欢迎访问《兵工学报》官方网站,今天是 分享到:

兵工学报 ›› 2022, Vol. 43 ›› Issue (12): 3062-3069.doi: 10.12382/bgxb.2021.0702

• 论文 • 上一篇    下一篇

基于局部大气修复的偏振导航方法

李宇阳1,2,王霞1,2,赵家碧1,2,李磊磊3,陈家斌3   

  1. (1.北京理工大学 光电学院,北京 100081;2.北京理工大学 光电成像技术与系统教育部重点实验室,北京 100081;3.北京理工大学 自动化学院,北京 100081)
  • 上线日期:2022-05-19
  • 作者简介:李宇阳(1998—), 女, 硕士研究生。 E-mail: yuweiyang1019@163.com
  • 基金资助:
    装备重大基础研究项目(5140505B03)

Polarization Navigation Algorithm Based on Local Atmosphere Image Restoration

LI Yuyang1,2, WANG Xia1,2, ZHAO Jiabi1,2, LI Leilei3, CHEN Jiabin3   

  1. (1.School of Optics and Photonics, Beijing Institute of Technology, Beijing 100081, China; 2.Key Laboratory of Optoelectronic Imaging Technology and System, Ministry of Education, Beijing Institute of Technology, Beijing 100081, China; 3.School of Automation, Beijing Institute of Technology, Beijing 100081, China)
  • Online:2022-05-19

摘要: 仿生偏振导航是一种利用天空光的偏振分布模式信息进行导航的方法,具有抗干扰性能好、误差不随时间积累等优点,受到了广泛的关注。但是其在图像采集过程中视野容易受到遮挡,导致图像信息丢失进而严重影响导航精度。针对上述问题,提出一种基于局部大气修复的偏振导航方法,建立基于图像相似块偏移量统计特性的修复模型。通过计算出天空中每一个图像块对应的偏移量直方图,得到每个像素点对应的偏移量,根据偏移量进行图像的修复。对修复后的天空图像进行处理,解算天空偏振分布模式,算出太阳方位角,实现导航的目的。仿真与实验结果表明:该算法能适应视野内部分遮挡环境,去除视野中的随机遮挡,恢复图像细节信息,部分遮挡条件下航向角误差降低至1°以内,比未修复前精度提高了约61%。

关键词: 偏振成像, 仿生导航, 图像修复, 相似块偏移量

Abstract: Bionic polarized light navigation is a method of navigation that uses skylight polarization patterns. Due to its strong anti-interference ability and high navigation accuracy, it has attracted extensive attention. However, the field of view is easily blocked during image acquisition, resulting in the loss of image information, which will seriously affect the navigation accuracy. To solve the problem, a polarization navigation algorithm based on local atmosphere image restoration is proposed, and an image inpainting model based on the offsets of similar patches is established. By calculating the nearest neighbor offset histogram of each patch in the image, the offset of each pixel can be obtained, and the image repaired accordingly. Using the restored images, solar azimuth can be determined by detecting the polarization distribution pattern of the atmosphere in the zenith region to achieve the purpose of navigation. The results of simulation and experiments show that the algorithm can adapt to a partially blocked view, remove random occlusion and improve image details. With partial occlusion, the heading angle error is reduced to about 1°, and the accuracy is improved by about 61%.

Key words: polarizationimaging, bionicnavigation, imageinpainting, offsetsofsimilarpatches

中图分类号: