Artifact identification method and device, nonvolatile storage medium and electronic equipment

Abstract

This application discloses an artifact recognition method, apparatus, non-volatile storage medium, and electronic device. The method includes: determining a first image similarity between a target scanned image and a first reference image in a scanned image sequence, and a second image similarity between the target scanned image and a second reference image, wherein the first and second reference images are adjacent to the target scanned image in the scanned image sequence, with the first reference image preceding the target scanned image and the second reference image following the target scanned image; determining an image similarity deviation value corresponding to the target scanned image based on the smaller of the first and second image similarities and a first baseline similarity; and determining an abnormal scanned image set in the scanned image sequence based on the image similarity deviation value. This application solves the technical problem in related technologies where image artifact recognition is easily affected by single-layer image noise, leading to inaccurate artifact recognition.

Description

Technical Field

[0001] This application relates to the field of image processing, and more specifically, to an artifact recognition method, apparatus, non-volatile storage medium, and electronic device. Background Technology

[0002] In related technologies, when determining whether artifacts exist in scanned images, they are easily affected by noise in single-layer images or abrupt changes in local anatomical structures, resulting in a large number of false positive identification results.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This application provides an artifact recognition method, apparatus, non-volatile storage medium, and electronic device to at least solve the technical problem in the related art where image artifacts are easily affected by single-layer image noise, resulting in inaccurate artifact recognition.

[0005] According to one aspect of the embodiments of this application, an artifact recognition method is provided, comprising: determining a first image similarity between a target scanned image and a first reference image in a scanned image sequence, and a second image similarity between the target scanned image and a second reference image, wherein the target scanned image is any image in the scanned image sequence excluding the first image and the last image, the first reference image and the second reference image are adjacent to the target scanned image in the scanned image sequence, and the first reference image is located before the target scanned image and the second reference image is located after the target scanned image in the scanned image sequence; determining an image similarity deviation value corresponding to the target scanned image based on the smaller image similarity between the first image similarity and the second image similarity, and a first benchmark similarity; and determining an abnormal scanned image set in the scanned image sequence based on the image similarity deviation value, wherein the abnormal scanned image set includes abnormal scanned images that are determined to contain artifacts.

[0006] Optionally, determining the abnormal scan image set in the scan image sequence based on the image similarity deviation value includes: determining a first preset threshold; determining the abnormal scan image set based on the first preset threshold and the image similarity deviation value corresponding to the target scan image, wherein the image similarity deviation value corresponding to the scan image in the abnormal scan image set is less than the first preset threshold, and the number of scan images in the abnormal scan image set is not less than a first preset number.

[0007] Optionally, determining the image similarity between the target scanned image and the reference image in the scanned image sequence includes: determining the human image region in the target scanned image and the human image region in the reference image; determining the structural similarity between the human image region in the target scanned image and the human image region in the reference image; and using the structural similarity as the image similarity between the target scanned image and the reference image.

[0008] Optionally, after determining the set of abnormal scanned images in the scanned image sequence, the method further includes: performing image block processing on the abnormal scanned images in the abnormal scanned image set to obtain multiple sets of image blocks, wherein the image blocks in the same image block set have the same relative position in the abnormal scanned images; determining the first image block similarity of the target image block in the image block set relative to the first reference image block, and the second image block similarity of the target image block relative to the second reference image block, wherein the target image block is any image block in the image block set other than the first and last image blocks. An image block is defined, wherein a first reference image block and a second reference image block are adjacent to a target image block, and in the image block set, the first reference image block is located before the target image block, and the second reference image block is located after the target image block; based on the smaller image block similarity between the first image block similarity and the second image block similarity, and a second baseline similarity, an image block similarity deviation value corresponding to the target image block is determined; based on the image block similarity deviation value, an artifact image block set is determined in the image block set, and based on the artifact image block set, an artifact region in the abnormal scan image is determined.

[0009] Optionally, determining the artifact image block set in the image block set based on the image block similarity deviation value includes: determining a second preset threshold; determining the artifact image block set based on the second preset threshold and the image block similarity deviation value corresponding to the target image block, wherein there are a second preset number of consecutive image blocks in the artifact image block set whose image block similarity deviation value is not less than the second preset threshold.

