Sample grading method, device and electronic equipment

Abstract

The present application provides a sample grading method, apparatus and electronic device. The present application grades a target sample by means of a multimodal imaging method, that is, an X-ray image obtained by scanning the target sample based on X-rays and at least one sample surface image of the surface of the target sample collected by at least one data acquisition module, so as to improve the accuracy of grading the target sample. Further, in this embodiment, by visually displaying the grading evaluation results of different surface regions in the target sample, such as the fat distribution characteristics of a piece of meat, the grading evaluation results of different surface regions in the target sample, such as the fat distribution characteristics of a piece of meat, are intuitively displayed, and then a more refined quality evaluation of the target sample, such as meat, is realized.

Description

Technical Field

[0001] The present application relates to data processing technologies, and particularly to a method and apparatus for sample grading and an electronic device. Background Art

[0002] In specific applications, samples are often graded to facilitate standardized management of the samples. Taking meat samples as an example, meat grading aims to standardize meat quality, facilitating market transactions and consumer purchases.

[0003] When grading samples, it often relies on the characteristic data of the samples. Still taking meat samples as an example, for meat samples, their characteristic data such as fat content. Currently, it often relies on the manual visual inspection method to grade samples such as meat samples. However, due to the relatively large volume of the samples and the uneven distribution of their characteristics, the manual visual inspection method often has the risk of inaccurate grading. Still taking meat samples as an example, for example, when grading a piece of meat, but because the piece of meat has a relatively large volume and is streaky pork with a very uneven fat and lean distribution, the manual visual inspection method often has the risk of inaccurate grading of the piece of meat. Summary of the Invention

[0004] The present application provides a method and apparatus for sample grading and an electronic device, so as to grade the sample by cooperating with the data of the sample collected in a multi-modal manner, and improve the accuracy of sample grading.

[0005] An embodiment of the present application provides a method for sample grading, and the method includes:

[0006] Determine the characteristic data of each image unit in the X-ray image; the X-ray image is an image obtained by scanning a target sample with X-rays of at least two energies; the characteristic data of any image unit is used to represent the component information of the position corresponding to the image unit in the target sample;

[0007] Perform image registration on at least one sample surface image and the X-ray image, and perform grading evaluation processing on different surface regions of the target sample according to the registration result and the characteristic data of each image unit; the at least one sample surface image is an image obtained by using at least one data acquisition module to collect the surface of the target sample; the grading evaluation result of any surface region is determined based on the predicted component information of the surface region and the region block characteristic data of the region block corresponding to the surface region; the region block corresponding to any surface region refers to a region block from the surface region to the bottom of the target sample, and the region block characteristic data of any region block is determined according to the characteristic data of each image unit in the region block, and the region block characteristic data of any region block characterizes the component information of the region block;

[0008] Visually display the grading evaluation results of different surface regions of the target sample.

[0009] A sample grading device, the device comprising:

[0010] A determination module for determining the characteristic data of each image unit in the X-ray image; the X-ray image is an image obtained by scanning a target sample with X-rays of at least two energies; the characteristic data of any image unit is used to represent the composition information of the position corresponding to the image unit in the target sample;

[0011] A processing module for performing image registration on at least one sample surface image and the X-ray image, and performing grading evaluation processing on different surface regions of the target sample according to the registration result and the characteristic data of each image unit; the at least one sample surface image is an image obtained by using at least one data acquisition module to acquire the surface of the target sample; the grading evaluation result of any surface region is determined based on the predicted composition information of the surface region and the region block characteristic data of the region block corresponding to the surface region; the region block corresponding to any surface region refers to a region block from the surface region to the bottom of the target sample in the target sample, the region block characteristic data of any region block is determined according to the characteristic data of each image unit in the region block, and the region block characteristic data of any region block characterizes the composition information of the region block;

[0012] A display module for visually displaying the grading evaluation results of different surface regions of the target sample.

[0013] An electronic device, the electronic device comprising a processor and a machine-readable storage medium; a plurality of computer instructions are stored on the machine-readable storage medium, and when the computer instructions are executed by the processor, the steps in the above method are implemented.

[0014] It can be seen from the above technical solutions that in this embodiment, the target sample is graded by means of a multimodal imaging method, that is, an X-ray image obtained by scanning the target sample with X-rays and at least one sample surface image of the surface of the target sample acquired by at least one data acquisition module are combined to improve the accuracy of grading the target sample.

[0015] Furthermore, in this embodiment, by visually displaying the grading evaluation results of different surface regions of the target sample, such as the fat distribution characteristics of a piece of meat, the grading evaluation results of different surface regions of the target sample, such as the fat distribution characteristics of a piece of meat, are intuitively displayed, and further, a more refined quality evaluation of the target sample, such as meat, is realized. Description of the Drawings

[0016] The drawings here are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure.

[0017] Figure 1Flowchart of the method provided by the embodiments of the present application;

[0018] Figure 2 Flowchart for implementing step 102 provided by the embodiments of the present application;

[0019] Figure 3 Schematic diagram of the device structure provided by the embodiments of the present application;

[0020] Figure 4 Structural diagram of the electronic device provided by the embodiments of the present application. Detailed implementation manners

[0021] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0022] The terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. The singular forms "a", "the", and "said" used in the present application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0023] In order to enable those skilled in the art to better understand the technical solutions provided by the embodiments of the present application and make the above objects, features, and advantages of the embodiments of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be further described in detail below with reference to the drawings.

[0024] See Figure 1 , Figure 1 which is the flowchart of the method provided by the embodiments of the present application. This method is applied to an electronic device. As Figure 1 shown, the process may include the following steps:

[0025] Step 101, determining the characteristic data of each image unit in the X-ray image; the X-ray image is an image obtained by scanning a target sample with X-rays of at least two energies; the characteristic data of any image unit is used to represent the composition information of the position corresponding to the image unit in the target sample.

[0026] As an embodiment, the target sample here may be a piece of meat.

[0027] In specific implementation, this embodiment will use a direct digital radiography system (DR: Digit Radiography), or use CT to emit X-rays of at least two energies to the target sample to scan and obtain a tomographic image of the target sample (i.e., the above-mentioned X-ray image). Optionally, the DR can be multi-energy or dual-energy DR. The CT can be multi-energy or dual-energy CT.

[0028] In this embodiment, the structure of the X-ray image can be a two-dimensional or three-dimensional image structure. For example, the structure of the X-ray image obtained by scanning with DR emitting X-rays is a two-dimensional image structure, and the structure of the X-ray image obtained by scanning with CT emitting X-rays is a three-dimensional image structure. If the structure of the X-ray image is different, the image units in the X-ray image are different. For example, for a two-dimensional X-ray image, its image unit is a single pixel point, while for a three-dimensional X-ray image, its image unit is a single volume element (abbreviated as voxel).

[0029] In this embodiment, the characteristic data of any image unit is used to represent the composition information of the position in the target sample corresponding to this image unit. Still taking the target sample as a piece of meat as an example, the characteristic data of any image unit is used to represent the composition information of the position in this piece of meat corresponding to this image unit, such as fat data or muscle data.

[0030] In specific implementation, the determination method of the characteristic data of any image unit in the X-ray image is related to the imaging mode of the X-ray image. If the imaging mode is different, the above determination method is also different. For example, if the imaging mode of the X-ray image is the DR mode (for example, at this time the X-ray image is obtained by scanning with DR emitting X-rays), the basis material decomposition algorithm is often used to determine the characteristic data of any image unit in the X-ray image, such as the fat ratio. For another example, if the imaging mode of the X-ray image is the CT mode (for example, at this time the X-ray image is obtained by scanning with CT emitting X-rays), the material identification and segmentation algorithm is often used to determine the characteristic data of any image unit in the X-ray image, such as the fat density value.

[0031] That is to say, this embodiment can apply the relatively mature basis material decomposition algorithm or material identification and segmentation algorithm that was not originally applied in the meat scenario to the meat grading scenario. Taking the basis material decomposition algorithm as an example, in meat detection, when dual-energy or multi-energy X-rays pass through a piece of meat, such as a piece of meat with fat and muscle superimposed, the signal received by the detector is:

[0032]

[0033] ;

[0034] Among them, is the signal emission intensity, represents the signal reception intensity received by the detector. is the mass attenuation coefficient of fat and muscle in the th energy interval, and this value can be obtained based on theory or experiment.

