CN117788461A - Magnetic resonance image quality evaluation system based on image analysis - Google Patents

Abstract

The invention discloses a magnetic resonance image quality assessment system based on image analysis, which belongs to the technical field of liver magnetic resonance imaging quality assessment and includes a platform end, a user end and a server; the user end includes an acquisition module, a confidentiality module and a transmission module; the acquisition module Used to collect the target image and generate the reference image corresponding to the target image; send the target image and the reference image to the confidentiality module for confidentiality processing of the target image and the reference image, and send the processed target image and reference image through the transmission module To the platform side; the transmission module is used for data transmission between the user side and the platform side; the platform side includes the first sub-item module, the second sub-item module and the quality assessment module; the first sub-item module is used to compare the target image and reference The image is analyzed and the first sub-item value is calculated; the second sub-item module is used to analyze the target image and calculate the second sub-item value; the quality assessment module is used to evaluate the quality score of the target image.

Description

A magnetic resonance image quality assessment system based on image analysis

Technical field

The invention belongs to the technical field of liver magnetic resonance imaging quality assessment, and is specifically a magnetic resonance image quality assessment system based on image analysis.

Background technique

Liver magnetic resonance imaging is a non-invasive, non-radiation imaging method that is widely used in the diagnosis and monitoring of liver diseases. However, liver MRI faces some technical challenges in practical applications, especially interference from multiple noises and artifacts. These include, but are not limited to, motion artifacts, metallic artifacts, and image distortion caused by magnetic field inhomogeneities. These problems can seriously affect the quality of images, making diagnosis and disease assessment difficult for doctors. Real-time and dynamic assessment of the imaging quality of magnetic resonance images is the focus of quality control work in hospital radiology departments. Currently, hospitals rely primarily on professional radiologists or reviewers to evaluate images. They mainly make subjective ratings based on factors such as the signal-to-noise ratio, clarity and detail performance of the image. While this method can assess image quality, it involves considerable subjectivity and is labor-intensive. Additionally, this subjective assessment method may lead to consistency differences between assessment results.

Therefore, based on this, the present invention provides a magnetic resonance image quality assessment system based on image analysis.

Contents of the invention

In order to solve the problems existing in the above solution, the present invention provides a magnetic resonance image quality assessment system based on image analysis.

The object of the present invention can be achieved through the following technical solutions:

A magnetic resonance image quality assessment system based on image analysis, including a platform end, a user end and a server; communication connections between the platform end, the user end and the server;

The user terminal includes a collection module, a security module and a transmission module;

The acquisition module is used to collect the corresponding magnetic resonance image, mark it as a target image, and generate a reference image corresponding to the target image; send the target image and the reference image to the security module.

The security module is used to perform security processing on the received target image and reference image, and send the processed target image and reference image to the platform through the transmission module.

Further, the method for confidentiality processing of the target image and the reference image includes:

Mark the received target image and reference image as images to be processed; preset confidentiality items and corresponding identification features; and set the confidentiality processing method corresponding to each of the confidentiality items;

Identify the image to be processed according to each confidential item and the corresponding identification feature, identify the corresponding private information; match the corresponding confidentiality processing method for each of the private information, and process the private information according to the obtained confidentiality processing method. Perform processing; obtain the target image and reference image after confidentiality processing.

The transmission module is used for data transmission between the user terminal and the platform terminal.

The platform end includes a first sub-item module, a second sub-item module and a quality assessment module;

The first sub-item module is used to perform comparative analysis on the target image and the reference image, and calculate the corresponding first sub-item value.

Further, the calculation method of the first component value includes:

Perform equivalent processing and grayscale processing on the target image and the reference image; obtain the corresponding target grayscale image and reference grayscale image;

Identify the grayscale value of each pixel in the target grayscale image and the reference grayscale image; generate the corresponding target grayscale matrix and reference grayscale matrix according to the grayscale value of each pixel and the corresponding position; convert the target grayscale The grayscale matrix and the reference grayscale matrix are respectively marked as and/> ; i=1, 2,..., n, n is a positive integer; j=1, 2,..., m, m is a positive integer;

Calculate the corresponding difference matrix according to the formula R ₃ =R ₁ -R ₂ ;

In the formula: R ₁ is the target gray level matrix; R ₂ is the reference gray level matrix; R ₃ is the difference matrix;

According to the formula Calculate the corresponding first component value; where: SUB _value is the first component value; ||R ₃ || _F represents the Frobenius norm of matrix R ₃ .

