Optimization method based on unmanned aerial vehicle camera image stitching algorithm
An optimization method based on an unmanned aerial vehicle photographed image stitching algorithm comprises the following steps: step 1, finding an intersection area of a plurality of photos, and taking the intersection area as a target area for feature point detection; 2, feature points are extract...
Gespeichert in:
Hauptverfasser: | , , , , , |
---|---|
Format: | Patent |
Sprache: | chi ; eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | An optimization method based on an unmanned aerial vehicle photographed image stitching algorithm comprises the following steps: step 1, finding an intersection area of a plurality of photos, and taking the intersection area as a target area for feature point detection; 2, feature points are extracted from the intersection area of the two images through an SIFT algorithm; step 3, adding an epipolar constraint method on the basis of original BF violent matching to carry out rough matching of feature point pairs; 4, performing fine purification on the feature point matching pairs by combining a grid-based motion statistics GMS algorithm and an RANSAC algorithm so as to obtain a more accurate spliced image; 5, aligning the two images by using the homography transformation matrix; and step 6, smoothing the spliced image by adopting a weighted average fusion algorithm to obtain a complete panorama. According to the method, the problems of low registration precision, low recovery rate and the like can be effectivel |
---|