紧凑折反式仿生复眼及图像快速拼接识别算法

doi:10.12382/bgxb.2021.0393

兵工学报 ›› 2022, Vol. 43 ›› Issue (8): 1845-1857.doi: 10.12382/bgxb.2021.0393

紧凑折反式仿生复眼及图像快速拼接识别算法

曹昭睿¹，郝永平¹，刘万成²，白帆¹，孙颢洋¹，张慧³，李宇海²

(1.沈阳理工大学装备工程学院, 辽宁沈阳 110159; 2.光电信息控制和安全技术重点实验室, 天津 300308;3.沈阳理工大学机械工程学院, 辽宁沈阳 110159)

上线日期:2022-07-24
作者简介:曹昭睿(1993—), 男, 博士研究生。E-mail: caozhaorui@163.com
基金资助:
国防科技基础加强计划技术领域基金项目(2020-JCJQ-JJ-422)；“十三五”装备预研兵器工业联合基金项目(6141B012841)；装备预研重点实验室基金项目(6142107190207)；辽宁省教育厅青年基金项目(LG201931)；国防科技重点实验室基金项目（2021JCJQLB055009）

Compact Catadioptric Bionic Compound Eye and Fast Image Mosaic Recognition Algorithm

CAO Zhaorui¹, HAO Yongping¹, LIU Wancheng², BAI Fan¹, SUN Haoyang¹, ZHANG Hui³, LI Yuhai²

(1. School of Equipment Engineering, Shenyang Ligong University, Shenyang 110159, Liaoning, China; 2. Science and Technology on Electro-optical Information Security Control Laboratory, 300308, Tianjin, China; 3. School of Mechanical Engineering, Shenyang Ligong University, Shenyang 110159, Liaoning, China)

Online:2022-07-24

摘要/Abstract

摘要： 针对多相机阵列复眼质量大与图像数据生成量高的问题，开展多通道-单探测器构型的大视场仿生复眼成像系统及其图像快速拼接识别算法的研究。通过光路折反与光程归一的方式，实现多视角子眼共像面与分区成像。根据折反式复眼成像特征，提出特征图级下基于区域结构相似性的图像快速拼接算法。配合目标识别卷积神经网络，实现紧凑空间下大视场全局图像的快速重建与目标识别。所提出的复眼系统有效光学视场角为138°×75°、光学尺寸为29.78 mm×19.74 mm× 6.86 mm、全局图像拼接速度为0.011 s。紧凑型折反式仿生复眼具有体积小、视场大、计算开销少的特点，能够为载荷与算力受限的小型无人设备与低速弹箭提供广域视觉成像与快速图像拼接识别能力。

关键词: 仿生复眼, 多孔径成像, 光路折反, 图像拼接, 目标识别

Abstract: To resolve the problems of high mass and high volume of image data generated by the multi-channel camera array in a bionic compound eye, a large FOV bionic compound eye imaging system and an image fast mosaic recognition algorithm based on multi-channel single detector is proposed. By using optical path refraction and normalization, multi view sub eye image plane coplanar and single photodetector partition imaging are realized. Based on the imaging characteristics of catadioptric eyes, an image mosaic algorithm based on the structural similarity at the feature map level is proposed. Fast reconstruction and target recognition of global images with a large field of view (FOV) in a compact space are realized by using a convolution neural network for target recognition. The proposed compound eye system has an effective optical FOV of 138°×75°, an optical dimension is 29.78 mm×19.74 mm×6.86 mm, and a global image mosaic speed is 0.011 s. The compact catadioptric compound eye has the advantages of small volume, large FOV, and low computational cost. It can provide wide area vision and fast image mosaic recognition capabilities for small unmanned equipment and low speed projectiles with limited load and computational power.

Key words: bioniccompoundeye, multi-apertureimaging, opticalpathcatadioptric, imagestitching, targetrecognition

中图分类号:

TH741

曹昭睿，郝永平，刘万成，白帆，孙颢洋，张慧，李宇海. 紧凑折反式仿生复眼及图像快速拼接识别算法[J]. 兵工学报, 2022, 43(8): 1845-1857.