[0010] Optionally, determining the artifact region in the abnormal scan image based on the artifact image block set includes: determining the artifact image block belonging to the target abnormal scan image in the artifact image block set, wherein the target abnormal scan image is any abnormal scan image; and stitching together the artifact image block belonging to the target abnormal scan image to obtain the artifact region in the target abnormal scan image.

[0011] Optionally, determining the artifact region in the abnormal scan image based on the artifact image block set includes: determining the relative position corresponding to the artifact image block set; and determining the artifact region in the abnormal scan image based on the relative position.

[0012] According to another aspect of the embodiments of this application, an artifact recognition device is also provided, comprising: a first processing module, configured to determine a first image similarity between a target scanned image and a first reference image in a scanned image sequence, and a second image similarity between the target scanned image and a second reference image, wherein the target scanned image is any image in the scanned image sequence excluding the first image and the last image, the first reference image and the second reference image are adjacent to the target scanned image in the scanned image sequence, and in the scanned image sequence, the first reference image is located before the target scanned image and the second reference image is located after the target scanned image; a second processing module, configured to determine an image similarity deviation value corresponding to the target scanned image based on the smaller image similarity between the first image similarity and the second image similarity, and a first benchmark similarity; and a third processing module, configured to determine an abnormal scanned image set in the scanned image sequence based on the image similarity deviation value, wherein the abnormal scanned image set includes abnormal scanned images that are determined to contain artifacts.

[0013] According to another aspect of the embodiments of this application, a non-volatile storage medium is also provided, wherein a program is stored in the non-volatile storage medium, and the program controls the device where the non-volatile storage medium is located to execute an artifact recognition method when it runs.

[0014] According to another aspect of the embodiments of this application, an electronic device is also provided, including: a memory and a processor, the processor being configured to run a program stored in the memory, wherein the program executes an artifact recognition method during runtime.

[0015] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the steps of an artifact recognition method.

[0016] In this embodiment, a method is employed to determine the first image similarity between the target scanned image and the first reference image in the scanned image sequence, and the second image similarity between the target scanned image and the second reference image. The target scanned image is any image in the scanned image sequence excluding the first and last images. The first and second reference images are adjacent to the target scanned image in the scanned image sequence, with the first reference image preceding the target scanned image and the second reference image following the target scanned image. Based on the smaller of the first and second image similarities and a first baseline similarity, an image similarity deviation value corresponding to the target scanned image is determined. Based on the image similarity deviation value, an abnormal scanned image set is determined in the scanned image sequence, including abnormal scanned images determined to contain artifacts. By determining the image similarity deviation value of the scanned images and determining the abnormal scanned image set based on the image similarity deviation value, the influence of single-layer image noise is avoided, thereby improving the technical effect of artifact recognition accuracy. This solves the technical problem in related technologies where image artifact recognition is easily affected by single-layer image noise, leading to inaccurate artifact recognition. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0018] Figure 1 This is a schematic diagram of the structure of a computer terminal (or mobile device) according to an embodiment of this application;

[0019] Figure 2 This is a flowchart illustrating an artifact recognition method provided according to an embodiment of this application;

[0020] Figure 3 This is a schematic diagram of an identification result provided according to an embodiment of this application;

[0021] Figure 4 This is a flowchart illustrating a process for determining artifact regions according to an embodiment of this application;

[0022] Figure 5 This is a schematic diagram of the structure of an artifact recognition device provided according to an embodiment of this application. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0024] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] In computed tomography (CT) and other imaging techniques, patient movement during scanning can cause structural misalignment, blurring, or ghosting between adjacent slices, known as motion artifacts. Motion artifacts reduce diagnostic reliability, therefore they need to be detected and labeled during quality control.

[0026] In related technologies, commonly used methods include single similarity thresholding and gradient- or edge-based methods. Single similarity thresholding calculates a certain similarity (such as SSIM or cross-correlation) between two adjacent frames; when the similarity is below a fixed threshold, motion is considered to exist at that location. This method is susceptible to noise in single-layer images and abrupt changes in local anatomical structures (such as organ boundaries, the junction of cavities and solid tissues), resulting in numerous false positives. Gradient- or edge-based methods infer motion from the degree of edge blurring, but they are sensitive to noise and struggle to distinguish between motion-induced blurring and low contrast caused by normal anatomy.

