Noisy CS-MRI reconstruction method for pyramid decomposition and dictionary learning
A dictionary learning and tower technology, applied in the field of compressed sensing and medical image processing, can solve the problem of blurring the edges and details of the image, and achieve the effect of improving visual effects, removing image noise, and accurately retaining
Inactive Publication Date: 2014-03-12
王勇
View PDF0 Cites 12 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0006] The technical problem solved by the present invention is to provide a method to solve the blurring of image edges and details caused by the traditional MRI denoising algorithm, which can use the combination of LP tower multi-scale decomposition and adaptive training of biorthogonal basis dictionary learning A compressive sensing MRI reconstruction method based on tower decomposition and dictionary learning
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
[0032] Noisy CS-MRI image reconstruction method of the present invention, such as figure 1 shown. First, the Laplacian Pyramid (LP) filter is used to decompose the MRI image at multiple scales. One LP decomposition decomposes the original MRI signal into low-frequency components and high-frequency components, and recursively decomposes the low-frequency components to make it Obtain the entire multi-resolution image; secondly, combine the K-SVD adaptive training and learning algorithm to sparsely represent the high-frequency components of each layer; then perform LP inverse transformation on the learned high-frequency signal and the low-frequency signal of the same layer to obtain the next The low-frequency information of the layer, until the lowest layer; finally, the LP inverse transform is performed on the image data of the first layer to obtain the final noise-reduced image, so as to achieve the purpose of image noise reduction.
[0033] The specific steps are:
[0034] ①...
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 invention discloses a noisy CS-MRI reconstruction method for pyramid decomposition and dictionary learning. The method comprises the following steps: firstly, a Laplacian pyramid (Laplacian Pyramid, LP) is adopted for decomposing an input noisy medical image MIR in a multi-scale mode, a noisy MIR signal is decomposed into low-frequency components and high-frequency components through primary LP decomposition, and the low-frequency components are decomposed recursively so as to obtain a whole multi-resolution image; secondly, the high-frequency components of each layer are sparsely represented in combination with a K-SVD adaptive training learning algorithm; thirdly, LP inverse transformation is simultaneously carried out on learned high-frequency signals and low-frequency signals of the same layer to obtain low-frequency information of a lower layer until the lowest layer; and finally, LP inverse transformation is simultaneously carried out on first-layer image data to obtain a de-noised medical image. The de-noising method provided by the invention is suitable for processing MRI images containing more Gauss noise, can better preserve edge information and detail information of images and effectively improve the visual effect and is very important for medical diagnosis, treatment and follow-up work.
Description
technical field [0001] The invention belongs to the fields of compressed sensing and medical image processing, and in particular relates to a method for denoising and accelerating noisy MRI images using compressed sensing using tower decomposition and dictionary learning. technical background [0002] Magnetic resonance imaging (also known as magnetic resonance imaging, Magnetic Resonance Imaging, MRI) is an important disease diagnosis technology applied in clinical medicine in recent years. It can obtain images of important parts of the human body, and its high-quality imaging is of great significance for practical applications. [0003] The main factor affecting the quality of MRI imaging comes from noise. Noise will blur the boundaries of some tissues in MRI images, making it difficult to distinguish fine structures. At the same time, the noise also limits the further improvement of MRI signal-to-noise ratio and sensitivity, and affects the imaging quality. Newman once...
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 IPC IPC(8): G06T5/00
Inventor 王勇张凤郑娜王灿陈楚楚高全学许录平
Owner 王勇