Image quality evaluation method and device, and medical system

Through the reference-free image quality evaluation method, the image quality score is calculated using pixel feature coefficients and gradient features, which solves the problem of difficult to ensure image comparability in medical image standardization processing, realizes automatic evaluation, and improves the accuracy and reliability of image analysis sex.

CN118918450APending Publication Date: 2024-11-08CANCER INST & HOSPITAL CHINESE ACADEMY OF MEDICAL SCI +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410824043.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-25
Publication Date
2024-11-08

Smart Images

  • Figure CN118918450A_ABST
    Figure CN118918450A_ABST
Patent Text Reader

Abstract

The invention discloses an image quality evaluation method and device and a medical system. The image quality evaluation method can comprise the steps of obtaining a to-be-evaluated image, and determining a feature coefficient corresponding to each pixel of the to-be-evaluated image; based on the feature coefficient, determining a plurality of gradient features corresponding to the to-be-evaluated image in a plurality of image orientations; for any image orientation, determining an orientation score of the image orientation based on the gradient features; and determining an image quality score of the to-be-evaluated image at least based on the plurality of orientation scores.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of medical data processing, and in particular to a method, device and medical system for evaluating the image quality of medical images without reference. Background Art

[0002] In the field of medical image processing and analysis, image quality evaluation is crucial for clinical diagnosis, disease monitoring and treatment planning. However, due to factors such as different scanning devices, scanning parameters and individual differences among patients, image characteristics are different. As a result, the results of image standardization (or image standardization) are unsatisfactory. Especially in multi-center clinical studies or long-term follow-up studies, how to ensure the comparability of image data (standardization can make images from different sources comparable) has become an urgent problem to be solved. Summary of the invention

[0003] The technical problem to be solved by this application is how to achieve real-time image quality assessment of medical images.

[0004] In order to solve the above problems, the present application discloses an image quality assessment method, device, equipment and storage medium. The image quality assessment method can realize automatic reference-free image assessment, and the assessment objectivity and consistency are high.

[0005] The first aspect of the present application discloses an image quality assessment method. The image quality assessment method may include: obtaining an image to be assessed, and determining a characteristic coefficient corresponding to each pixel of the image to be assessed; based on the characteristic coefficient, determining a plurality of gradient features corresponding to each of the image to be assessed at a plurality of image orientations; for any image orientation, determining an orientation score of the image orientation based on the gradient feature; and determining an image quality score of the image to be assessed based at least on a plurality of orientation scores.

[0006] According to some embodiments of the present application, determining a characteristic coefficient corresponding to a pixel may include: obtaining multiple attribute parameters of the pixel; performing a first mathematical operation based on the attribute parameters to determine the characteristic coefficient; the first mathematical operation is performed based on a comprehensive formula of four arithmetic operations.

[0007] According to some embodiments of the present application, determining the gradient feature corresponding to the image orientation may include: for any pixel, determining an offset value based on a feature coefficient of the pixel and feature coefficients of neighboring pixels adjacent to the pixel at the image orientation; and determining the gradient feature based on the offset values ​​corresponding to all pixels of the image to be evaluated at the image orientation.

[0008] According to some embodiments of the present application, determining the orientation score of the image orientation based on the gradient feature may include: determining multiple fitting parameters based on the gradient feature; determining a weight corresponding to each fitting parameter; and determining the orientation score based on the multiple fitting parameters and multiple weights.

[0009] According to some embodiments of the present application, determining a plurality of fitting parameters based on the gradient feature may include: performing mathematical fitting on the gradient feature to determine the fitting parameters including a shape parameter, a left scale parameter, and a right scale parameter.

[0010] According to some embodiments of the present application, determining the weight corresponding to each fitting parameter may include: performing a feature extraction operation using the fitting parameter to determine the weight.

[0011] According to some embodiments of the present application, determining the image quality score of the image to be evaluated based at least on multiple orientation scores may include: obtaining an image information measurement value corresponding to each orientation score; and determining the image quality score based on the multiple orientation scores and the multiple image information measurement values.

[0012] According to some embodiments of the present application, the image information metric value may include a median absolute deviation; and determining the image quality score may include: performing a second mathematical operation based on the multiple orientation scores and the multiple image information metric values ​​to determine the image quality score. The second mathematical operation includes a proportional weighted sum.

[0013] A second aspect of the present application provides an image quality assessment method, which may include: acquiring a first image and a second image, wherein the second image is obtained by processing the first image through one or more image processing operations; performing the operations in the image quality assessment method as described above to determine a first image quality score corresponding to the first image and a second image quality score corresponding to the second image; based on the first image quality score and the second image quality score, at least determining whether the one or more image processing operations are applicable.

[0014] According to some embodiments of the present application, the image to be evaluated or the first image may include a real-time medical scanning image.

[0015] A third aspect of the present application provides an image quality assessment device, which may include: an image acquisition module configured to acquire an image to be evaluated and determine a feature coefficient corresponding to each pixel of the image to be evaluated; a feature determination module configured to determine, based on the feature coefficients, a plurality of gradient features corresponding to each of the image to be evaluated at a plurality of image orientations; a score calculation module configured to determine, for any image orientation, an orientation score of the image orientation based on the gradient features; and a quality assessment module configured to determine an image quality score of the image to be evaluated based on a plurality of orientation scores.

[0016] In a fourth aspect of the present application, there is provided an image quality assessment device, which may include: an image acquisition module configured to acquire a first image and a second image, wherein the second image is obtained by processing the first image through one or more image processing operations; a quality scoring module configured to perform the operations in the image quality assessment method as described above to determine a first image quality score corresponding to the first image and a second image quality score corresponding to the second image; and a processing determination module configured to at least determine whether the one or more image processing operations are applicable based on the first image quality score and the second image quality score.

[0017] A fifth aspect of the present application provides an image processing device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the image quality assessment method described above when executed by the processor.

[0018] A sixth aspect of the present application provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the image quality assessment method described above are implemented.

[0019] A seventh aspect of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the image quality assessment method described above are implemented.

[0020] An eighth aspect of the present application provides a medical system, wherein the medical equipment may include the image processing apparatus as described above or the image processing equipment as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The present application will be further described in the form of exemplary embodiments, which will be described in detail by way of the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0022] Figure 1is a schematic diagram of an exemplary image processing system according to some embodiments of the present application;

[0023] Figure 2 is a block diagram of an exemplary processing device according to some embodiments of the present application;

[0024] Figure 3 is an exemplary flow chart of an image quality assessment method according to some embodiments of the present application;

[0025] Figure 4 is an exemplary flow chart of a method for determining an orientation score according to some embodiments of the present application;

[0026] Figure 5 is another exemplary flow chart of an image quality assessment method according to some embodiments of the present application;

[0027] Figure 6 is an exemplary schematic diagram of image pixels according to some embodiments of the present application;

[0028] Figure 7 is an exemplary comparison diagram of image quality scores according to some embodiments of the present application;

[0029] Figure 8 is an exemplary module diagram of a processing system for image quality assessment according to some embodiments of the present application;

[0030] Fig. 9 is another exemplary module diagram of a processing system for image processing according to some embodiments of the present application. DETAILED DESCRIPTION

[0031] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present application, so the present application is not limited by the specific embodiments disclosed below.

