Image sharpness lifting method based on sparse expression
A technology of image clarity and sparse representation, which is applied in the field of image processing to achieve the effect of enriching detailed information, clear details, and improving image clarity
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0027] Embodiment 1: A method for improving image clarity based on sparse representation, the specific steps of the method are as follows:
[0028] Step1, the input two pieces of CT and MRI image that pixel size is 256*256 (as image 3 (a), 3(b)), perform low-rank decomposition respectively, and obtain sparse partial images and low-rank partial images respectively; (that is, obtain a low-rank partial image A after CT image decomposition 1 and a sparse partial image A 2 , the MRI image is decomposed to obtain a low-rank partial image B 1 and a sparse partial image B 2 );
[0029] Step2, use the dictionary learning model to select the image set Y (such as Figure 8 As shown, using a high-resolution non-medical image set, this embodiment selects 6 pictures to construct an image set) for training, and obtains a low-rank dictionary D L and a sparse dictionary D S ; The dictionary learning model is:
[0030]
[0031] s.t.||Z S || 0 ≤T 0 ,||Z L || 0 ≤T 1
[0032] whe...
Embodiment 2
[0036] Embodiment 2: a method for improving image clarity based on sparse representation, the specific steps of the method are as follows:
[0037] Step1, input two noisy CT and MRI images with a pixel size of 256×256 (such as image 3 (c), 3(d)), perform low-rank decomposition respectively, and obtain sparse partial images and low-rank partial images respectively; (that is, after CT image decomposition, a low-rank partial image A 1 and a sparse partial image A 2 , the MRI image is decomposed to obtain a low-rank partial image B 1 and a sparse partial image B 2 );
[0038] Step2, use the dictionary learning model to select the image set Y (such as Figure 8 As shown, using a high-resolution non-medical image set, this embodiment selects 6 pictures to construct) for training, and obtains a low-rank dictionary D L and a sparse dictionary D S ; The dictionary learning model is:
[0039]
[0040] s.t.||Z S || 0 ≤T 0 ,||Z L || 0 ≤T 1
[0041] where Y is denoted as ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