The second sub-item module is used to analyze the target image and calculate the corresponding second sub-item value.

Furthermore, the calculation method of the second sub-item value includes:

Acquire a target image, identify the size of the target image, divide the target image into equal parts according to the identified size of the target image, and obtain K unit blocks, wherein k' is a positive integer in the formula k'=K ^0.5 , 2≤k'≤10; K is a positive integer;

Identify the contrast representative value, resolution representative value and brightness representative value corresponding to each of the unit blocks; generate the corresponding contrast matrix, resolution matrix and brightness matrix based on the position of each of the unit blocks; the contrast matrix, resolution matrix and brightness The matrices are all square matrices of order k´;

Calculate the corresponding merge matrix according to the formula R _s =b ₁ ×R ₄ +b ₂ ×R ₅ +b ₃ ×R ₆ ;

In the formula: R _s is the merging matrix; R ₄ is the contrast matrix; R ₅ is the resolution matrix; R ₆ is the brightness matrix; b ₁ , b ₂ , b ₃ are all proportional coefficients, and the value range is 0<b ₁ ≤ 1, 0<b ₂ ≤1, 0<b ₃ ≤1;

Mark the merged matrix as ; Mark the elements in the merged matrix as s _qp ; q=1, 2, ..., k´, k´ is a positive integer; p = 1, 2, ..., k´, k´ is a positive integer;

According to the formula Calculate the corresponding second component value;

Where: SRB _value is the second sub-item value.

Further, methods for dividing the target image into equal parts include:

Perform simulated segmentation on the target image according to the value range of k´, obtain 9 groups of segmented images, and mark a single image in each group of segmented images as a unit block;

Evaluate the contrast representative value, resolution representative value and brightness representative value corresponding to the unit blocks in each group of segmented images; when any contrast representative value, resolution representative value and brightness representative value in each group of segmented images is 0, remove the segmented images of the corresponding group;

Calculate the segmentation evaluation value of each group of segmented images through a preset segmentation evaluation formula, identify the segmented image corresponding to the highest segmentation evaluation value, and segment the target image according to the segmentation method corresponding to the segmented image.

Further, the evaluation methods of contrast representative values include:

Identify the value of K corresponding to the segmented image, divide each unit block equally according to K, and obtain K subdivided blocks; identify the contrast corresponding to each subdivided block and mark it as a single contrast;

According to the formula Calculate the contrast representative value of the corresponding unit block;

In the formula: CON _t is the contrast representative value of the corresponding unit block; t represents the corresponding unit block, t=1, 2,...,K; v represents the corresponding subdivision block, v=1, 2,...,K ; DC _tv represents the single contrast corresponding to the subdivided block; G(A) is the judgment model, and A is the area of the subdivided block.

Further, methods for obtaining the resolution representative value include:

Identify the value of K corresponding to the segmented image, divide each unit block equally according to K, and obtain K subdivided blocks; identify the corresponding resolution of each subdivided block and mark it as a single resolution;

According to the formula Calculate the resolution representative value of the corresponding unit block;

In the formula: RE _t is the representative value of resolution of the corresponding unit block; t represents the corresponding unit block, t=1, 2,...,K; v represents the corresponding subdivision block, v=1, 2,..., K; DR _tv represents the single resolution of the corresponding subdivision block; G(A) is the judgment model, and A is the area of the subdivision block.

Further, the method for obtaining the brightness representative value includes:

Identify the value of K corresponding to the segmented image, divide each unit block equally according to K, and obtain K subdivided blocks; identify the brightness corresponding to each subdivided block and mark it as a single brightness; obtain the corresponding reference brightness, and calculate the brightness rate of the corresponding subdivided block according to brightness rate = single brightness ÷ reference brightness;

According to the formula Calculate the brightness representative value of the corresponding unit block;

Wherein: BRIGH _t is the brightness representative value of the corresponding unit block; t represents the corresponding unit block, t=1, 2, ..., K; v represents the corresponding subdivision block, v=1, 2, ..., K; DB _tv represents the single resolution of the corresponding subdivision block; G(A) is the judgment model, and A is the area of the subdivision block.