[0027] To address the aforementioned issues, this application provides relevant solutions, which are detailed below.

[0028] According to an embodiment of this application, a method embodiment for artifact recognition is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0029] The methods and embodiments provided in this application can be executed on mobile terminals, computer terminals, or similar computing devices. This method is not only applicable to CT / PET images, but also to scenarios such as industrial X-ray inspection and visible light surface defect detection. Figure 1 A hardware block diagram of a computer terminal (or mobile device) for implementing an artifact recognition method is shown. Figure 1 As shown, the computer terminal 10 (or mobile device 10) may include one or more processors 102 (shown as 102a, 102b, ..., 102n in the figure) 102 (processor 102 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0030] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0031] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the artifact recognition method in this embodiment. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned artifact recognition method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0032] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0033] The display may be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0034] Under the above operating environment, this application provides an artifact recognition method, such as... Figure 2 As shown, the method includes the following steps:

[0035] Step S202: Determine the first image similarity between the target scanned image and the first reference image in the scanned image sequence, and the second image similarity between the target scanned image and the second reference image. The target scanned image is any image in the scanned image sequence other than the first image and the last image. The first reference image and the second reference image are adjacent to the target scanned image in the scanned image sequence. In the scanned image sequence, the first reference image is located before the target scanned image, and the second reference image is located after the target scanned image.

[0036] In the technical solution provided in step S202, the step of determining the image similarity between the target scan image and the reference image in the scan image sequence includes: determining the human image region in the target scan image and the human image region in the reference image; determining the structural similarity between the human image region in the target scan image and the human image region in the reference image; and using the structural similarity as the image similarity between the target scan image and the reference image.

[0037] In some embodiments of this application, the above-described scan image sequence is also referred to as a tomographic image sequence or a CT tomographic scan image sequence. Assume the tomographic image sequence has N frames after being sorted by slice number (or physical location), and denote the i-th frame as... ,in The human body region mask can then be obtained through thresholding combined with morphological processing. This eliminates interference from background areas in the image.

[0038] Step S204: Determine the image similarity deviation value corresponding to the target scanned image based on the smaller image similarity between the first image similarity and the second image similarity, and the first benchmark similarity.

[0039] In some embodiments of this application, the first baseline similarity value can be 1, and the image similarity deviation value can be equal to the first baseline similarity minus the smaller of the first image similarity and the second image similarity.

[0040] Optionally, a similarity deviation value can be defined. Let the preset motion determination threshold (i.e., the first preset threshold) be τ (corresponding to...). Figure 3 The SSIM threshold line in the code, such as τ = 0.08 corresponding to SSIM = 0.92). At that time, the layer was marked as a suspicious flag. . Figure 3 The continuous decreasing interval in the context refers to the interval in which the image similarity deviation values ​​corresponding to multiple consecutive frames (also known as multi-layer images) are not greater than the first preset threshold.

[0041] In some embodiments of this application, for two adjacent scanned images in a scanned image sequence and The following formula can be used to calculate its position in the human body's union region. Internal structural similarity index :

[0042]

[0043] Step S206: Based on the image similarity deviation value, determine the abnormal scan image set in the scan image sequence, wherein the abnormal scan image set includes abnormal scan images that are determined to have artifacts.

[0044] In the technical solution provided in step S206, the step of determining the abnormal scan image set in the scan image sequence based on the image similarity change information includes: determining a first preset threshold; determining the abnormal scan image set based on the first preset threshold and the image similarity deviation value corresponding to the target scan image, wherein the image similarity deviation value corresponding to the scan image in the abnormal scan image set is less than the first preset threshold, and the number of scan images in the abnormal scan image set is not less than the first preset number.