[0032] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by those skilled in the art to which the present application belongs. The terms used herein in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. The "first", "second" and similar words used herein do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships, and when the absolute position of the described object changes, the relative positional relationship may also change accordingly. The terms "and / or" or "and / or" used herein include any and all combinations of one or more related listed items.

[0033] Image standardization can be understood as a preprocessing step for images to improve the accuracy and reliability of subsequent image analysis and processing. However, image standardization is difficult to unify. First, different scanning methods and equipment will cause significant differences in image grayscale, contrast, etc. Even images of the same patient taken with the same scanning instrument at different time points may cause variations in image features due to physiological changes, body position adjustments, involuntary gastrointestinal motility, respiratory movements, etc. These differences make direct comparison and analysis of images difficult.

[0034] In order to evaluate image standardization, the quality of the standardized images can be evaluated to confirm whether the standardization is used. Image quality assessment is mainly divided into two methods: reference image evaluation and non-reference image evaluation. When applied to the field of medical imaging, not only efficient and accurate evaluation algorithms are required, but also in-depth understanding of the background knowledge and clinical needs of medical imaging is required to ensure that the evaluation results have practical significance and application value for medical practice. The evaluation technology used needs to overcome the special challenges of medical imaging, such as the characteristic differences of different imaging modalities (CT, MRI, X-ray, etc.), the complexity of physiological and pathological structures in the image, and the strict requirements for high accuracy and interpretability. Therefore, most reference image evaluations are currently used to ensure that the image quality meets the high standards of diagnosis and treatment. Secondly, with the rapid development of medical imaging technology, new scanning technologies and equipment are constantly emerging. These new technologies often lack sufficient sample sets as references, making the traditional image evaluation method based on a large number of samples (that is, reference image evaluation) no longer applicable. In this case, it is particularly important to establish a non-reference evaluation system.

[0035] The non-reference image evaluation standard, that is, the evaluation method that does not rely on a specific sample set or standard image, can effectively solve the above problems. This evaluation system evaluates the quality of the image by analyzing the intrinsic characteristics of the image, such as texture, shape, signal distribution, etc. It can quickly evaluate the image without the need for complex sample collection and statistical analysis.

[0036] In addition, the establishment of a reference-free evaluation system is of great significance for improving the accuracy and reliability of medical image analysis. It can reduce the interference of human factors and improve the objectivity and consistency of evaluation. At the same time, it can also provide technical support for automated image analysis, auxiliary diagnosis systems, or quantification and control of tumor movement, and improve the efficiency and quality of medical services. It can not only cope with the challenges brought by the diversity of scanning instruments, but also adapt to individual differences and changes in the treatment process, providing a powerful tool for automated analysis of medical images and personalized diagnosis and treatment.

[0037] However, establishing an effective no-reference evaluation standard is not an easy task. It requires in-depth research on the theory and technology of image processing and analysis, and the development of new algorithms and models to adapt to different medical image data. In addition, a large number of experiments and verifications are required to ensure the accuracy and robustness of the evaluation system.

[0038] The image quality assessment method disclosed in this application can realize automated assessment, that is, an automated method without external reference to realize image quality assessment. This can avoid manual subjective judgment and improve the objectivity and consistency of the assessment. In this way, an evaluation standard can be provided for image standardization.

[0039] Some preferred embodiments of the present application are described below with reference to the accompanying drawings. It should be noted that the following description is for the purpose of illustration and is not intended to limit the scope of protection of the present application. The flowchart used is used to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the described operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or a certain step or several steps of operations can be removed from these processes.

[0040] Figure 1 FIG. 1 is a schematic diagram of an exemplary image processing system according to some embodiments of the present invention. Figure 1As shown, the image processing system 100 may include an imaging device 110, a processing device 120, one or more terminals 130, a network 140, and a storage device 150. The various components in the image processing system 100 may be connected in a variety of ways. For example, the imaging device 110, the terminal 130, and / or the storage device may be connected to the processing device 120 via the network 140, or may be directly connected to the processing device 120. Image-guided radiotherapy device

[0041] The imaging device 110 may be any device with imaging functions. For example, a camera, an optical sensor, a medical imaging device, etc. In some embodiments, the imaging device 110 may be a medical imaging device including but not limited to CT, MRI, PET, SPET, PET-CT, UI, DR, DSA, etc. In some embodiments, the imaging device 110 may be a medical device that integrates medical scanning imaging and radiotherapy, and is used to perform scanning imaging while performing radiotherapy on a patient to guide the radiotherapy. Optionally or preferably, the imaging device 110 may be a medical device that combines MRI with radiotherapy equipment. Alternatively, the imaging device 110 may be a magnetic resonance guided radiotherapy (MRgRT) device.

[0042] The processing device 120 can process data and / or information obtained from the imaging device 110, the terminal 130 and / or the storage device 150, and perform one or more steps of the technical solution disclosed in the present application. For example, the processing device 120 can obtain the image to be evaluated and determine the characteristic coefficient corresponding to each pixel in the image to be evaluated. For another example, the processing device 120 can determine multiple gradient features corresponding to the image to be evaluated at multiple image orientations based on the characteristic coefficients, and determine the orientation score of each image orientation based on the gradient features. For another example, the processing device 120 can determine the image quality score of the image to be evaluated based on at least multiple orientation scores. In some embodiments, the processing device 120 can be a single server or a server group. The server group can be centralized or distributed (for example, the processing device 120 can be a distributed system). In some embodiments, the processing device 120 can be local or remote. In some embodiments, the processing device 120 can obtain information and / or data from the imaging device 110, the terminal 130 and / or the storage device 150 through the network 140. Alternatively, the processing device 120 may also be directly connected to the imaging device 110 and / or the storage device 150 to access information and / or data. In some embodiments, the processing device 120 may be implemented on a cloud platform. For example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, between clouds, multiple clouds, etc., or any combination of the above. In some embodiments, the processing device 120 may be connected to the imaging device 110 and / or the storage device 150 to access information and / or data. Figure 2For example, the processing device 120 may be implemented on a computing device such as Figure 2 The processing device 120 is implemented on a computing device 200 shown, including one or more components of the computing device 200. In some embodiments, the processing device 120 may be a part of the imaging device 110. For example, when the imaging device 110 is a medical imaging device, the processing device 120 may be integrated into an operating console of the imaging device 110.

