, Zach C. SPP-net: Deep absolute pose regression with synthetic views[DB/OL]. arXiv Preprint: 1712.03452, 2017.
[23]	Liu W, Anguelov D, Erhan D, et al. SSD: Single shot MultiBox detector[C]// Computer Vision-ECCV 2016. Cham: Springer International Publishing, 2016: 21-37.
[24]	Arya M C, Rawat A. A review on YOLO (You Look Only One): An algorithm for real time object detection[J]. International Journal of Engineering Science 2020, 11: 554-557.
[25]	Redmon J, Divvala S, Girshick R, et al. You only look once: Unified, real-time object detection[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE Press, 2016: 779-788.
[26]	Redmon J, Farhadi A. YOLO9000: Better, faster, stronger[C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE Press, 2017: 6517-6525.
[27]	Redmon J, Farhadi A. YOLOv3: An incremental improvement[DB/OL]. arXiv Preprint: 1804.02767, 2018.
[28]	Bochkovskiy A, Wang C Y, Liao H Y M. YOLOv4: Optimal speed and accuracy of object detection[DB/OL]. arXiv Preprint: 2004.10934, 2004.
[29]	Adarsh P, Rathi P, Kumar M. YOLOv3-tiny: Object Detection and Recognition using one stage improved model[C]// 2020 6th International Conference on Advanced Computing and Communication Systems (ICACCS). Piscataway:IEEE Press, 2020: 687-694.
[30]	Carion N, Massa F, Synnaeve G, et al. End-to-end object detection with transformers[C]// Computer Vision-ECCV 2020. Cham: Springer International Publishing, 2020: 213-229.
[31]	Stewart R, Andriluka M, Ng A Y. End-to-end people detection in crowded scenes[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE Press, 2016: 2325-2333.
[32]	Hosang J, Benenson R, Schiele B. Learning non-maximum suppression[C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE Press, 2017: 6469-6477.

复杂背景目标探测识别技术综述

Overview of Target Detection and Recognition Technology in Complex Background

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 7

参考文献 32

相关文章 0

编辑推荐

Metrics

访问统计

友情链接

联系我们

官方微信公众号