[0045] After calculating the similarity deviation value of the target scan image and obtaining the suspicious markers corresponding to each layer. Then, suspicious markers in each layer of the scanned image sequence can be identified. Perform a continuity consistency check. For example... Figure 3 As shown, this determination process is achieved by identifying the width of abnormal intervals in the structural similarity index curve on the spatial axis. During the determination, the corresponding images of each layer in the scanned image sequence are sequentially traversed. Numerical values, statistically suspicious indicators =1 The number of consecutively occurring layers, as obtained from statistics. =1 When the number of consecutively occurring layers reaches or exceeds a preset first number K, the consecutive interval is determined to be an abnormal scan image interval, and motion artifacts are confirmed to exist in the scan image within this interval. If a suspicious flag is displayed... =1 If the number of single-point occurrences or consecutive occurrences in layers does not reach the first preset quantity K, then this type of anomaly is determined to be a single anomaly caused by noise or artifact interference, and motion artifact marking is not performed on the scan image corresponding to this type of anomaly. After completing the continuity consistency determination, the abnormal scan image intervals with motion artifacts are marked, and the starting layer number a and ending layer number b corresponding to the abnormal scan image interval are recorded. The interval [a, b] is taken as the final output motion artifact occurrence range, which is the above-mentioned abnormal scan image set. At the same time, the corresponding layers in this interval are... Assigning a value of 1 allows us to pinpoint the exact location and coverage area of ​​motion artifacts in the scanned image sequence using this interval information. The value is used to identify whether the scanned image of this layer is an abnormal image. A value of 1 indicates that the image is an abnormal scanned image.

[0046] In some embodiments of this application, such as Figure 4 As shown, after identifying the set of abnormal scan images in the scanned image sequence, the artifact regions can be further determined using the following methods:

[0047] Step S402: Perform image block processing on the abnormal scan images in the abnormal scan image set to obtain multiple sets of image blocks, wherein the image blocks in the same set of image blocks have the same relative position in the abnormal scan images.

[0048] Step S404: Determine the similarity between the target image block and the first image block of the first reference image block in the image block set, and the similarity between the target image block and the second image block of the second reference image block. The target image block is any image block in the image block set except for the first and last image blocks. The first reference image block and the second reference image block are adjacent to the target image block. In the image block set, the first reference image block is located before the target image block, and the second reference image block is located after the target image block.

[0049] Step S406: Based on the smaller image patch similarity between the first image patch similarity and the second image patch similarity, and the second baseline similarity, determine the image patch similarity deviation value corresponding to the target image patch;

[0050] Step S408: Based on the image block similarity deviation value, determine the artifact image block set in the image block set, and determine the artifact region in the abnormal scan image based on the artifact image block set.

[0051] In some embodiments of this application, determining the artifact image block set in the image block set based on the image block similarity deviation value includes: determining a second preset threshold; determining the artifact image block set based on the second preset threshold and the image block similarity deviation value corresponding to the target image block, wherein there are a second preset number of consecutive image blocks in the artifact image block set whose image block similarity deviation value is not less than the second preset threshold. In some embodiments of this application, determining the artifact region in an abnormal scan image based on the artifact image block set includes: determining artifact image blocks belonging to the target abnormal scan image in the artifact image block set, wherein the target abnormal scan image is any abnormal scan image; stitching together the artifact image blocks belonging to the target abnormal scan image to obtain the artifact region in the target abnormal scan image.

[0052] After identifying the abnormal scanned image interval [a, b] containing motion artifacts in the scanned image sequence, to further improve the accuracy of motion artifact localization, a unified image block processing is performed on all abnormal scanned images within this interval. Each abnormal scanned image is divided into multiple image blocks according to the same block division rules. Based on the division results, multiple sets of image blocks are obtained. The relative positions of image blocks in the same image block set remain consistent in the corresponding abnormal scanned image. Subsequently, local structural similarity analysis is performed on each set of image blocks to determine the similarity between the target image block in each set and the first image block of the first reference image block, as well as the similarity between the target image block and the first image block of the first reference image block. The second reference image block similarity is calculated, where the target image block is any image block in the image block set except for the first and last image blocks. Both the first and second reference image blocks are adjacent to the target image block, with the first reference image block preceding the target image block and the second reference image block following the target image block in the image block set. Then, based on the smaller similarity value between the first and second reference image blocks, and combined with the second baseline similarity, the image block similarity deviation value corresponding to each target image block is calculated. .

