a New Approach To Fast Mosaic Uav Images

  • Liu Q
  • Liu W
  • Zou L
 et al. 
  • 31


    Mendeley users who have this article in their library.
  • 6


    Citations of this article.


Unmanned Aerial Vehicles (UAVs) have been widely used to acquire high quality terrain images of the areas of interest, particularly when such a task could potentially risk human life or even impossible as the areas cannot be accessed easily by surveyors. Once the images have been obtained, traditional photogrammetric processing process can be used to establish a relative orientation model and then, absolute orientation model with the procedures of space resection and intersection. In many such applications, the geo- referenced images which are stitched together to represent the geospatial relationships for the feature objects are sufficient. A fast or near real-time processing approach for UAV images using GPS/INS data has being investigated for years. One beneficial application of such approach is the capability of quick production of geo-referenced images for various engineering or business activities, such as urban and road planning, the site selection of factories and bridges, etc. In this paper, we have proposed a new fast processing approach for the UAV images collected with an integrated GPS/INS/Vision system. The approach features that the corresponding points between images have been determined, and then coordinate transformation is carried out to implement image stitching. The accuracy of corresponding points normally affects the quality of stitched images, but the results of our experiments revealed that the image stitching errors were obvious even the accuracy of corresponding points was high. The stitching errors could be caused by the changes of surface elevation.

Author-supplied keywords

  • coordinate transformations
  • displacement of images
  • image mosaic errors
  • mosaic uav images

Get free article suggestions today

Mendeley saves you time finding and organizing research

Sign up here
Already have an account ?Sign in

Find this document


  • Q. Liu

  • W. Liu

  • L. Zou

  • J. Wang

  • Y. Liu

Cite this document

Choose a citation style from the tabs below

Save time finding and organizing research with Mendeley

Sign up for free