[0043] The terminal 130 may include one or more devices with data communication functions, such as a smart mobile device 130-1, a tablet computer 130-2, a laptop computer 130-3, a desktop computer 130-4, etc. In some embodiments, the smart mobile device 120-1 may include but is not limited to a smart phone, smart glasses, a smart watch, a wearable device, a virtual display device, a display enhancement device, etc. or any combination thereof. In some embodiments, the terminal 130 may be a part of the processing device 120. For example, the computing power (e.g., CPU, GPU, etc.) of the terminal 130 may be used to implement some functions of the processing device 120. In some embodiments, the terminal 130 may be integrated into the operating console of the imaging device 110. Through the input port of the terminal 130 (e.g., the touch virtual keyboard of the mobile device 130-1 and the tablet computer 130-2 such as a smart phone smart tablet, the mouse and keyboard of the laptop computer 130-3 and the desktop computer 130-4, etc.), the user can issue control instructions to the imaging device 110, such as imaging.

[0044] The network 140 can facilitate the exchange of information and / or data. In some embodiments, one or more components of the image processing system 100 (e.g., the imaging device 110, the processing device 120, the terminal 130, the storage device 150, etc.) can send information and / or data to other components of the image processing system 100 through the network 140. For example, the processing device 120 can obtain data (e.g., a medical scan image) from the imaging device 110 through the network 140. In some embodiments, the network 140 can include a public network (e.g., the Internet), a private network (e.g., a local area network (LAN), a wide area network (WAN), etc.), a wired network (e.g., an Ethernet), a wireless network (e.g., an 802.11 network, a wireless Wi-Fi network, etc.), a cellular network (e.g., a long-term evolution (LTE) network), a frame relay network, a virtual private network (VPN), a satellite network, a telephone network, a router, a hub, a server computer, etc., or a combination thereof. For example, the network 140 may include a wired network, a fiber optic network, a telecommunications network, an intranet, the Internet, a local area network (LAN), a wide area network (WAN), a wireless local area network (WLAN), a metropolitan area network (MAN), a wide area network (WAN), a public switched telephone network (PSTN), a Bluetooth TMNetwork, ZigBee TM The network 140 may include one or more combinations of a wireless communication network, a near field communication (NFC) network, a global system for mobile communications (GSM) network, a code division multiple access (CDMA) network, a time division multiple access (TDMA) network, a general packet radio service (GPRS) network, an enhanced data rate for GSM evolution (EDGE) network, a wideband code division multiple access (WCDMA) network, a high speed downlink packet access (HSDPA) network, a long term evolution (LTE) network, a user datagram protocol (UDP) network, a transmission control protocol / internet protocol (TCP / IP) network, a short message service (SMS) network, a wireless application protocol (WAP) network, an ultra-wideband (UWB) network, a mobile communication (1G, 2G, 3G, 4G, 5G, 6G) network, Wi-Fi, Li-Fi, a narrowband Internet of Things (NB-IoT), infrared communication, etc. In some embodiments, the network 140 may include one or more network access points. For example, network 140 may include wired and / or wireless network access points, such as base stations and / or Internet exchange points, through which one or more components of image processing system 100 may connect to network 140 to exchange data and / or information.

[0045] The storage device 150 may store data and / or instructions. In some embodiments, the storage device 150 may store data obtained from the imaging device 110, the processing device 120, and the terminal 130. In some embodiments, the storage device 150 may store data and / or instructions for execution or use by the processing device 120, and the processing device 120 may implement the exemplary method described in the present application by executing or using the data and / or instructions. In some embodiments, the storage device 150 may include a large capacity memory, a removable memory, a volatile read-write memory, a read-only memory (ROM), etc., or any combination of the above examples. Exemplary large capacity memory may include a magnetic disk, an optical disk, a solid state hard disk, etc. Exemplary removable memory may include a flash disk, a floppy disk, an optical disk, a memory card, a compressed hard disk, a magnetic tape, etc. Exemplary volatile read-only memory may include a random access memory (RAM). Exemplary random access memory may include a dynamic random access memory (DRAM), a double data rate synchronous dynamic random access memory (DDRSDRAM), a static random access memory (SRAM), a thyristor random access memory (T-RAM), and a zero capacitance memory (Z-RAM), etc. Exemplary read-only memories may include masked read-only memories (MROMs), programmable read-only memories (PROMs), erasable programmable read-only memories (EPROMs), electrically erasable programmable read-only memories (EEPROMs), compact disk read-only memories (CD-ROMs), and digital multifunction disk read-only memories, etc. In some embodiments, the storage device 150 may be implemented on a cloud platform. Just to give an example, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, between clouds, multiple clouds, etc., or any combination of the above examples. For example, some algorithms or data in the present application may be stored on a cloud platform and updated regularly, and the processing device 120 accesses these algorithms or data through the network to achieve the unification and interaction of algorithms or data of the entire platform.

[0046] In some embodiments, the storage device 150 may be connected to the network 140 to enable communication with one or more components of the image processing system 100 (e.g., the imaging device 110, the processing device 120, the terminal 130, etc.). One or more components of the image processing system 100 may access data or instructions stored in the storage device 150 through the network 140. In some embodiments, the storage device 150 may be directly connected to or communicate with one or more components of the image processing system 100 (e.g., the imaging device 110, the processing device 120, the terminal 130, etc.). In some embodiments, the storage device 150 may be a part of the processing device 120.

[0047] Figure 21 is a block diagram of an exemplary processing device according to some embodiments of the present application. The computing device 200 may include any components used to implement the system described in the embodiments of the present application. For example, the computing device 200 may be implemented by hardware, software programs, firmware, or a combination thereof. For convenience, only one processing device is drawn in the figure, but the computing functions related to the image processing system 100 described in the embodiments of the present application can be implemented in a distributed manner by a group of similar platforms to disperse the processing load of the system.

[0048] In some embodiments, the computing device 200 may include a processor 210, a memory 220, an input / output component 230, and a communication port 240. In some embodiments, the processor (e.g., CPU) 210 may execute program instructions in the form of one or more processors. In some embodiments, the memory 220 includes different forms of program memory and data memory, such as a hard disk, a read-only memory (ROM), a random access memory (RAM), etc., for storing various data files processed and / or transmitted by a computer. In some embodiments, the input / output component 230 may be used to support input / output between the processing device 110 and other components. In some embodiments, the communication port 240 may be connected to a network for data communication. An exemplary processing device may include program instructions executed by the processor 210 stored in a read-only memory (ROM), a random access memory (RAM), and / or other types of non-temporary storage media. The method and / or process of the embodiment of the present application may be implemented in the form of program instructions. The computing device 200 may also receive the programs and data disclosed in the present application through network communication.

[0049] For ease of understanding, Figure 2 Only one processor is drawn as an example. However, it should be noted that the computing device 200 in the embodiment of the present application may include multiple processors, so the operations and / or methods implemented by one processor described in the embodiment of the present application may also be implemented jointly or independently by multiple processors. For example, if in the present application, the processor of the computing device 200 performs operations A and B, it should be understood that operations A and B may also be performed jointly or independently by two different processors of the computing device 200 (for example, the first processor performs operation A, the second processor performs operation B, or the first and second processors perform operations A and B jointly).