[0053] After calculating the image block similarity deviation values ​​for all target image blocks, a second preset threshold for local determination is determined. The image block similarity deviation value corresponding to each image block is calculated. With the second preset threshold Compare and statistically analyze the similarity deviation values ​​of image patches. The number of consecutively occurring layers in an image block is not less than the second preset threshold τ_b. When the number of consecutively occurring layers reaches or exceeds the second preset number L, it is determined that there is motion artifact in the spatial region corresponding to the image block, and the image block of this type is included in the artifact image block set. If the image block similarity deviation value of the image block is... Only a single point exceeds the second preset threshold If the number of consecutive layers exceeding the threshold is less than the second preset number L, then the anomaly of this type of image block is determined to be caused by noise or artifact interference, and it will not be included in the artifact image block set.

[0054] After determining the set of artifact image blocks, artifact image blocks belonging to each target anomalous scan image are selected from the set. All artifact image blocks belonging to the same target anomalous scan image are spatially merged. The merged region is used to form an artifact region mask. Finally, the specific spatial distribution of motion artifacts in each anomalous scan image is accurately obtained through this mask, so as to achieve accurate positioning of motion artifacts in the anomalous scan image interval [a, b].

[0055] As an optional implementation, the step of determining the artifact region in the abnormal scan image based on the artifact image block set includes: determining the relative position corresponding to the artifact image block set; and determining the artifact region in the abnormal scan image based on the relative position.

[0056] After identifying the abnormal scanned image interval [a, b] containing motion artifacts in the scanned image sequence and completing the screening and determination of the artifact image block set, as an optional implementation method, the artifact region in the abnormal image is determined based on the relative position corresponding to the artifact image block set. First, basic information is extracted and organized for all image blocks in the artifact image block set. The extracted information includes the image block set number to which each artifact image block belongs, the layer sequence position of the image block in the corresponding image block set, and, most importantly, the relative position information of the image block in the corresponding abnormal scanned image. The relative position information includes the row and column coordinate range of the image block in the abnormal scanned image, the coordinates of the starting and ending pixels of the block region, the size of the block, and other key positioning information. This information has been calibrated during the previous unified image block processing of the abnormal scanned images, and all abnormal scanned images follow the same block rules. Therefore, the relative position information of each image block is unique and uniform, and can be directly used as the basis for artifact region localization.

[0057] After information extraction is completed, the relative position information of all extracted artifact image blocks can be classified and integrated. They can be grouped according to the target abnormal scan image to which they belong. That is, the relative position information of all artifact image blocks belonging to the same abnormal scan image is collected into the same group. At the same time, invalid or duplicate data that may occur during the information extraction process is removed to ensure that each group of relative position information can accurately correspond to a single abnormal scan image, thus laying an accurate and orderly data foundation for the subsequent determination of artifact regions.

[0058] After extracting and classifying the relative position information of the artifact image blocks, the artifact regions of each target anomalous scan image can be delineated and determined one by one based on the integrated relative position information. For the relative position information of each group of corresponding single anomalous scan images, the spatial region corresponding to each artifact image block is marked one by one in the pixel coordinate system of the anomalous scan image according to the row and column coordinates, pixel coordinates, etc. During the marking process, the size and coordinate calibration rules of the blocks in the early stage are strictly followed to ensure that the marked area of ​​each artifact image block completely coincides with the actual block area, without any offset or deviation.

[0059] After marking all artifact image blocks, spatial correlation analysis can be performed on all marked artifact image blocks in the same anomalous scan image to determine whether adjacent artifact image blocks have overlapping pixels or adjacent row and column positions. For artifact image blocks with overlapping or adjacent relationships, they are spatially merged to form a continuous whole region. For independent and unrelated artifact image blocks, their independent region shapes are preserved without fusion processing. After completing region marking and fusion, the overall boundary range of the merged whole region and independent regions is determined based on the pixel coordinate system of the anomalous scan image. This includes the minimum row coordinates, maximum row coordinates, minimum column coordinates, maximum column coordinates of the region, as well as the total number of pixels and pixel range covered by the overall region. The spatial region defined by these boundary ranges can then be used as the finally determined artifact region in the anomalous scan image.

[0060] For all target anomalous scanned images within the anomalous scanned image interval [a, b], the artifact regions are determined sequentially according to the same steps and rules described above, ensuring that the artifact regions of each anomalous scanned image can be accurately delineated based on the relative position information of the artifact image blocks. The entire process relies on the standardized relative position information formed by the unified block division in the early stage, eliminating the need for additional stitching processing of the artifact image blocks. The determination of artifact regions can be completed directly through the marking, integration, and fusion of position information.