[0035] If the matrix of the attenuation coefficient is , with a size of , is the total number of energy intervals, where , . If the image matrix is , with a size of , ; if the mass thickness matrix is , with a size of , , and the definition of mass thickness is density multiplied by thickness. Based on the basis material decomposition algorithm, is solved to obtain , then the fat mass , muscle mass and total mass at the position of each pixel point in the target sample can be obtained. Correspondingly, the fat percentage at this position is .

[0036] Here, taking the fat ratio as the characteristic data as an example, in this embodiment, the characteristic data of each image unit, that is, the fat ratio, will be integrated in the spatial domain, and finally a full-field quantization map containing the fat distribution characteristics will be generated.

[0037] Step 102, perform image registration on at least one sample surface image and the above X-ray image, and perform hierarchical evaluation processing on different surface regions of the target sample according to the registration result and the characteristic data of each image unit.

[0038] In a specific application, since the X-ray image lacks the depth information of each position in the target sample, relying solely on the X-ray image to grade the target sample may also have the risk of inaccurate grading. Based on this, this embodiment will also use at least one data acquisition module, such as at least one of a visible light imaging device, a multispectral sensor, and a hyperspectral acquisition device, to acquire the surface image of the target sample to obtain at least one sample surface image. Still taking the target sample as a piece of meat as an example, the obtained sample surface images characterize the surface information of this piece of meat.

[0039] As an embodiment, after obtaining at least one sample surface image, necessary preprocessing, such as at least one of image correction, image noise reduction, and principal component analysis, can be performed on each sample surface image to reduce the data volume and remove noise data.

[0040] After that, the surface images of various samples can be image-registered with the above X-ray images. Optionally, in this embodiment, the surface images of various samples can be aligned with the X-ray images according to a set alignment method, so that the surface images of various samples and the X-ray images are spatially aligned. As an embodiment, the set alignment method is, for example, at least one of methods such as feature matching algorithm, Scale-Invariant Feature Transform (SIFT), etc.

[0041] After that, different surface regions in the target sample are subjected to hierarchical evaluation processing according to the registration results and the feature data of each image unit. Here, the hierarchical evaluation result of any surface region is determined based on the predicted composition information of the surface region and the region block feature data of the region block corresponding to the surface region; the region block corresponding to any surface region refers to a region block in the target sample from the surface region to the bottom of the target sample, and the region block feature data of any region block is determined based on the feature data of each image unit in the region block, and the region block feature data of any region block characterizes the composition information of the region block. How to perform hierarchical evaluation processing on different surface regions in the target sample according to the registration results and the feature data of each image unit will be described by way of example below.

[0042] Step 103, visually display the hierarchical evaluation results of different surface regions in the target sample.

[0043] For example, the hierarchical evaluation results are projected onto an interactive screen in real time through a projection display device to form a dynamic visualization interface for real-time quality grade determination of the target sample, such as meat, based on display results such as fat distribution characteristics. Applied to the meat grading scenario, this visual display can be used to guide subsequent operations such as cutting the meat.

[0044] So far, the Figure 1 shown process is completed.

[0045] Through Figure 1 As can be seen from the shown process, in this embodiment, the target sample is graded by means of a multimodal imaging method, that is, an X-ray image obtained by X-ray scanning the target sample and at least one sample surface image of the surface of the target sample collected by at least one data acquisition module are combined to improve the accuracy of grading the target sample.

[0046] Furthermore, in this embodiment, by visually displaying the hierarchical evaluation results of different surface regions in the target sample, such as the fat distribution characteristics of a piece of meat, the hierarchical evaluation results of different surface regions in the target sample, such as the fat distribution characteristics of a piece of meat, are intuitively displayed, and further, a more refined quality evaluation of the target sample, such as meat, is realized.

[0047] The following describes how to perform hierarchical evaluation processing on different surface regions in the target sample according to the registration results and the feature data of each image unit in step 102:

[0048] Optionally, the grading evaluation result of any surface area indicates the consistency judgment result between the predicted component information of the surface area and the component information of the region block corresponding to the surface area. The following is a description:

[0049] See Figure 2 , Figure 2 which is the flowchart for implementing step 102 provided by the embodiment of the present application. As Figure 2 shown, the process may include the following steps:

[0050] Step 201, obtain the predicted component information of each different surface area on the upper surface of the target sample.

