激光与光电子学进展, 2020, 57 (12): 120002, 网络出版: 2020-06-03   

大视场域的目标检测与识别算法综述 下载: 2011次

Review on Object Detection and Recognition in Large Field of View
作者单位
1 中国科学院上海微系统与信息技术研究所微系统技术重点实验室, 上海 201800
2 中国科学院大学, 北京 100049
摘要
目标精确感知与识别为信息化战争提供了一个重要的技术增长点,全景视觉传感设备因其拥有大视场(LFOV)范围优势而逐渐被应用于安防及**领域中的目标检测与识别任务。首先从相机成像模型、图像成像质量以及目标物体的非对称性三个方面对存在的困难以及挑战进行阐述。基于是否进行畸变校正预处理,将近年来LFOV域的目标检测与识别算法分为基于畸变校正的目标检测与识别算法和基于原始LFOV图像的目标检测与识别算法两类,并针对这两类算法进行了全面梳理和总结,对当前LFOV域的目标检测与识别各类算法的统一性和差异性进行思考分析,探讨其未来发展趋势。
Abstract
Precise object detection and recognition plays an important role in information-based warfare. Due to panoramic vision sensors’ large field of view (LFOV), they are gradually applied to security and military areas. In this paper, first, the difficulties and challenges in the development of object detection and recognition in LFOV are presented from three aspects: camera imaging model, image imaging quality, and asymmetry of object. Then, based on whether the distortion correction preprocessing is carried out or not, the object detection and recognition algorithms in LFOV are classified into two categories: distortion correction based algorithms and original LFOV image based algorithms. These two kinds of algorithms are comprehensively combed and summarized. Finally, the paper analyzes the unity and difference of various algorithms for object detection and recognition in LFOV and discusses their future development trend.

李唐薇, 童官军, 李宝清, 卢晓洋. 大视场域的目标检测与识别算法综述[J]. 激光与光电子学进展, 2020, 57(12): 120002. Tangwei Li, Guanjun Tong, Baoqing Li, Xiaoyang Lu. Review on Object Detection and Recognition in Large Field of View[J]. Laser & Optoelectronics Progress, 2020, 57(12): 120002.

本文已被 5 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!