Alignment method for double camera images for hyperspectral imaging platform
A hyperspectral image and image alignment technology, applied in the field of computational photography, can solve problems such as the lack of alignment methods for high-resolution RGB images and low-resolution hyperspectral images, and achieve the effect of improving accuracy and accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0051] A method for dual-camera image alignment in a hyperspectral imaging platform disclosed in this embodiment is applied to an image fusion platform of hyperspectral imaging, using a hyperspectral camera to acquire a low-resolution hyperspectral image of a scene, and simultaneously using an RGB camera to acquire High-resolution RGB images of the same scene; spatial downsampling of RGB images to obtain low-resolution RGB images, and spectral domain downsampling of hyperspectral images to obtain low-resolution RGB images of the same size as spatial downsampling; Two low-resolution RGB images establish an image alignment model; use an alignment model solving algorithm to solve the homography transformation matrix in the alignment model; use the homography transformation matrix to transform the high-resolution RGB image so that the high-resolution The RGB image is aligned with the low-resolution hyperspectral image, thereby improving the accuracy of the hyperspectral imaging alg...
Embodiment 2
[0069] This embodiment also discloses a method for image alignment of dual cameras in a hyperspectral imaging platform, which is applied to an image fusion platform of hyperspectral imaging, using a hyperspectral camera to acquire a low-resolution hyperspectral image of a scene, and simultaneously using an RGB camera to acquire High-resolution RGB images of the same scene; the homography transformation matrix is initialized as an identity matrix; iteratively carry out the following steps of hyperspectral reconstruction and alignment until a preset number of iterations is reached: the high-resolution RGB image and the low-resolution The hyperspectral image is reconstructed using the image fusion algorithm to obtain the reconstructed hyperspectral image; the reconstructed hyperspectral image is spatially down-sampled; the collected low-resolution hyperspectral image and the reconstructed hyperspectral image after spatial downsampling image, establish an alignment model; use an ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More