[0051] It should be noted that in this embodiment, the predicted component information of each different surface area on the upper surface of the target sample can be implemented based on an image classification module. Taking the target sample as a piece of meat as an example, the image classification module is used to distinguish whether the component information of each area (denoted as the surface area) on the upper surface of the piece of meat is muscle or fat, so as to distinguish the components (muscle or fat) of each different surface area on the upper surface of the piece of meat. The image classification module can be a region component recognition model constructed by at least one of SVM, convolutional neural network, etc.

[0052] Step 202, for each surface area, determine the region block corresponding to the surface area from the target sample based on the above registration result, and determine the region block feature data of the region block according to the feature data of each image unit in the region block. The region block feature data characterizes the component information of the region block; perform a consistency judgment on the region block feature data and the predicted component information of the surface area.

[0053] For example, if the target sample is a piece of meat, assuming a surface area is determined, the region block corresponding to the surface area is a small piece of meat framed from top to bottom along the region edge of the surface area.

[0054] After determining the region block, determine the region block feature data of the region block according to the feature data of each image unit in the region block. For example, taking the target sample as a piece of meat and the feature data of the image unit as the fat ratio as an example, calculate the average value of the fat ratios of each image unit in the region block to obtain the average fat ratio, and the average fat ratio is the region block feature data of the region block, which characterizes the components of the entire region block. For example, if the average fat ratio of the region block is greater than or equal to the fat setting threshold, then determine that the entire region block is fat; conversely, if the average fat ratio of the region block is less than the fat setting threshold, then determine that the entire region block is lean meat.

[0055] After that, a consistency judgment is made on the region block feature data and the predicted component information of the surface region obtained in the above step 201 (such as the component information of the surface region predicted based on the above image classification module). For example: compare whether the region block feature data and the predicted component information of the surface region obtained in the above step 201 (such as the component information of the surface region predicted based on the above image classification module) meet the set consistency requirements. If the set consistency requirements are met, it is determined that the surface region is a region with good consistency (at this time, the region can have a corresponding identification value such as 1, indicating that the set consistency requirements are met). At this time, the component of the surface region can be the predicted component information of the surface region. Otherwise, it is determined that the surface region is a region with poor consistency (at this time, the region can have a corresponding identification value such as 0, indicating that the set consistency requirements are not met). At this time, it can be marked that the component of the surface region is uncertain or may be different from the predicted component information of the surface region).

[0056] Taking the target sample as a piece of meat and the region block feature data as the average fat percentage of the region block as an example, the above region block feature data and the predicted component information of the surface region meeting the set consistency requirements means that: if the predicted component information of the surface region is fat and the average fat percentage is greater than or equal to the set fat threshold, it is determined that the set consistency requirements are met. Vice versa.

[0057] Optionally, in this embodiment, for each surface region, it can have a corresponding consistency degree to characterize the degree to which the region block feature data of the surface region and the predicted component information of the surface region (such as the component information of the surface region predicted based on the above image classification module) meet the set consistency requirements. For example, the consistency degree of the surface region can be represented by the residual between the characteristic value corresponding to the predicted component information of the surface region (used to represent the component information) and the region block feature data of the surface region. Optionally, in this embodiment, the characteristic value corresponding to the predicted component information of any surface region is related to the component information. For example, when the component information is fat, the characteristic value is 0, and when the component information is muscle, the characteristic value is 1, etc. This embodiment does not specifically limit.

[0058] So far, the Figure 2 shown process is completed.

[0059] Through the Figure 2 shown process, how to perform hierarchical evaluation processing on different surface regions of the target sample is realized.

[0060] Optionally, in this embodiment, the characteristic data of each image unit in the above X-ray image can also be subjected to component analysis, and the spatial distribution characteristics of the target sample are visually displayed; the spatial distribution characteristics are represented by the quantization indexes of each image unit; the quantization index of any image unit is the mass of the position corresponding to the image unit in the target sample, and / or the component distribution information determined based on the characteristic data of the image unit. Preferably, visualization presentation means such as three-dimensional stereoscopic rendering technology or pseudo-three-dimensional projection technology can be used to visually display the spatial distribution characteristics of the target sample. Additionally, if the target sample is a piece of meat, the component distribution information determined based on the characteristic data of the image unit is: fat proportion and / or muscle proportion, fat and / or muscle mass distribution.

[0061] It should be noted that the above spatial distribution characteristics can be at least one of a heat map, a contour map, a two-dimensional histogram, etc.

