Super-resolution reconstruction method based on compound motion and adaptive non-local prior
A technology of super-resolution reconstruction and compound motion, which is applied in the field of super-resolution image reconstruction based on compound motion and adaptive non-local prior, and can solve the problem of not being able to achieve parameter adaptation
Active Publication Date: 2017-02-01
CAS OF CHENGDU INFORMATION TECH CO LTD
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, since there are many parameters that need to be adjusted manually in the non-local prior, parameter adaptation cannot be achieved.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0039] The specific implementation manner of the present invention will be described in detail below in conjunction with the accompanying drawings.
[0040] The super-resolution reconstruction of the image is to restore the observed image to an ideal image. The observed image is a series of low-resolution images, and the ideal image is the desired high-resolution image. Given a high-resolution image X of a certain scene, after a series of geometric motion, optical blur, sub-sampling and additional noise degradation processes, p low-resolution observation images Y are generated k , using a commonly used image observation model to describe the relationship between the ideal image and the observed image, the observation model is: Y k =DB k m k X+n k , k=1,...,p, where M k is the motion change matrix, B k is the fuzzy matrix, D is the downsampling matrix, n k for additional noise.
[0041] Based on the above observation model, the present invention uses the maximum a poste...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The present invention provides a super-resolution reconstruction method based on compound motion and self-adaptive non-local prior, comprising the following steps: selecting reference frame images and non-reference frame images from p low-resolution images; using global motion parameters and local The optical flow method is used for image registration to obtain the motion field mk(x) of the non-reference frame image relative to the reference frame image, and use mk(x) to construct the motion transformation matrix Mk; calculate the interpolation image of the reference frame image, non-local prior Parameters hi, j and Euclidean threshold; calculate the similarity weight wNLM[i, j; s, t] between each pixel and other pixels, and use wNLM to construct a non-local weight matrix S about the high-resolution image X; use the motion transformation matrix Mk and the non-local weight matrix S solve the target functional to obtain the reconstructed high-resolution estimated image. Compared with the prior art, the present invention effectively solves the disadvantages of large amount of calculation, poor scalability, and low precision of current motion estimation by adopting a compound motion model, and adopts adaptive non-local prior to reduce reconstruction images. distortion.
Description
technical field [0001] The invention relates to the technical fields of image processing and computer vision, in particular to a super-resolution image reconstruction method based on compound motion and self-adaptive non-local a priori which improves the spatial resolution of images in the field of image clarity. Background technique [0002] In the field of imaging, images with high spatial resolution have always been one of the pursued goals. Images with high spatial resolution fully record the detailed information of objects, and can provide richer information for human and computer reasoning, judgment, and decision-making. Therefore, in many imaging applications, high-resolution images are usually very necessary, such as: video surveillance, medical diagnosis, military reconnaissance, remote sensing and other applications. There are two ways to improve the spatial resolution of images, "hardware way" and "software way". Under normal circumstances, people mainly obtain ...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06T5/50
Inventor 陈帅陈斌何易德赵雪专
Owner CAS OF CHENGDU INFORMATION TECH CO LTD