Sequence image splicing method and system of low-altitude unmanned vehicle
A technology of unmanned aerial vehicles and sequence images, which is applied in image communication, TV system components, TV, etc., and can solve the problems of sequence image error accumulation and so on
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0042] Example one
[0043] In this embodiment, SURF feature points and HARRIS-AFFINE feature points, as well as the RANSAC fault-tolerant algorithm and epipolar geometric constraints are used to solve the homography matrix to improve the quality of splicing.
[0044] figure 1 Shown is a flowchart of a method for splicing a sequence image of a low-altitude unmanned aerial vehicle according to the first embodiment of the present invention, including:
[0045] Step 11: Correction of camera distortion.
[0046] Step 12, the unmanned aerial vehicle sequence image acquisition.
[0047] Step 13, image correction, determining the homography matrix between adjacent images, including the following sub-steps:
[0048] Step 131, feature extraction, extract SURF feature points and HARRIS-AFFINE feature points from the sequence image;
[0049] Step 132, image matching, includes the following sub-steps:
[0050] Step 1321: Use Mahalanobis distance to calculate the similarity of feature points between ad...
Example Embodiment
[0136] Example two
[0137] The difference from the first embodiment is that the second embodiment adds a matching verification step. Although the first embodiment calculates the homography matrix and matches the image, whether the matching result is correct requires further verification. Especially for the image sequence acquired by the aircraft, bad images such as complete yaw are mixed inside, which need to be automatically removed during stitching.
[0138] figure 2 Shown is a flowchart of a method for splicing serial images of a low-altitude unmanned aerial vehicle according to the second embodiment of the present invention, including:
[0139] Step 21: Correction of camera distortion.
[0140] Step 22: Acquire serial images of the unmanned aerial vehicle.
[0141] Step 23, select an image in the sequence of images.
[0142] Step 24, image correction, determining the homography matrix between adjacent images, includes the following sub-steps:
[0143] Step 241, feature extraction, ...
Example Embodiment
[0166] Example three
[0167] The difference from the second embodiment is that the third embodiment further adds the steps of global optimization and adjustment of the homography matrix.
[0168] image 3 Shown is a flowchart of a method for splicing a sequence image of a low-altitude unmanned aerial vehicle according to the third embodiment of the present invention, including:
[0169] Step 31: Correction of camera distortion.
[0170] Step 32, the unmanned aerial vehicle sequence image acquisition.
[0171] Step 33: Select an image in the sequence of images.
[0172] Step 34, image correction, determining the homography matrix between adjacent images, includes the following sub-steps:
[0173] Step 341, feature extraction, extract SURF features and HARRIS-AFFINE features from the sequence image;
[0174] Step 342, image matching, includes the following sub-steps:
[0175] Step 3421: Use Mahalanobis distance to calculate the similarity of feature points between adjacent images for initial...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap