Template image global searching method based on mask matrix and fast Fourier transform
A Fourier transform, template image technology, applied in the field of image processing, can solve the problem of high computational complexity, achieve the effect of overcoming complexity, improving computational speed, and saving processing time
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0046] Example 1
[0047] Aiming at the problems of high algorithm complexity in the template search method in the field of digital image processing, repeated calculations cannot be used multiple times, and loss of certain calculation accuracy, the present invention proposes a global search of template images based on a mask matrix and fast Fourier transform Method, see figure 1 , Including the following steps:
[0048] (1) Copy template image and full image: read the original template image and full image data, where the original template image is as follows image 3 , Its size is smaller than the full image, the full image is the image of lena in the gallery, see figure 2 .
[0049] (2) Use the original template image to generate a mask image:
[0050] (2a) Generate a template with the original Figure one A full array of sample size is the mask image;
[0051] (2b) When there is no need to match the target background, the element in the mask matrix representing the position of the ...
Example Embodiment
[0076] Example 2
[0077] The global search method of the template image based on the mask matrix and the fast Fourier transform is the same as in embodiment 1. The template expansion image and the full image described in step (6) are used to perform circular convolution using the following formula:
[0078] c o r r = F 2 - 1 { C [ F 2 ( t _ e x t ) ] . * F 2 ( i m g ) }
[0079] Among them, F -1 Stands for inverse fast Fourier transform, C stands for conjugate, F stands for fast Fourier transform, subscript 2 stands for two-dimensional transform, t_ext stands for template expansion image, img stands for full image, corr stands for output template full image correlation matrix.
[0080] The actual project can be implemented using matlab language. If it is implemented using matlab, the above formula can be expressed as a function:
[0081] corr=ifft2(conj(fft2(t_ext)).*fft2(img))
[0082] Among them, fft2 is...
Example Embodiment
[0087] Example 3
[0088] The global search method of the template image based on the mask matrix and the fast Fourier transform is the same as in the embodiment 1-2, and the calculation of the energy matrix of the whole image described in step (7) is expressed as follows:
[0089] i m g _ e n e r g y = F 2 - 1 { C [ F 2 ( m a s k _ e x t ) ] . * F 2 ( i m g . * i m g ) }
[0090] Among them, F -1 Stands for inverse fast Fourier transform, C stands for conjugate, F stands for fast Fourier transform, subscript 2 stands for two-dimensional transform, mask_ext stands for mask extension, img stands for full image, img_energy stands for output full image energy matrix.
[0091] If you use matlab language, the above formula can be expressed as:
[0092] img_energy=ifft2(conj(fft2(mask_ext)).*fft2(img.*img))
[0093] Among them, fft2 is the fast two-dimensional Fourier transform functio...
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