Furthermore, the expression of the judgment model is .

Furthermore, the segmentation evaluation formula is ;

In the formula: A _sse is the segmentation evaluation value; b ₁ , b ₂ , and b ₃ are all proportional coefficients, and the value range is 0<b ₁ ≤1, 0<b ₂ ≤1, 0<b ₃ ≤1; t represents Corresponding unit block, t=1, 2,...,K; CON _t is the representative value of contrast of the corresponding unit block; RE _t is the representative value of resolution of the corresponding unit block; BRIGH _t is the representative value of brightness of the corresponding unit block.

The quality assessment module is used to assess the quality score of the target image, obtain the first sub-item value and the second sub-item value, and calculate the corresponding comprehensive value according to the formula: _Prior = _b4 × SUB _value + _b5 × SRB _value ;

In the formula: _Prior is the comprehensive value; SUB _value is the first component value; SRB _value is the second component value; b ₄ and b ₅ are both proportional coefficients, and the value range is 0<b ₄ ≤1, 0< b ₅ ≤1;

Acquire corresponding training data, analyze the training data, and obtain a corresponding comprehensive value set; form a number of scoring coordinates according to the quality score and the comprehensive value set corresponding to the training data; and set a corresponding evaluation function according to each of the scoring coordinates;

The comprehensive value is substituted into the evaluation function to calculate the corresponding quality score; and the quality score is sent to the user terminal.

Compared with the prior art, the beneficial effects of the present invention are:

Through the mutual cooperation between various modules, intelligent evaluation of magnetic resonance images can be achieved and corresponding quality scores can be obtained. Through the separation between the platform side and the user side, the role of invested resources will be maximized, which will greatly benefit Reduce the hospital's cost investment; achieve one-to-many, providing services to multiple user hospitals; and through the management and operation and maintenance of the platform, there will be more professional staff to ensure the normal operation of the system; and by setting up a confidentiality module to protect patients Privacy, to prevent the leakage of patient privacy; by evaluating the quality of liver magnetic resonance images, we can understand the advantages and disadvantages of image processing algorithms, further optimize the image processing process, and improve the effect and efficiency of image processing; at the same time, we can provide doctors with more accurate and reliable The diagnostic basis helps doctors make more scientific and reasonable clinical decisions.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following will briefly introduce the drawings needed to describe the embodiments or the prior art. Obviously, the drawings in the following description are only These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without exerting creative efforts.

Figure 1 is a functional block diagram of the present invention;

Figure 2 is a flow chart of the method of the present invention.

Detailed ways

The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

As shown in FIG. 1 and FIG. 2 , a magnetic resonance image quality assessment system based on image analysis includes a platform end, a user end and a server; the platform end, the user end and the server are communicatively connected.

The platform side is used by the platform side for application, management and maintenance.

The client is used by registered hospital users.

In order to achieve the quality assessment of magnetic resonance images, corresponding equipment and software are required. In terms of hospital needs and transformation costs, if the hospital invests and uses it independently, it will be difficult to use all the invested resources, resulting in a waste of resources; therefore, Adopting the relationship between the platform and users for management will maximize the utilization of invested resources and greatly reduce the hospital's cost investment; realize one-to-many and provide services to multiple user hospitals; and through the platform For management and operation and maintenance, there will be more professional staff to ensure the normal operation of the system.

The user terminal includes a collection module, a confidentiality module, and a transmission module;

The acquisition module is used to collect the corresponding magnetic resonance image when quality assessment is required, mark it as a target image, and generate a reference image of the target image; send the obtained target image and reference image to the confidentiality module.

The reference image refers to a clear, noise-free, and distortion-free magnetic resonance image similar to the target image. When acquiring the target image, the corresponding image set can be collected, and the collected images can be selected for preprocessing, including denoising, correction, Normalization and other operations are performed to ensure the quality and stability of the image; a high-quality original magnetic resonance image is selected from the pre-processed image as a reference image. Or you can also use other existing methods to obtain the reference image.