[0050] Figure 3is an exemplary flow chart of an image quality assessment method according to some embodiments of the present application. In some embodiments, process 300 may be executed by processing system 800. For example, process 300 may be stored in a storage device (such as a built-in storage unit of processing system 800 or an external storage device) in the form of a program or instruction, and when the program or instruction is executed, process 300 may be implemented. Figure 3 As shown, process 300 may include the following operations.

[0051] Step 310 , obtaining an image to be evaluated, and determining a characteristic coefficient corresponding to each pixel of the image to be evaluated. This step may be performed by the image acquisition module 810 .

[0052] In some embodiments, the image to be evaluated may be an image for which image quality scoring is required, including images obtained using any imaging device. The imaging device may be an optical imaging device such as a camera, an optical sensor, etc., or may be various imaging devices used in fields such as medical imaging, radiation detection, high energy physics, precision analysis, optical communication, etc., such as CT, MRI, PET, SPET, PET-CT, C-arm (such as DSA), DR, etc., which use the principle of high energy ray conversion to achieve imaging.

[0053] In some embodiments, the image to be evaluated may include a medical image acquired in real time, for example, a real-time medical scan image. Exemplarily, the image to be evaluated may be a medical image obtained by imaging a region of interest (for example, at least a region of a tumor that needs to be radiotherapy) of a patient undergoing radiotherapy using a magnetic resonance imaging device (MRI) during a guided radiotherapy process (for example, magnetic resonance guided radiotherapy (MRgRT)). The image to be evaluated may be acquired after a medical scan of the region of interest under specific scanning parameters and / or patient status. For example, during guided radiotherapy, the image to be evaluated is obtained by scanning the region of interest in a specific radiation field direction, a corresponding imaging plane, and when the patient is breathing freely or with restricted breathing.

[0054] In some embodiments, the image to be evaluated may be an original image directly obtained by using the imaging device, or a processed image obtained by using the original image after one or more image processing operations. In some examples, histogram equalization, adaptive histogram equalization, gamma transformation, linear grayscale transformation, unsharp mask filter enhancement, local contrast enhancement, exponential grayscale transformation, logarithmic grayscale transformation, image sharpening operator, Gaussian filtering, etc. may be used as the image processing operation.

[0055] In some embodiments, the image to be evaluated may be predetermined and stored. For example, it may be stored in a built-in storage unit or an external storage device of the processing system 800. The image acquisition module 810 may communicate with the built-in storage unit or the external storage device to acquire the image to be evaluated. In some embodiments, when the image to be evaluated is a real-time medical scan image, the image acquisition module 810 may communicate with an imaging device to acquire the image to be evaluated.

[0056] It can be known that pixels are the basic units that constitute images. Multiple pixels are arranged in rows and columns and display relevant information such as color, grayscale, position, texture features, edge features, shape features, etc., so as to display the image macroscopically. Therefore, the characteristic information of each pixel is concentrated together and can be used to represent the characteristic trend of the entire image. In the present application, the characteristic information of the pixel can be characterized by a characteristic coefficient. The characteristic coefficient can be determined using the attribute parameters of the pixel, including but not limited to the grayscale value, neighborhood mean, neighborhood variance and other parameters of the pixel. In some embodiments, the image acquisition module 810 can perform a first mathematical operation using these attribute parameters to determine the characteristic coefficient. The first mathematical operation can be performed based on a comprehensive formula of four arithmetic operations. Exemplarily, the characteristic coefficient It can be determined based on the following formula:

[0057]

[0058] Wherein, I(i,j) represents the gray value of the pixel, μ(i,j) represents the neighborhood mean (or local mean), σ(i,j) represents the neighborhood variance (or local variance), and C represents a constant. In some embodiments, C may be 1. The image acquisition module 810 may process the image to be evaluated using Gaussian blur to obtain μ(i,j) and σ(i,j) for each pixel. The first mathematical operation is then performed to determine the characteristic coefficient corresponding to each pixel. Existing calculation methods may be used to determine the neighborhood mean and neighborhood variance, and the technical solution of the present application is not limited thereto.

[0059] Step 120 , based on the feature coefficients, determining a plurality of gradient features corresponding to the image to be evaluated at a plurality of image orientations. This step may be performed by the feature determination module 820 .

[0060] It can be understood that the image quality is not only related to the quality of the multiple pixels that make up the image, but also to the relationship between the pixel and its adjacent pixels. For example, the adjacent pixels of a high-quality pixel are all relatively low-quality pixels. Even if the quality of these pixels meets the requirements, the area formed by these pixels is generally a low-quality area. In order to represent the association information between pixels, the present application uses the gradient features on the image orientation to achieve this. For any pixel in the image, its adjacent pixels are distributed on the horizontal line, vertical line, main diagonal line and secondary diagonal line of the pixel. Therefore, the image orientation may include horizontal orientation, vertical orientation, main diagonal orientation and secondary diagonal orientation, and the obtained gradient features may include horizontal gradient features, vertical gradient features, left diagonal gradient features and right diagonal gradient features.

[0061] In some embodiments, the feature determination module 820 may determine the gradient feature based on the offset value of the feature coefficient of each pixel of the image to be evaluated at each image orientation. Figure 6 This process is described exemplarily. Figure 6 An exemplary schematic diagram of image pixels is shown. For pixel (i, j), the feature determination module 820 can use the feature coefficient of pixel (i, j) and the feature coefficient of the neighboring pixels of pixel (i, j) in four image orientations to calculate the four offset values ​​of pixel (i, j). For the horizontal orientation, the neighboring pixel may be the pixel (i, j+1) located to the right of pixel (i, j). For the vertical orientation, the neighboring pixel may be the pixel (i+1, j+1) located below pixel (i, j). For the vertical main diagonal orientation, the neighboring pixel may be the pixel (i+1, j+1) located to the lower right of pixel (i, j). For the secondary diagonal orientation, the neighboring pixel may be the pixel (i+1, j-1) located to the lower left of pixel (i, j). Exemplarily, the offset values ​​corresponding to pixel (i, j) in four image orientations may be determined based on the following formula:

[0062]

[0063] Among them, H(i,j) represents the offset value in the horizontal direction, V(i,j) represents the offset value in the vertical direction, D1(i,j) represents the offset value in the main diagonal direction, and D2(i,j) represents the offset value in the secondary diagonal direction.

[0064] After determining the offset value of each pixel in the four image orientations, the feature determination module 820 can use the offset value to arrange the pixels in the original image, that is, the row and column positions in the image to be evaluated, to obtain four corresponding offset value matrices. The offset value matrix can represent the change of features in a specific image orientation, and can thus be designated as the gradient feature. Of course, for some pixels that do not have neighboring pixels in a certain direction, their corresponding offset values ​​can be discarded in the above-mentioned offset value matrix. For example, assuming that the size of the image to be evaluated is 100×100 pixels, the size of the offset value matrix corresponding to the horizontal orientation can be 100×99, the size of the offset value matrix corresponding to the vertical orientation can be 99×100, the size of the offset value matrix for the main diagonal orientation is 99×99, and the size of the offset value matrix for the secondary diagonal orientation is 99×99.