[0061] This method employs a method to determine the first image similarity between a target scanned image and a first reference image, and the second image similarity between the target scanned image and a second reference image in a scanned image sequence. The target scanned image is any image in the scanned image sequence excluding the first and last images. The first and second reference images are adjacent to the target scanned image in the scanned image sequence, with the first reference image preceding the target scanned image and the second reference image following it. Based on the smaller of the first and second image similarities and a first baseline similarity, an image similarity deviation value corresponding to the target scanned image is determined. Based on the image similarity deviation value, an abnormal scanned image set is determined in the scanned image sequence, including abnormal scanned images determined to contain artifacts. By determining the image similarity deviation value of the scanned images and determining the abnormal scanned image set based on the image similarity deviation value, the influence of single-layer image noise is avoided, thereby improving the technical effect of artifact recognition accuracy. This solves the technical problem in related technologies where image artifact recognition is easily affected by single-layer image noise, leading to inaccurate artifact recognition.

[0062] This application provides an artifact recognition device. Figure 5This is a schematic diagram of the device. From... Figure 5 As can be seen from the diagram, the device includes: a first processing module 50, used to determine the first image similarity between a target scanned image and a first reference image, and the second image similarity between a target scanned image and a second reference image in the scanned image sequence, wherein the target scanned image is any image in the scanned image sequence other than the first image and the last image, the first reference image and the second reference image are adjacent to the target scanned image in the scanned image sequence, and in the scanned image sequence, the first reference image is located before the target scanned image and the second reference image is located after the target scanned image; a second processing module 52, used to determine the image similarity deviation value corresponding to the target scanned image based on the smaller image similarity between the first image similarity and the second image similarity, and a first reference similarity; and a third processing module 54, used to determine an abnormal scanned image set in the scanned image sequence based on the image similarity deviation value, wherein the abnormal scanned image set includes abnormal scanned images that are determined to have artifacts.

[0063] In some embodiments of this application, the step of the first processing module 50 in determining the image similarity between the target scan image and the reference image in the scan image sequence includes: determining a human image region in the target scan image and a human image region in the reference image; determining the structural similarity between the human image region in the target scan image and the human image region in the reference image; and using the structural similarity as the image similarity between the target scan image and the reference image.

[0064] In some embodiments of this application, the third processing module 54 determines an abnormal scan image set in the scan image sequence based on the image similarity deviation value, including: determining a first preset threshold; determining an abnormal scan image set based on the first preset threshold and the image similarity deviation value corresponding to the target scan image, wherein the image similarity deviation value corresponding to the scan image in the abnormal scan image set is less than the first preset threshold, and the number of scan images in the abnormal scan image set is not less than a first preset number.

[0065] In some embodiments of this application, after determining the abnormal scan image set in the scan image sequence, the third processing module 54 is further configured to: perform image block processing on the abnormal scan images in the abnormal scan image set to obtain multiple sets of image block sets, wherein the image blocks in the same image block set have the same relative position in the abnormal scan images; determine the first image block similarity of the target image block in the image block set relative to the first reference image block, and the second image block similarity of the target image block relative to the second reference image block, wherein the target image block is the image block set excluding the first and last ones. For any image block outside the image block set, the first reference image block and the second reference image block are adjacent to the target image block, and in the image block set, the first reference image block is located before the target image block and the second reference image block is located after the target image block; based on the smaller image block similarity between the first image block similarity and the second image block similarity, and the second benchmark similarity, the image block similarity deviation value corresponding to the target image block is determined; based on the image block similarity deviation value, the artifact image block set is determined in the image block set, and the artifact region in the abnormal scan image is determined based on the artifact image block set.

[0066] In some embodiments of this application, the third processing module 54 determines the artifact image block set in the image block set based on the image block similarity deviation value, including: determining a second preset threshold; determining the artifact image block set based on the second preset threshold and the image block similarity deviation value corresponding to the target image block, wherein there are a second preset number of consecutive image blocks in the artifact image block set whose image block similarity deviation value is not less than the second preset threshold.