The security module is used to perform security processing on the received target image and reference image, and mark the received target image and reference image as images to be processed; preset security items and corresponding identification features; that is, the magnetic resonance image may contain Data items including patient privacy information are combined with the identification features of the data items in the image to set identification features corresponding to each confidential item;

Preset the confidentiality processing methods corresponding to each confidentiality item, such as deleting corresponding information, adapting information, code replacement and other confidentiality methods; you can select the confidentiality processing method according to your needs if there is no corresponding method for image quality evaluation to be processed.

Identify the image to be processed according to the preset confidentiality items and corresponding identification features, and identify the corresponding private information; match the corresponding confidentiality processing method for each private information, and process the corresponding private information according to the obtained confidentiality processing method; obtain The target image and reference image after confidentiality processing are sent to the platform through the transmission module.

The transmission module is used for data transmission between the user terminal and the platform terminal.

The platform includes a first sub-item module, a second sub-item module and a quality assessment module;

The first sub-item module is used to compare and analyze the target image and the reference image, and calculate the corresponding first sub-item value.

The specific calculation method is as follows:

Perform equivalence processing and grayscale processing on the target image and the reference image; equivalence processing means processing the target image and the reference image into the same size and resolution.

Obtain the target grayscale image and reference grayscale image corresponding to the processed target image and reference image respectively;

Identify the grayscale value of each pixel in the target grayscale image and the reference grayscale image; generate the corresponding target grayscale matrix and reference grayscale matrix according to the corresponding grayscale value and corresponding position of each pixel; convert the obtained target grayscale matrix and the reference grayscale matrix are respectively marked as and/> ; i=1, 2,..., n, n is a positive integer; j=1, 2,..., m, m is a positive integer;

Calculate the corresponding difference matrix according to the formula R ₃ =R ₁ -R ₂ ; in the formula: R ₁ is the target grayscale matrix; R ₂ is the reference grayscale matrix; R ₃ is the difference matrix;

According to the formula Calculate the corresponding first component value; where: SUB _value is the first component value; ||R ₃ || _F represents the Frobenius norm of matrix R ₃ .

The second sub-item module is used to analyze the target image and calculate the corresponding second sub-item value. The process is as follows:

Obtain the target image, identify the target image size, segment the target image according to the recognized target image size, and obtain K unit blocks, and k´=K ^0.5 is a positive integer, 2≤k´≤10;

Identify the contrast representative value, resolution representative value and brightness representative value corresponding to each unit block; generate the corresponding contrast matrix, resolution matrix and brightness matrix based on the position of each unit block; that is, use the position of each unit block as the corresponding element in the matrix bit, fill in the corresponding contrast representative value, resolution representative value or brightness representative value to form the corresponding matrix; the contrast matrix, resolution matrix and brightness matrix are all k´ order square matrices; marked R ₄ and R ₅ respectively. , _R6 ;

Calculate the corresponding merging matrix according to the formula R _s =b ₁ ×R ₄ +b ₂ ×R ₅ +b ₃ ×R ₆ ; where: R _s is the merging matrix; R ₄ is the contrast matrix; R ₅ is the resolution matrix; R ₆ is the brightness matrix; b ₁ , b ₂ , and b ₃ are all proportional coefficients, and the value range is 0<b ₁ ≤1, 0<b ₂ ≤1, 0<b ₃ ≤1; and b in the segmentation evaluation formula ₁ , _b2 , and _b3 have the same proportional coefficients.

Mark the merged matrix as ; Mark the elements in the merged matrix as s _qp ; q=1, 2, ..., k´, k´ is a positive integer; p = 1, 2, ..., k´, k´ is a positive integer;

According to the formula Calculate the corresponding second component value;

In the formula: SRB _value is the second component value.

Methods for segmenting target images include:

Set the segmentation evaluation formula. The segmentation evaluation formula is ;

In the formula: A _sse is the segmentation evaluation value; b ₁ , b ₂ , and b ₃ are all proportional coefficients, and the value range is 0<b ₁ ≤1, 0<b ₂ ≤1, 0<b ₃ ≤1; t represents Corresponding unit block, t=1, 2,...,K; CON _t is the representative value of contrast of the corresponding unit block; RE _t is the representative value of resolution of the corresponding unit block; BRIGH _t is the representative value of brightness of the corresponding unit block.