[0065] Step 330 : For any image orientation, determine the orientation score of the image orientation based on the gradient feature. This step may be performed by the score calculation module 830 .

[0066] In some embodiments, the orientation score may be a quality score for the object to be evaluated at the corresponding image orientation. In combination with the above description, the association information between adjacent pixels also affects the image quality, and the quality change of the image at each image orientation jointly affects the quality of the entire image. Therefore, the final image quality score may be determined based on the orientation scores of multiple image orientations. It can be understood that the clarity, contrast, brightness distribution, darkness distribution, etc. of the image all affect the quality score of the image. These properties can be quantified using corresponding parameters. To determine the orientation score corresponding to any image orientation, the score calculation module 830 may mathematically fit the gradient features of any image orientation to determine multiple fitting parameters. The fitting parameters may be used to quantify the influence of the properties of the image to be evaluated on the image quality at the corresponding image orientation, or may be referred to as the quality features of the image to be evaluated. Any applicable fitting method may be used for this operation, such as linear fitting, polynomial fitting, Gaussian fitting, surface fitting, nonlinear fitting, etc. In addition, the score calculation module 830 may further determine the weight of each fitting parameter. The weight may be used to determine the contribution value of the corresponding fitting parameter to the influence of the image quality. In other words, participation. The scoring calculation module 830 can perform feature extraction operations on the fitting parameters to determine the weight corresponding to each fitting parameter. For example, the scoring calculation module 830 can use a data set (for example, a fitting parameter data set) including the fitting parameters, which is composed of fitting parameters of multiple associated images related to the image to be evaluated, to perform feature extraction to determine the weight corresponding to the fitting parameters. Subsequently, the scoring calculation module 830 can use the fitting parameters and the weights to perform calculations such as weighted summation to determine the orientation score. For other instructions on the orientation score, please refer to the present application. Figure 4 Part, no more elaboration.

[0067] Step 340 , determining the image quality score of the image to be evaluated based at least on the multiple orientation scores. This step may be performed by the quality evaluation module 840 .

[0068] In some embodiments, the quality assessment module 840 can determine the image information metric associated with each orientation score, and determine the image quality score based on multiple orientation scores and multiple image information metric values. It can be understood that in the field of medical imaging, a single medical image cannot present complete information. For example, during magnetic resonance guided radiotherapy (MRgRT), the MRI scanning imaging of the patient is generally performed continuously during 3-5 respiratory cycles based on the same scanning parameters and / or patient status, and a plurality of continuous image frames, such as 50 image frames, are obtained. These 50 image frames together show the changes in the shape, position, etc. of the patient's radiotherapy area (that is, the tumor target area). After processing, it is possible to accurately track the target area in real time, improve the quality of radiotherapy, and reduce patient damage. Therefore, when performing image quality assessment, since these 50 image frames have the same image orientation, if the information is rich in a certain image orientation, it can be considered that the light and dark contrast, texture, and other information contrast in this image orientation are prominent, so the data fluctuation will be greater. This fluctuation will affect all image frames. The image information metric value is used to indicate the degree of fluctuation of the image to be evaluated in the "image set".

[0069] In some embodiments, for the above-mentioned images to be evaluated in the "image set", such as real-time medical scanning images, the image information metric value can be determined using the image information metric value of each image in the "image set". It can be understood that each image in this "image set" can be subjected to image quality evaluation to better carry out subsequent work. For example, the 50 image frames in the aforementioned example can be screened according to the image quality score of each image frame, and image frames with high quality scores can be selected for tumor tracking to achieve better tracking effects. In some embodiments, for a single image to be evaluated, such as a photographic image, the image information metric value can be the same at each image orientation. For example, the image information metric values ​​at the four image orientations can all be 1.

[0070] For the image to be evaluated in the "image set", for each image orientation, the quality assessment module 840 can use the median absolute deviation (MAD) of all orientation scores at the image orientation as the image information metric value, that is, the image information metric value corresponding to the orientation score of the image to be evaluated at the image orientation. Multiple orientation scores and multiple image information metric values ​​can be used to perform a second mathematical operation to determine the image quality score. The second mathematical operation can include at least a proportional weighted sum. Exemplarily, the quality assessment module 840 can determine the image quality score based on the following formula:

[0071] QI=w H ·QI H +w V QI V+w D1 QI D1 +w D2 QI D2

[0072]

[0073]

[0074] Wherein, QI represents the image quality score, QI H The orientation score indicating horizontal orientation, QI V The orientation score representing the vertical orientation, QI D1 Indicates the orientation score of the main diagonal orientation, QI D2 represents the orientation score of the sub-diagonal orientation, w H Indicates the proportion of image information measurement value in the horizontal direction in the whole, w V Indicates the proportion of the image information measurement value in the vertical direction of the water in the whole, w D1 Indicates the proportion of the image information measurement value on the main diagonal direction in the whole, w D2 Indicates the proportion of the image information measurement value in the sub-diagonal direction in the whole.

[0075] For a single object to be evaluated, since the image information metric values ​​at each image orientation are the same, the quality evaluation module 840 may determine the image quality score based on the following formula:

[0076] QI=1 / 4·(QI H +QI V +QI D1 +QI D2 )

[0077] It should be noted that the above Figure 3 The description of each step in the embodiment is only for example and explanation, and does not limit the scope of application of the present application. For those skilled in the art, under the guidance of the present application, Figure 3 Various modifications and changes are made to the various steps in the present invention. However, these modifications and changes are still within the scope of this application.

[0078] The image quality assessment method disclosed in the present application can realize automatic reference-free image assessment, and can more accurately assess distortion caused by motion blur, partial volume failure, etc., thereby improving the accuracy of image quality assessment, especially medical image assessment.

[0079] Figure 4is an exemplary flow chart of a method for determining an orientation score according to some embodiments of the present application. In some embodiments, process 400 can be executed for each image orientation to determine the orientation scores corresponding to each of the multiple image orientations. Process 400 can be executed by the score calculation module 830 of the processing system 800. For example, process 400 can be stored in a storage device (such as a built-in storage unit of the processing system 400 or an external storage device) in the form of a program or instruction, and the program or instruction can implement process 400 when executed. Figure 4 As shown, process 400 may include the following operations.

[0080] Step 410: determining a plurality of fitting parameters based on the gradient characteristics.

[0081] In some embodiments, the score calculation module 830 may perform mathematical fitting on the gradient feature to determine a plurality of fitting parameters. Exemplarily, the score calculation module 830 may fit the gradient feature to an asymmetric generalized Gaussian distribution (AGGD) to obtain corresponding fitting parameters. For example, a shape parameter, a left scale parameter, and a right scale parameter may be included. The shape parameter may be used to indicate the effect of the clarity and contrast of the image on the image quality, or may be referred to as a clarity / contrast feature parameter, the left scale parameter may be used to indicate the effect of the darkness distribution of the image on the image quality, or may be referred to as a darkness distribution feature parameter, and the right scale parameter may be used to indicate the effect of the brightness portion of the image on the image quality, or may be referred to as a brightness distribution parameter.

[0082] In step 420, a weight corresponding to each fitting parameter is determined.

[0083] In some embodiments, the weights may be used to evaluate the influence of the above fitting parameters on the image quality, or the contribution of the fitting parameters to the image quality. The score calculation module 830 may perform a feature extraction operation on the fitting parameters to determine the weights. Figure 3As described in the relevant description, the association information between adjacent pixels is actually related to all images in the "image set" where the image to be evaluated is located. Therefore, the score calculation module 830 can use a clustering algorithm to extract features from the fitting parameters of all images in the "image set" (for example, determined by the relevant operations of process 300) to determine the corresponding weights. Exemplary clustering algorithms may include but are not limited to K-means clustering algorithm, hierarchical clustering algorithm, DBSCAN, OPTICS, mean shift, Gaussian Mixture Model, etc. This step is exemplified by taking the K-means algorithm as an example. The fitting parameters corresponding to all images in the "image set" can be used to create a data set, and the data is standardized so that the numerical ranges of different fitting parameters are consistent. Randomly divide K initial clusters (or clusters), for each fitting parameter in the data set, calculate its distance to each centroid (or cluster center), and assign the fitting parameter to the cluster to which the nearest centroid belongs. For each cluster, calculate the mean of all samples in the cluster, and use the mean as the new centroid. Repeat the above steps until the centroid no longer changes or the preset number of iterations is reached. The value of the centroid can be assigned as the weight corresponding to the fitting parameter.

[0084] In some embodiments, machine learning models such as logistic regression, support vector machine, decision tree, KNN algorithm, naive Bayes, random forest, AdaBoost, GBDT, XGBoost, artificial neural network, etc. can also be used to automatically implement the feature extraction operation. For example, the score calculation module 830 can input the fitting parameters into one of the above trained machine learning models to obtain the corresponding weights. Optionally or preferably, the XGBoost model can be used to process the fitting parameters to predict the weights. The XGBoost model can be pre-trained, for example, using a large number of fitting parameters corresponding to the image and the related weights as training data (for example, as sample input values ​​and corresponding true values) for model training. After the process of initialization, iterative tree construction, residual calculation, model update, regularization and loss function optimization, when the preset maximum number of iterations is reached, or the change of the loss function is less than the preset threshold, or the validation set is used for early stopping (Early Stopping) When the performance on the validation set is no longer improved, the training can be stopped to obtain the trained XGBoost model. The model can be stored in the built-in storage unit or external storage device of the processing system 800. The score calculation module 830 can communicate with the built-in storage unit or the external storage device to call the trained XGBoost model.

[0085] Step 430: Determine the orientation score based on the multiple fitting parameters and multiple weights.

[0086] In some embodiments, the score calculation module 830 performs weighted summation using multiple fitting parameters and their corresponding weights to obtain the orientation score. Exemplarily, the orientation score can be determined using the following formula:

[0087] QI x =w 1x α x +w 2x β 1x +w 3x β 2x

[0088] Among them, QI x represents the orientation score, and x can refer to four different image orientations, including the horizontal orientation H, the vertical orientation V, the main diagonal orientation D1, and the secondary diagonal orientation D2. α, β1, and β2 can respectively represent the fitting parameters, including the shape parameter, the left scale parameter, and the right scale parameter. w1, w2, and w3 can respectively represent the weights corresponding to the fitting parameters. After performing the above process for each image orientation, the corresponding orientation score can be obtained. For example, the four orientation scores QI on the four image orientations are H ,QI V ,QI D1 and QI D2 .

[0089] It should be noted that the above Figure 4 The description of each step in the embodiment is only for example and explanation, and does not limit the scope of application of the present application. For those skilled in the art, under the guidance of the present application, Figure 4 Various modifications and changes are made to the various steps in the present invention. However, these modifications and changes are still within the scope of this application.

[0090] Figure 5 is an exemplary flow chart of an image quality assessment method according to some embodiments of the present application. In some embodiments, process 500 may be executed by processing system 900. For example, process 500 may be stored in a storage device (such as a built-in storage unit of processing system 900 or an external storage device) in the form of a program or instruction, and when the program or instruction is executed, process 500 may be implemented. Figure 5 As shown, process 500 may include the following operations.

[0091] Step 510 , acquiring a first image and a second image. This step may be performed by the image processing module 910 .

[0092] In some embodiments, the first image may be the same as or similar to the image to be evaluated in the above description. Reference may be made to the previous description, which will not be repeated here. The second image may be obtained by processing the first image through one or more image processing operations. Exemplary image processing operations may include, but are not limited to, logarithmic transformation, gamma transformation, histogram equalization, adaptive histogram equalization, linear grayscale transformation, unsharp mask filter enhancement, local contrast enhancement, exponential grayscale transformation, image sharpening operator, Gaussian filtering, etc. or any combination thereof. For example, logarithmic transformation can be used to strengthen low grayscale areas in an image, gamma transformation can adjust the overall contrast of an image, histogram equalization can be used to uniformly distribute grayscale values ​​of an image, adaptive histogram equalization can adjust the local grayscale value distribution, linear grayscale transformation can be used to linearly adjust the grayscale value of an image, unsharp mask filter enhancement can be used to enhance image details, local contrast enhancement can improve the local contrast effect of an image, exponential grayscale transformation can be used to enhance the bright details of an image, image sharpening operator can be used to improve the edge clarity of an image, Gaussian filtering can smooth an image and reduce the influence of noise, and optimize the quality of an image. Other image processing operations such as Laplace enhancement, non-local mean filtering, etc. may also be applied. The image processing module 910 may apply the above exemplary image processing operations to the first image to obtain the second image. It should be noted that the above image processing operations are only for illustrative purposes and are not intended to limit the present application.

[0093] Step 520 : Determine a first image quality score corresponding to the first image and a second image quality score corresponding to the second image. This step may be performed by the quality scoring module 920 .

[0094] In some embodiments, the determination of the first image quality score and the second image quality score may be based on the present application Figure 3-Figure 4 For details, please refer to the relevant instructions, which will not be repeated here.

[0095] Step 530 : Based on the first image quality score and the second image quality score, at least determine whether the one or more image processing operations are applicable. This step may be performed by the processing determination module 930 .