CAO Zhaorui, HAO Yongping, LIU Wancheng, BAI Fan, SUN Haoyang, ZHANG Hui, LI Yuhai. Compact Catadioptric Bionic Compound Eye and Fast Image Mosaic Recognition Algorithm[J]. Acta Armamentarii, 2022, 43(8): 1845-1857.

参考文献

［1］周琼, 闫杰. 大跟踪场图像导引头位标器设计[J]. 南京理工大学学报(自然科学版), 2009, 33(1): 93-95.
ZHOU Q, YAN J. Design of coordinator of image seeker with large track field[J]. Journal of Nanjing University of Science and Technology (Natural Science), 2009, 33(1): 93-95. (in Chinese)
[2］ LI Y X, LIM Y J, XU Q K, et al. Raster adaptive optics for video rate aberration correction and large FOV multi-photon imaging[J]. Biomedical optics express, 2020, 11(2): 1032-1042.
[3］陈婷. 球面/平面多尺度多孔径光学成像系统研究[D]. 哈尔滨:哈尔滨工业大学, 2016.
CHENG T. Sphere/plane multi-scale & multi-aperture optical imaging systems[D]. Harbin: Harbin Institute of Technology, 2016.(in Chinese)
[4］王红. 大相对孔径大视场紫外告警相机光学系统[J]. 光学精密工程, 2014, 22(12): 3212-3216.
WANG H. Optical system for ultraviolet warning camera with wide field of view and large relative aperture[J]. Optics and Precision Engineering, 2014, 22(12):3212-3216. (in Chinese)
[5］燕飞, 郭云芝, 史立芳, 等. 基于仿生复眼系统的图像拼接技术[J]. 航空兵器, 2017(6): 49-53.
YAN F, GUO Y Z, SHI L F, et al. Research of image mosaic algorithm based on bionic compound eye system[J]. Aero Weaponry, 2017(6): 49-53. (in Chinese)
[6］曹阿秀. 小型化复眼结构多维信息探测关键技术研究[D]. 中国科学院大学(中国科学院光电技术研究所), 2019.
CAO A X. Research on multi-dimensional information detection technology by miniaturized compound eye structure[D]. Chinese Academy of Science (Institute of Optics and Electronics), 2019. (in Chinese)
[7］王玉伟. 仿生复眼全景立体成像关键技术研究[D]. 中国科学技术大学, 2017.
WANG Y W. Research on key technologies of bionic compound eye for panoramic stereo imaging[D]. University of Science and Technology of China, 2017. (in Chinese)
[8］杨超, 裘溯, 金伟其, 等. 基于多微面光纤面板的仿生复眼图像拼接及定位算法[J]. 兵工学报, 2018, 39(6): 1144-1150.
YANG C, QIU S, JIN W Q, et al. Image mosaic and positioning algorithms of bionic compound eye based on fiber faceplate[J]. Acta Armamentarii, 2018, 39(6): 1144-1150. (in Chinese)
[9］徐扬. 仿生眼图像拼接技术研究[D]. 长春:长春理工大学, 2019.
XU Y. Study on splicing technology of bionic eye image [D]. Changchun: Changchun University of Science and Technology, 2019. (in Chinese)
[10］张家铭, 陈宇, 覃海棋, 等. 大视场并列型仿生复眼光学系统[J]. 光学精密工程, 2020, 28(5): 1012-1020.
ZHANG J M, CHEN Y, TAN H Q, et al. Optical system of bionic compound eye with large field of view[J]. Optics and Precision Engineering, 2020, 28(5): 1012-1020. (in Chinese)
[11］徐浩然. 小型无人机光电吊舱视轴稳定技术研究[D]. 长春:长春理工大学,2020.
XU H R. Research on visual axis stabilization technology of small UAV photoelectric pod[D]. Changchun: Changchun University of Science and Technology, 2020. (in Chinese)
[12］ PAN J W, WU G Q, YU J H, et al. Detecting the early infarct core on non-contrast CT images with a deep learning residual network[J]. Journal of stroke and cerebrovascular diseases:the official journal of National Stroke Association, 2021, 30(6): 105752.
[13］ LIU T, PANG B, AI S M, et al. Study on visual detection algorithm of sea surface targets based on improved YOLOv3[J]. Sensors, 2020, 20(24): 7263.
[14］ LIN M B, JI R R, WANG Y, et al. HRank: filter pruning using high-rank feature map[C]∥Proceedings of 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle, WA, US: IEEE, 2021: 19874247.
[15］吕晨, 程德强, 寇旗旗, 等. 基于YOLOv3和ASMS的目标跟踪算法[J]. 光电工程, 2021, 48(2): 70-80.
L C, CHENG D Q, KOU Q Q, et al. Target tracking algorithm based on YOLOv3 and ASMS[J]. Opto-Electronic Engineering, 2021, 48(2): 70-80. (in Chinese)
[16］魏鑫. 基于冗余图像筛选的全景图像拼接技术研究[D]. 烟台:烟台大学, 2021.
WEI X. Research on Image Redundancy Filtering for Panorama Stitching[D]. Yantai: Yantai University, 2021. (in Chinese)