Divide the target image into equal parts according to the value range of k´, that is, divide it into equal parts according to the number of 2 ² , 3 ² , 4 ² , ..., etc.;

Nine groups of segmented images are obtained, and a single image in each group of segmented images is marked as a unit block; the segmentation evaluation value corresponding to each group of segmented images is evaluated through the segmentation evaluation formula, and the value of K corresponding to the highest segmentation evaluation value is selected for segmentation.

Methods for obtaining the contrast representative value include:

Identify the value of K corresponding to this group of segmented images, divide each unit block equally according to K, and obtain K subdivided blocks; identify the contrast corresponding to each subdivided block, mark it as a single contrast, according to the formula Calculate the contrast representative value of the corresponding unit block; where: CON _t is the contrast representative value of the corresponding unit block; t represents the corresponding unit block, t=1, 2,...,K; v represents the corresponding subdivision block, v =1, 2,...,K; DC _tv represents the single contrast of the corresponding subdivision block; G(A) is the judgment model, A is the area of the subdivision block; the expression of the judgment model is/> .

Methods for obtaining the representative value of resolution include:

Identify the value of K corresponding to the group of segmented images, divide each unit block equally according to K, and obtain K subdivided blocks; identify the corresponding resolution of each subdivided block, mark it as a single resolution, according to the formula Calculate the resolution representative value of the corresponding unit block; where: RE _t is the resolution representative value of the corresponding unit block; t represents the corresponding unit block, t=1, 2,...,K; v represents the corresponding subdivision block , v=1, 2,...,K; DR _tv represents the single resolution of the corresponding subdivision block; G(A) is the judgment model, and A is the area of the subdivision block.

Methods for obtaining representative brightness values include:

Identify the value of K corresponding to the group of segmented images, divide each unit block equally according to K, and obtain K subdivided blocks; identify the brightness corresponding to each subdivided block, and mark it as a single brightness; obtain the highest brightness that may appear in the magnetic resonance image, mark it as a reference brightness, and calculate the brightness rate corresponding to the subdivided block according to the formula brightness rate = single brightness ÷ reference brightness;

According to the formula Calculate the brightness representative value of the corresponding unit block; where: BRIGH _t is the brightness representative value of the corresponding unit block; t represents the corresponding unit block, t=1, 2,...,K; v represents the corresponding subdivision block, v =1, 2,...,K; DB _tv represents the single resolution of the corresponding subdivision block; G(A) is the judgment model, and A is the area of the subdivision block.

The quality assessment module is used to evaluate the quality score of the target image based on the first sub-item value and the second sub-item value, and obtain the first sub-item value and the second sub-item value, according to the formula _Prior =b ₄ ×SUB _value + _{Calculate the corresponding comprehensive value} by _{b 5} The range is 0<b ₄ ≤1, 0<b ₅ ≤1;

Acquire a large number of historical magnetic resonance images, which refers to historical magnetic resonance images that have been rated by professionals and are marked as training data; analyze according to the above steps, obtain several historical comprehensive values, and integrate the historical comprehensive values with the corresponding quality scores is the corresponding scoring coordinate; after a large number of integrations, several scoring coordinates will be obtained. The obtained scoring coordinates are input into the coordinate system. The horizontal axis is the comprehensive value, the vertical axis is the quality score, and the adjacent scoring coordinates in the coordinate system are Connect to form the corresponding evaluation curve, fit the evaluation curve, and obtain the corresponding evaluation function;

Input the obtained comprehensive value into the evaluation function and output the corresponding quality score;

Send the obtained quality score to the transfer module.

The above formulas are all numerical calculations after removing the dimensions. The formula is a formula closest to the real situation obtained by collecting a large amount of data for software simulation. The preset parameters and preset thresholds in the formula are set by those skilled in the field according to the actual situation. Or obtain a large amount of data through simulation.

The above embodiments are only used to illustrate the technical methods of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical methods of the present invention can be modified or equivalently substituted. without departing from the spirit and scope of the technical method of the present invention.