[0096] It can be understood that the one or more image processing operations are to improve the quality of the image. The quantitative parameter of the image quality is the image quality score. It is easy to know that when the first image quality score (also referred to as QI1) of the first image is compared with the second image quality score (also referred to as QI2) of the second image, if QI1>QI2, it means that the image quality of the second image is worse than that of the first image. The one or more image processing operations performed on the first image do not improve the image quality, but have side effects. Therefore, the processing determination module 930 can determine that the one or more image processing operations are not suitable for quality enhancement of the first image. On the contrary, if QI1<QI2, it means that the image quality of the second image is better than that of the first image. The one or more image processing operations performed on the first image improve the image quality and play a positive role. Therefore, the processing determination module 930 can determine that the one or more image processing operations are suitable for quality enhancement of the first image and can be used in subsequent processes. For example, in the field of medical image processing, for a set of medical scan images for a patient (such as a set of 50 image frames of MRI scan images acquired during guided radiotherapy), after executing process 500 on the previous image frame and determining that the image processing operation is applicable, the same image processing operation can be performed on all subsequent image frames.

[0097] refer to Figure 7 To intuitively illustrate the comparison of image quality scores. Figure 7 As shown, 5 groups of medical images are included, each group includes 50 image frames. The image quality score (i.e., QI2) of the processed image determined after the image processing operation is greater than the image quality score (i.e., QI1) of the image before processing. In this way, the applied image processing operation can be determined to be applicable.

[0098] In some embodiments, the image quality score can also be used for image screening. For example, for a group of medical images processed by the same and applicable image processing method, 5% of the score distribution of all image quality scores can be determined as the lower threshold. Images below the lower threshold can be considered to still not meet the quality standards after processing and can be discarded. Or reprocessed. Of course, the lower threshold can be adjusted according to actual conditions, such as 1%, 5%, 10%, etc., which does not limit this application.

[0099] It should be noted that the above Figure 5 The description of each step in the embodiment is only for example and explanation, and does not limit the scope of application of the present application. For those skilled in the art, under the guidance of the present application, Figure 5 Various modifications and changes are made to the various steps in the present invention. However, these modifications and changes are still within the scope of this application.

[0100] Figure 8 is an exemplary module diagram of a processing system for image quality assessment according to some embodiments of the present application. The processing system can implement a reference-free quality score for an image. Figure 8 As shown, the processing system 800 may include an image acquisition module 810 , a feature determination module 820 , a score calculation module 830 , and a quality assessment module 840 .

[0101] The image acquisition module 810 may be configured to acquire an image to be evaluated and determine a characteristic coefficient corresponding to each pixel of the image to be evaluated. The image to be evaluated may be an image for which image quality scoring is required, including an image obtained using any imaging device. The characteristic coefficient may be characteristic information for characterizing pixels constituting the image to be evaluated. The image acquisition module 810 may determine the characteristic coefficient by using a comprehensive formula based on four arithmetic operations using attribute parameters of pixels (e.g., gray value, neighborhood mean, neighborhood variance, etc.).

[0102] The feature determination module 820 may be configured to determine a plurality of gradient features corresponding to the image to be evaluated at a plurality of image orientations based on the feature coefficients. The feature determination module 820 may determine the gradient feature based on the offset value of the feature coefficient of each pixel of the image to be evaluated at each image orientation, and use the offset values ​​to arrange the pixels according to the row and column positions in the original image, i.e., the image to be evaluated, to obtain four corresponding offset value matrices. The offset value matrix may represent the change of features at a specific image orientation, and thus may be designated as the gradient feature.

[0103] The quality assessment module 840 may be configured to determine the image quality score of the image to be assessed based at least on the multiple orientation scores. The quality assessment module 840 may determine an image information metric associated with each orientation score, and determine the image quality score based on the multiple orientation scores and the multiple image information metric values. A second mathematical operation may be performed using the multiple orientation scores and the multiple image information metric values ​​to determine the image quality score. The second mathematical operation may include at least a proportional weighted sum.

[0104] Fig. 9 is another exemplary module diagram of a processing system for image processing according to some embodiments of the present application. The processing system can implement non-reference quality scoring of images and admission determination of image processing operations. Fig. 9 As shown, the processing system 900 may include an image processing module 910 , a quality scoring module 920 , and a processing determination module 930 .

[0105] The image processing module 910 may be configured to acquire a first image and a second image. The first image may be the same as or similar to the image to be evaluated described above. The second image may be obtained by processing the first image with one or more image processing operations. The image processing module 910 may perform one or more image processing operations such as logarithmic transformation, gamma transformation, histogram equalization, adaptive histogram equalization, linear grayscale transformation, unsharp mask filter enhancement, local contrast enhancement, exponential grayscale transformation, image sharpening operator, Gaussian filtering, etc. on the first image to obtain the second image.

[0106] The quality scoring module 920 may be configured to determine a first image quality score corresponding to the first image and a second image quality score corresponding to the second image. The quality scoring module 920 may execute processes 300 and 400 to determine the first image quality score and the second quality score.

[0107] The processing determination module 930 may be configured to at least determine whether the one or more image processing operations are applicable based on the first image quality score and the second image quality score. The processing determination module 930 may compare the first image quality score with the second image quality score. If the first image quality score is greater than the second image quality score, it means that the one or more image processing operations are not applicable to the quality enhancement of the first image. If the first image quality score is less than the second image quality score, it means that the one or more image processing operations are applicable to the quality enhancement of the first image.

[0108] The processing system 900 can also perform image screening, for example, based on the quality scores of a group of medical images processed by the same and applicable image processing method, to screen out images that do not meet the image quality requirements.

[0109] For other descriptions of the above modules, please refer to other parts of this application, for example, Figures 3 to 5 .

[0110] It should be understood that Figure 8 and Fig. 9The system and its modules shown can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software or a combination of software and hardware. Among them, the hardware part can be implemented using dedicated logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or a dedicated design hardware. Those skilled in the art will understand that the above-mentioned method and system can be implemented using computer executable instructions and / or included in a processor control code, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. Such code is provided on the carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of the present application can not only be implemented by hardware circuits such as ultra-large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but can also be implemented by software such as executed by various types of processors, and can also be implemented by a combination of the above-mentioned hardware circuits and software (for example, firmware).

[0111] It should be noted that the above description of the modules is only for convenience of description and does not limit the present application to the scope of the embodiments. It is understandable that for those skilled in the art, after understanding the principle of the system, it is possible to arbitrarily combine the modules or form a subsystem connected to other modules without deviating from this principle. For example, the processing system 800 and the processing system 900 can be the same system, such as the processing system 800 is a part of the processing system 900. For another example, the modules can share a storage module, or the modules can have their own storage modules. Such variations are all within the scope of protection of the present application.

[0112] Some embodiments of the present application also disclose a medical system. The medical system may include the device / system / apparatus for image quality scoring as described above. Exemplarily, the medical system may include an imaging device for acquiring a medical scan image, and an operating console for processing the medical scan image. The operating console may be used to implement the process of the above-mentioned image quality scoring method.