[0067] In some embodiments of this application, the step of the third processing module 54 in determining the artifact region in the abnormal scan image based on the artifact image block set includes: determining the artifact image block belonging to the target abnormal scan image in the artifact image block set, wherein the target abnormal scan image is any abnormal scan image; and stitching together the artifact image block belonging to the target abnormal scan image to obtain the artifact region in the target abnormal scan image.

[0068] In some embodiments of this application, the step of the third processing module 54 in determining the artifact region in the abnormal scan image based on the artifact image block set includes: determining the relative position corresponding to the artifact image block set; and determining the artifact region in the abnormal scan image based on the relative position.

[0069] It should be noted that each module in the above-mentioned artifact recognition device can be a program module (for example, a set of program instructions to implement a certain function) or a hardware module. For the latter, it can be manifested in the following forms, but is not limited to them: each of the above modules is manifested as a processor, or the functions of each of the above modules are implemented by a processor.

[0070] According to an embodiment of this application, a non-volatile storage medium is also provided, which stores a program. When the program runs, it controls the device containing the non-volatile storage medium to execute the following artifact recognition method: determining a first image similarity between a target scanned image and a first reference image, and a second image similarity between the target scanned image and a second reference image in a scanned image sequence. The target scanned image is any image in the scanned image sequence excluding the first and last images. The first and second reference images are adjacent to the target scanned image in the scanned image sequence, with the first reference image preceding the target scanned image and the second reference image following the target scanned image. Based on the smaller of the first and second image similarities and a first baseline similarity, an image similarity deviation value corresponding to the target scanned image is determined. Based on the image similarity deviation value, an abnormal scanned image set is determined in the scanned image sequence, wherein the abnormal scanned image set includes abnormal scanned images determined to contain artifacts.

[0071] According to an embodiment of this application, an electronic device is also provided, including: a memory and a processor, wherein the processor is configured to run a program stored in the memory, wherein the program executes the following artifact recognition method: determining a first image similarity between a target scanned image and a first reference image in a scanned image sequence, and a second image similarity between the target scanned image and a second reference image, wherein the target scanned image is any image in the scanned image sequence excluding the first image and the last image, the first reference image and the second reference image are adjacent to the target scanned image in the scanned image sequence, and in the scanned image sequence, the first reference image is located before the target scanned image and the second reference image is located after the target scanned image; determining an image similarity deviation value corresponding to the target scanned image based on the smaller image similarity between the first image similarity and the second image similarity, and a first benchmark similarity; determining an abnormal scanned image set in the scanned image sequence based on the image similarity deviation value, wherein the abnormal scanned image set includes abnormal scanned images that are determined to have artifacts.

[0072] According to an embodiment of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the following steps of an artifact recognition method: determining a first image similarity between a target scanned image and a first reference image in a scanned image sequence, and a second image similarity between the target scanned image and a second reference image, wherein the target scanned image is any image in the scanned image sequence excluding the first image and the last image, the first reference image and the second reference image are adjacent to the target scanned image in the scanned image sequence, and the first reference image is located before the target scanned image and the second reference image is located after the target scanned image in the scanned image sequence; determining an image similarity deviation value corresponding to the target scanned image based on the smaller image similarity between the first image similarity and the second image similarity, and a first baseline similarity; and determining an abnormal scanned image set in the scanned image sequence based on the image similarity deviation value, wherein the abnormal scanned image set includes abnormal scanned images that are determined to have artifacts.

[0073] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0074] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0075] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0076] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0077] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to related technologies, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0078] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for artifact recognition, characterized in that, include: Determine the first image similarity between a target scanned image and a first reference image in a scanned image sequence, and the second image similarity between the target scanned image and a second reference image, wherein the target scanned image is any image in the scanned image sequence other than the first image and the last image, the first reference image and the second reference image are adjacent to the target scanned image in the scanned image sequence, and in the scanned image sequence, the first reference image is located before the target scanned image, and the second reference image is located after the target scanned image; Based on the smaller image similarity between the first image similarity and the second image similarity, and the first baseline similarity, the image similarity deviation value corresponding to the target scanned image is determined; Based on the image similarity deviation value, an abnormal scan image set is determined in the scan image sequence, wherein the abnormal scan image set includes abnormal scan images that are determined to contain artifacts.