Claims (10)

1. A magnetic resonance image quality assessment system based on image analysis, characterized in that it includes a platform end, a user end and a server; the platform end, the user end and the server are connected in communication; The user terminal includes a collection module, a security module and a transmission module; The acquisition module is used to collect the corresponding magnetic resonance image, mark it as a target image, and generate a reference image corresponding to the target image; send the target image and the reference image to the security module; The security module is used to perform security processing on the received target image and reference image, and send the processed target image and reference image to the platform through the transmission module; The transmission module is used for data transmission between the user terminal and the platform terminal; The platform includes a first sub-item module, a second sub-item module and a quality assessment module; The first sub-item module is used to compare and analyze the target image and the reference image, and calculate the corresponding first sub-item value; The second sub-item module is used to analyze the target image and calculate the corresponding second sub-item value; The quality assessment module is used to evaluate the quality score of the target image, obtain the first sub-item value and the second sub-item value, and calculate the corresponding comprehensive value according to the formula _Prior = b ₄ ×SUB _value +b ₅ ×SRB _value ; Wherein: _Prior is the comprehensive value; SUB _value is the first sub-item value; SRB _value is the second sub-item value; _b4 and _b5 are both proportional coefficients, with a value range of 0< _b4 ≤1, 0< _b5 ≤1; Acquire corresponding training data, analyze the training data, and obtain a corresponding comprehensive value set; form a number of scoring coordinates according to the quality score and the comprehensive value set corresponding to the training data; and set a corresponding evaluation function according to each of the scoring coordinates; Substituting the comprehensive value into the evaluation function to calculate the corresponding quality score; and sending the quality score to the user end. 2. A magnetic resonance image quality assessment system based on image analysis according to claim 1, characterized in that the method for confidential processing of the target image and the reference image includes: Marking the received target image and reference image as images to be processed; presetting confidentiality items and corresponding identification features; and setting confidentiality processing methods corresponding to each of the confidentiality items; Identify the image to be processed according to each confidential item and the corresponding identification feature, and identify the corresponding private information; match the corresponding confidentiality processing method for each of the private information, and process the private information according to the obtained confidentiality processing method. Perform processing; obtain the target image and reference image after confidentiality processing. 3. The magnetic resonance image quality assessment system based on image analysis according to claim 1, wherein the calculation method of the first sub-item value comprises: Perform equivalent processing and grayscale processing on the target image and the reference image; obtain the corresponding target grayscale image and reference grayscale image; Identify the grayscale value of each pixel in the target grayscale image and the reference grayscale image; generate the corresponding target grayscale matrix and reference grayscale matrix according to the grayscale value of each pixel and the corresponding position; convert the target grayscale The grayscale matrix and the reference grayscale matrix are respectively marked as and/> ; i=1, 2,..., n, n is a positive integer; j=1, 2,..., m, m is a positive integer; Calculate the corresponding difference matrix according to the formula R ₃ =R ₁ -R ₂ ; In the formula: R ₁ is the target gray level matrix; R ₂ is the reference gray level matrix; R ₃ is the difference matrix; According to the formula Calculate the corresponding first component value; where: SUB _value is the first component value; ||R ₃ || _F represents the Frobenius norm of matrix R ₃ . 4. A magnetic resonance image quality assessment system based on image analysis according to claim 1, characterized in that the calculation method of the second component value includes: Obtain the target image, identify the size of the target image, divide the target image equally according to the identified target image size, and obtain K unit blocks, and in the formula k´=K ^0.5 , k´ is a positive integer, 2≤k´ ≤10; K is a positive integer; Identify the contrast representative value, resolution representative value and brightness representative value corresponding to each of the unit blocks; generate a corresponding contrast matrix, resolution matrix and brightness matrix based on the position of each of the unit blocks; the contrast matrix, resolution matrix and brightness matrix are all k'-order square matrices; Calculate the corresponding merging matrix according to the formula R _s =b ₁ ×R ₄ +b ₂ ×R ₅ +b ₃ ×R ₆ ; In the formula: R _s is the merging matrix; R ₄ is the contrast matrix; R ₅ is the resolution matrix; R ₆ is the brightness matrix; b ₁ , b ₂ , b ₃ are all proportional coefficients, and the value range is 0<b ₁ ≤ 1, 0<b ₂ ≤1, 0<b ₃ ≤1; Mark the merged matrix as ; Mark the elements in the merged matrix as s _qp ; q=1, 2, ..., k´, k´ is a positive integer; p = 1, 2, ..., k´, k´ is a positive integer; According to the formula Calculate the corresponding second component value; In the formula: SRB _value is the second component value. 