This paper designs and implements a real-time image mosaic method for UAV, which can mosaic UAV real-time video stream online. This paper first filters and cleans the real-time image transmitted by UAV by the distortion degree and distortion continuity index of transmission transformation, then extracts the image into gray scale image, and creates the SURF feature description of the image, and then use the optimization algorithm to match the SURF feature among the continuous images, and then use the random sampling consistency algorithm to filter the matching, According to the matching relationship between the images, the homography transformation matrix is calculated, and the homography transformation is used to complete the image mosaic, and finally the image fusion is carried out. The UAV image mosaic method proposed in this paper improves the practicability of traditional image mosaic methods, and improves the timeliness of UAV image mosaic to a certain extent.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

UAV Real Time Image Mosaic Method Based on SURF Algorithm

  • Jian Liu,
  • Zihan Yuan,
  • Guoqi Zeng,
  • Minqiang Zhao,
  • Zifan Niu,
  • Delin Hao

摘要

This paper designs and implements a real-time image mosaic method for UAV, which can mosaic UAV real-time video stream online. This paper first filters and cleans the real-time image transmitted by UAV by the distortion degree and distortion continuity index of transmission transformation, then extracts the image into gray scale image, and creates the SURF feature description of the image, and then use the optimization algorithm to match the SURF feature among the continuous images, and then use the random sampling consistency algorithm to filter the matching, According to the matching relationship between the images, the homography transformation matrix is calculated, and the homography transformation is used to complete the image mosaic, and finally the image fusion is carried out. The UAV image mosaic method proposed in this paper improves the practicability of traditional image mosaic methods, and improves the timeliness of UAV image mosaic to a certain extent.