[1]	宋晓茹, 刘康, 高嵩, 陈超波, 阎坤. 复杂战场环境下改进YOLOv5军事目标识别算法研究[J]. 兵工学报, 2024, 45(3): 934-947.
[2]	赵晓冬, 张洵颖. 基于YOLOv5的无人车自主目标识别优化算法[J]. 兵工学报, 2023, 44(9): 2732-2744.
[3]	刘冰, 郝新红, 周文, 杨瑾. 基于BAS-BPNN的调频无线电引信目标与扫频干扰识别方法[J]. 兵工学报, 2023, 44(8): 2391-2403.
[4]	霍健, 陈慧敏, 马云飞, 郭鹏宇, 杨旭, 孟祥盛. 基于MEMS激光雷达的车辆目标识别算法[J]. 兵工学报, 2023, 44(4): 940-948.
[5]	丁伯圣, 张睿恒, 徐立新, 陈慧敏. 基于灰度补偿和特征融合的沙尘图像修复方法[J]. 兵工学报, 2023, 44(10): 3115-3126.
[6]	王亮，陈建华，李烨. 一种基于深度学习的无人艇海上目标识别技术[J]. 兵工学报, 2022, 43(S2): 13-19.
[7]	李良福，陈卫东，高强，许开銮，刘轩，何曦，钱钧. 基于深度学习的光电系统智能目标识别[J]. 兵工学报, 2022, 43(S1): 162-168.
[8]	薛灵芝, 曾向阳, 杨爽. 基于生成对抗网络的水声目标识别算法[J]. 兵工学报, 2021, 42(11): 2444-2452.
[9]	王健，秦春霞，杨珂，任萍. 基于导向重构与降噪稀疏自编码器的合成孔径雷达目标识别[J]. 兵工学报, 2020, 41(9): 1861-1870.
[10]	黄莎玲，朱鸿志，程祥利，谢玉斌. 自适应阈值侵彻引信层目标识别算法[J]. 兵工学报, 2020, 41(9): 1762-1771.
[11]	邹汝平，刘建书. 基于概率假设密度滤波与无迹Kalman滤波的多目标跟踪与识别[J]. 兵工学报, 2020, 41(8): 1502-1508.
[12]	杨超，裘溯，金伟其，戴佳琳. 基于多微面光纤面板的仿生复眼图像拼接及定位算法[J]. 兵工学报, 2018, 39(6): 1144-1150.
[13]	曾溢良，蓝金辉，邹金霖. 滑动置信度约束的红外弱小目标跟踪算法研究[J]. 兵工学报, 2017, 38(9): 1771-1778.
[14]	黄莹，郝新红，孔志杰，张彪. 基于熵特征的调频引信目标与干扰信号识别[J]. 兵工学报, 2017, 38(2): 254-260.
[15]	王曙光，曾向阳，王征，王强. 水下目标的Gammatone子带降噪和希尔伯特-黄变换特征提取[J]. 兵工学报, 2015, 36(9): 1704-1709.

紧凑折反式仿生复眼及图像快速拼接识别算法

Compact Catadioptric Bionic Compound Eye and Fast Image Mosaic Recognition Algorithm

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价