5. A magnetic resonance image quality assessment system based on image analysis according to claim 4, characterized in that the method of dividing the target image into equal parts includes: Perform simulated segmentation on the target image according to the value range of k´, obtain 9 groups of segmented images, and mark a single image in each group of segmented images as a unit block; Evaluate the contrast representative value, resolution representative value and brightness representative value corresponding to the unit block in each group of segmented images; when any contrast representative value, resolution representative value and brightness representative value in each group of segmented images are 0, eliminate the corresponding group segmented image; Calculate the segmentation evaluation value of each group of segmented images through a preset segmentation evaluation formula, identify the segmented image corresponding to the highest segmentation evaluation value, and segment the target image according to the segmentation method corresponding to the segmented image. 6. The magnetic resonance image quality assessment system based on image analysis according to claim 5, wherein the contrast representative value assessment method comprises: Identify the value of K corresponding to the segmented image, divide each unit block equally according to K, and obtain K subdivided blocks; identify the contrast corresponding to each subdivided block and mark it as a single contrast; According to the formula Calculate the contrast representative value of the corresponding unit block; In the formula: CON _t is the contrast representative value of the corresponding unit block; t represents the corresponding unit block, t=1, 2,...,K; v represents the corresponding subdivision block, v=1, 2,...,K ; DC _tv represents the single contrast corresponding to the subdivided block; G(A) is the judgment model, and A is the area of the subdivided block. 7. A magnetic resonance image quality assessment system based on image analysis according to claim 5, characterized in that the method for obtaining the resolution representative value includes: Identify the value of K corresponding to the segmented image, divide each unit block equally according to the value, and obtain K subdivided blocks; identify the corresponding resolution of each subdivided block and mark it as a single resolution; According to the formula Calculate the resolution representative value of the corresponding unit block; In the formula: RE _t is the representative value of resolution of the corresponding unit block; t represents the corresponding unit block, t=1, 2,...,K; v represents the corresponding subdivision block, v=1, 2,..., K; DR _tv represents the single resolution of the corresponding subdivision block; G(A) is the judgment model, and A is the area of the subdivision block. 8. A magnetic resonance image quality assessment system based on image analysis according to claim 5, characterized in that the method for obtaining the brightness representative value includes: Identify the value of K corresponding to the segmented image, divide each unit block equally according to K, and obtain K subdivided blocks; identify the brightness corresponding to each subdivided block and mark it as a single brightness; obtain the corresponding reference brightness, according to the brightness rate = Single brightness ÷ reference brightness calculates the brightness rate of the corresponding subdivision block; According to the formula Calculate the brightness representative value of the corresponding unit block; In the formula: BRIGH _t is the brightness representative value of the corresponding unit block; t represents the corresponding unit block, t=1, 2,...,K; v represents the corresponding subdivision block, v=1, 2,...,K ; DB _tv represents the single resolution of the corresponding subdivision block; G(A) is the judgment model, and A is the area of the subdivision block. 9. A magnetic resonance image quality assessment system based on image analysis according to any one of claims 6, 7 or 8, characterized in that the expression of the judgment model is . 10. A magnetic resonance image quality evaluation system based on image analysis according to claim 5, characterized in that the segmentation evaluation formula is ; In the formula: A _sse is the segmentation evaluation value; b ₁ , b ₂ , and b ₃ are all proportional coefficients, and the value range is 0<b ₁ ≤1, 0<b ₂ ≤1, 0<b ₃ ≤1; t represents Corresponding unit block, t=1, 2,...,K; CON _t is the representative value of contrast of the corresponding unit block; RE _t is the representative value of resolution of the corresponding unit block; BRIGH _t is the representative value of brightness of the corresponding unit block.