2. The artifact recognition method according to claim 1, characterized in that, Based on the image similarity deviation value, the set of abnormal scanned images in the scanned image sequence includes: Determine the first preset threshold; Based on the first preset threshold and the image similarity deviation value corresponding to the target scanned image, the abnormal scanned image set is determined, wherein the image similarity deviation value corresponding to the scanned images in the abnormal scanned image set is less than the first preset threshold, and the number of scanned images in the abnormal scanned image set is not less than a first preset number.

3. The artifact recognition method according to claim 1, characterized in that, Determining the image similarity between a target scan image and the target scan image in a scan image sequence includes: Determine the human body image region in the target scan image, and the human body image region in the reference image; Determine the structural similarity between the human body image region in the target scan image and the human body image region in the reference image; The structural similarity is used as the image similarity between the target scanned image and the reference image.

4. The artifact recognition method according to claim 1, characterized in that, After identifying the set of abnormal scan images in the scanned image sequence, the method further includes: The abnormal scan images in the abnormal scan image set are processed by image block segmentation to obtain multiple sets of image blocks, wherein the image blocks in the same set of image blocks have the same relative position in the abnormal scan images; Determine the first image block similarity in the image block set relative to the first reference image block, and the second image block similarity in the image block set relative to the second reference image block, wherein the target image block is any image block in the image block set except for the first and last image blocks, the first reference image block and the second reference image block are adjacent to the target image block, and in the image block set, the first reference image block is located before the target image block, and the second reference image block is located after the target image block; Based on the smaller image patch similarity between the first image patch similarity and the second image patch similarity, and the second baseline similarity, the image patch similarity deviation value corresponding to the target image patch is determined; Based on the image block similarity deviation value, an artifact image block set is determined in the image block set, and an artifact region in the abnormal scan image is determined based on the artifact image block set.

5. The artifact recognition method according to claim 4, characterized in that, Based on the image patch similarity deviation value, determining the artifact image patch set in the image patch set includes: Determine the second preset threshold; Based on the second preset threshold and the image block similarity deviation value corresponding to the target image block, the artifact image block set is determined, wherein there are a second preset number of consecutive image blocks in the artifact image block set whose image block similarity deviation value is not less than the second preset threshold.

6. The artifact recognition method according to claim 4, characterized in that, Determining the artifact regions in the abnormal scan image based on the artifact image block set includes: In the set of artifact image blocks, artifact image blocks belonging to the target abnormal scan image are determined, wherein the target abnormal scan image is any one of the abnormal scan images; By stitching together artifact image blocks belonging to the target abnormal scan image, the artifact region in the target abnormal scan image is obtained.

7. The artifact recognition method according to claim 4, characterized in that, Determining the artifact regions in the abnormal scan image based on the artifact image block set includes: Determine the relative positions corresponding to the set of artifact image blocks; The artifact region in the abnormal scan image is determined based on the relative position.

8. An artifact detection device, characterized in that, include: A first processing module is configured to determine the first image similarity between a target scanned image and a first reference image in a scanned image sequence, and the second image similarity between the target scanned image and a second reference image, wherein the target scanned image is any image in the scanned image sequence other than the first image and the last image, the first reference image and the second reference image are adjacent to the target scanned image in the scanned image sequence, and in the scanned image sequence, the first reference image is located before the target scanned image, and the second reference image is located after the target scanned image; The second processing module is used to determine the image similarity deviation value corresponding to the target scanned image based on the smaller image similarity between the first image similarity and the second image similarity, and the first benchmark similarity. The third processing module is used to determine an abnormal scan image set in the scan image sequence based on the image similarity deviation value, wherein the abnormal scan image set includes abnormal scan images that are determined to have artifacts.

9. A non-volatile storage medium, characterized in that, The non-volatile storage medium stores a program, wherein when the program is executed, it controls the device where the non-volatile storage medium is located to execute the artifact recognition method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, include: A memory and a processor, the processor being configured to run a program stored in the memory, wherein the program, when running, performs the artifact recognition method according to any one of claims 1 to 7.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the artifact recognition method according to any one of claims 1 to 7.

Images