[0113] The basic concepts have been described herein. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of the present application. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements and amendments to the present application. Such modifications, improvements and amendments are suggested in the present application, so such modifications, improvements and amendments still belong to the spirit and scope of the exemplary embodiments of the present application.

[0114] At the same time, the present application uses specific words to describe the embodiments of the present application. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present application. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or multiple times in different positions in the present application does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present application can be appropriately combined.

[0115] In addition, it will be appreciated by those skilled in the art that various aspects of the present application may be illustrated and described by a number of patentable categories or situations, including any new and useful process, machine, product or combination of substances, or any new and useful improvements thereto. Accordingly, various aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may all be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". In addition, various aspects of the present application may be represented as a computer product located in one or more computer-readable media, which includes computer-readable program code.

[0116] A computer storage medium may include a propagated data signal containing computer program code, such as in baseband or as part of a carrier wave. The propagated signal may be in a variety of forms, including electromagnetic, optical, etc., or a suitable combination. A computer storage medium may be any computer-readable medium other than a computer-readable storage medium, which can be connected to a command control system, device or equipment to communicate, propagate or transmit the program for use. The program code on the computer storage medium may be transmitted via any suitable medium, including radio, cable, fiber optic cable, RF, or similar media, or any combination of the above media.

[0117] The computer program coding required for the operation of each part of the application can be written in any one or more programming languages, including object-oriented programming languages ​​such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB.NET, Python, etc., conventional procedural programming languages ​​such as C language, Visual Basic, Fortran 2003, Perl, COBOL 2002, PHP, ABAP, dynamic programming languages ​​such as Python, Ruby and Groovy, or other programming languages, etc. The program coding can be run completely on the user's computer, or run on the user's computer as an independent software package, or run partly on the user's computer and partly on the remote computer, or run completely on the remote computer or server. In the latter case, the remote computer can be connected to the user's computer through any network form, such as a local area network (LAN) or a wide area network (WAN), or connected to an external computer (e.g., via the Internet), or in a cloud computing environment, or used as a service such as software as a service (SaaS).

[0118] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this application, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this application. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this application. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0119] Similarly, it should be noted that in order to simplify the description of the disclosure of this application and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this application, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the object of this application are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0120] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of the present application are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0121] Each patent, patent application, patent application disclosure, and other materials, such as articles, books, instructions, publications, documents, etc., cited in this application are hereby incorporated by reference in their entirety. Except for application history documents that are inconsistent with or conflicting with the content of this application, documents that limit the broadest scope of the claims of this application (currently or later attached to this application) are also excluded. It should be noted that if the descriptions, definitions, and / or use of terms in the attached materials of this application are inconsistent or conflicting with the content described in this application, the descriptions, definitions, and / or use of terms in this application shall prevail.

[0122] Finally, it should be understood that the embodiments described in this application are only used to illustrate the principles of the embodiments of the present application. Other variations may also fall within the scope of the present application. Therefore, as an example and not a limitation, the alternative configurations of the embodiments of the present application may be considered to be consistent with the teachings of the present application. Accordingly, the embodiments of the present application are not limited to the embodiments explicitly introduced and described in the present application.

Claims

1. A method for evaluating image quality, characterized in that: The method comprises: Acquire an image to be evaluated, and determine a characteristic coefficient corresponding to each pixel of the image to be evaluated; Based on the characteristic coefficients, determining a plurality of gradient features corresponding to the image to be evaluated at a plurality of image orientations; For any image orientation, determining an orientation score of the image orientation based on the gradient feature; An image quality score of the image to be evaluated is determined based at least on the multiple orientation scores.

2. The image quality assessment method according to claim 1, characterized in that: Determine the characteristic coefficients corresponding to the pixels, including: Obtaining multiple attribute parameters of the pixel; A first mathematical operation is performed based on the attribute parameters to determine the characteristic coefficient; the first mathematical operation is performed based on a comprehensive formula of four arithmetic operations.

3. The image quality assessment method according to claim 1, characterized in that: Determining the gradient feature corresponding to the image orientation includes: For any pixel, Determining an offset value based on a characteristic coefficient of the pixel and characteristic coefficients of neighboring pixels adjacent to the pixel in the image orientation; The gradient feature is determined based on the offset values ​​corresponding to all pixels of the image to be evaluated at the image position.

4. The image quality assessment method according to claim 1, characterized in that: The determining of the orientation score of the image orientation based on the gradient feature comprises: Based on the gradient characteristics, determining a plurality of fitting parameters; determining a weight corresponding to each fitting parameter; The orientation score is determined based on the plurality of fitting parameters and a plurality of weights.

5. The image quality assessment method according to claim 4, characterized in that: The determining of a plurality of fitting parameters based on the gradient feature comprises: performing mathematical fitting on the gradient feature to determine the fitting parameters including a shape parameter, a left scale parameter and a right scale parameter; and the determining of a weight corresponding to each fitting parameter comprises: performing a feature extraction operation using the fitting parameters to determine the weight.

6. The image quality assessment method according to claim 1, characterized in that: The step of determining the image quality score of the image to be evaluated based at least on the multiple orientation scores includes: Obtain the image information measurement value corresponding to each orientation score; A second mathematical operation is performed based on the plurality of orientation scores and the plurality of image information metric values ​​to determine the image quality score; the second mathematical operation comprises a proportional weighted sum.

7. A method for evaluating image quality, characterized in that: The method comprises: Acquire a first image and a second image, wherein the second image is obtained by processing the first image through one or more image processing operations; Performing the operations in the image quality assessment method according to any one of claims 1 to 6 to determine a first image quality score corresponding to the first image and a second image quality score corresponding to the second image; Based on the first image quality score and the second image quality score, at least it is determined whether the one or more image processing operations are applicable.

8. An image quality assessment device, characterized in that: The image quality assessment device comprises: An image acquisition module is configured to acquire an image to be evaluated and determine a characteristic coefficient corresponding to each pixel of the image to be evaluated; A feature determination module is configured to determine, based on the feature coefficients, a plurality of gradient features corresponding to the image to be evaluated at a plurality of image orientations; a score calculation module configured to determine, for any image orientation, an orientation score of the image orientation based on the gradient feature; and The quality assessment module is configured to determine the image quality score of the image to be assessed based on multiple orientation scores.

9. An image quality assessment device, characterized in that: The image quality assessment device comprises: An image acquisition module is configured to acquire a first image and a second image, wherein the second image is obtained by processing the first image through one or more image processing operations; a quality scoring module, configured to perform the operations in the image quality assessment method according to any one of claims 1 to 6, so as to determine a first image quality score corresponding to the first image and a second image quality score corresponding to the second image; The processing determination module is configured to at least determine whether the one or more image processing operations are applicable based on the first image quality score and the second image quality score.

10. A medical system, characterized in that: The medical device comprises the image processing apparatus according to claim 8 or 9.