Method for matching feature points among multiple images
A feature point matching, multi-image technology, applied in image analysis, image data processing, graphic image conversion and other directions, can solve the problems of large matching calculation and high mismatch rate, achieve broad application prospects, avoid the probability of mismatching, The effect of reducing computational complexity
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0035] A method for matching feature points between multiple images involved in this embodiment has the following specific process steps:
[0036] Step 1: Calculate the rotation matrix of each image using the mobile phone gyroscope data:
[0037] The world coordinate system is established with the true north direction parallel to the ground as the X-axis, the vertical direction of the X-axis parallel to the ground as the Y-axis, and the direction perpendicular to the ground as the Z-axis, and the world coordinate system is a left-handed coordinate system; The system is a local coordinate system, the mobile phone screen is the X-Y plane, the upper left corner of the mobile phone screen is the coordinate origin, the screen height is the X axis, the width is the Y axis, the main optical axis of the mobile phone is the Z axis, and is a left-handed coordinate system; read from the gyroscope Take the angle between the mobile phone coordinate axis and the world coordinate axis, and s...
Embodiment 2
[0056] This embodiment applies the feature point matching method between multiple images described in Embodiment 1 to match real images captured from multiple angles, such as figure 2 As shown in the figure, the black part in the middle is the three-dimensional object to be photographed, the multiple gray-white pyramids represent the selected shooting angles in various directions, and the images taken at adjacent angles are similar;
[0057] After using the existing image matching method to process the captured image, the image pair that needs to be matched is obtained, such as image 3 After processing the captured pictures by using the feature point matching method between multiple images of the present invention, the image pairs that need to be matched are obtained as follows: Figure 4 shown; in image 3 and Figure 4 , each point represents an image pair, by putting image 3 and Figure 4 It can be seen from the comparison that, after processing the captured pictures...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