[0062] Optionally, in this embodiment, the dynamic rendering engine can be configured to adjust the rendering parameters in real time according to the user interaction instruction to achieve intuitive display. For example, by configuring the dynamic rendering engine to adjust the rendering parameters in real time according to the user interaction instruction, the spatial distribution characteristics and quantization ratios of different tissue components in the target sample can be intuitively displayed, thereby helping the user to perform a more refined quality assessment of the target sample such as meat.

[0063] Furthermore, in this embodiment, a digital identification code can be generated based on the hierarchical evaluation results of different surface regions of the target sample; the digital identification code is represented by a quality parameter matrix, and the quality parameter matrix is determined based on the hierarchical evaluation results of different surface regions of the target sample.

[0064] As an alternative implementation, based on the description of the above hierarchical evaluation process, optionally, the quality parameter matrix here can be the identification value (whether the set consistency requirement is met) corresponding to different surface regions of the target sample. Table 1 below exemplifies the quality parameter matrix:

[0065] Table 1

[0066]

[0067] As another alternative implementation, based on the description of the above hierarchical evaluation process, optionally, the quality parameter matrix here can be the degree of consistency corresponding to different surface regions of the target sample. Table 2 below exemplifies the quality parameter matrix:

[0068] Table 2

[0069]

[0070] Optionally, in this embodiment, the digital identification code is attached to the packaging surface of the target sample. For example, the digital identification code is printed on a waterproof material substrate through a thermal transfer device to form a traceable physical label and attached to the product packaging surface of the target sample.

[0071] The method provided by the embodiments of the present application has been described above. Next, the device provided by the embodiments of the present application will be described:

[0072] See Figure 3 , Figure 3 which is the structural diagram of the device provided by the embodiments of the present application. As Figure 3 shown, the device includes:

[0073] A determination module, configured to determine the feature data of each image unit in the X-ray image; the X-ray image is an image obtained by scanning the target sample with X-rays of at least two energies; the feature data of any image unit is used to represent the component information of the position corresponding to the image unit in the target sample;

[0074] A processing module, configured to perform image registration on at least one sample surface image and the X-ray image, and perform hierarchical evaluation processing on different surface regions of the target sample according to the registration result and the feature data of each image unit; the at least one sample surface image is an image obtained by using at least one data acquisition module to acquire the surface of the target sample; the hierarchical evaluation result of any surface region is determined based on the predicted component information of the surface region and the region block feature data of the region block corresponding to the surface region; the region block corresponding to any surface region refers to a region block from the surface region to the bottom of the target sample, and the region block feature data of any region block is determined according to the feature data of each image unit in the region block, and the region block feature data of any region block characterizes the component information of the region block;

[0075] A display module, configured to visually display the hierarchical evaluation results of different surface regions of the target sample.

[0076] Optionally, the performing image registration on at least one sample surface image and the X-ray image includes:

[0077] Aligning the sample surface image and the X-ray image according to a set alignment method so that the sample surface image and the X-ray image are spatially aligned.

[0078] Optionally, the performing hierarchical evaluation processing on different surface regions of the target sample according to the registration result and the feature data of each image unit includes:

[0079] Obtaining the predicted component information of each different surface region in the upper surface of the target sample;

[0080] For each surface region, based on the registration result, determine the corresponding region block from the target sample, and determine the region block feature data of the region block according to the feature data of each image unit in the region block. The region block feature data indicates the composition information of the region block; judge the consistency between the region block feature data and the predicted composition information of the surface region;

[0081] Wherein, the grading evaluation result of any surface region indicates the consistency judgment result between the predicted composition information of the surface region and the composition information of the region block corresponding to the surface region.

[0082] Optionally, the display module is used to visually display the spatial distribution characteristics of the target sample; the spatial distribution characteristics are represented by the quantization indexes of each image unit; the quantization index of any image unit includes mass, and / or the composition distribution information determined based on the feature data of the image unit.

[0083] Optionally, the target sample is a piece of meat; the composition distribution information determined based on the feature data of the image unit is: fat ratio and / or muscle ratio.

[0084] Optionally, the processing module is used to generate a digital identification code based on the grading evaluation results of different surface regions of the target sample; the digital identification code is represented by a quality parameter matrix, and the quality parameter matrix is determined based on the grading evaluation results of different surface regions of the target sample; the digital identification code is attached to the packaging surface of the target sample.

[0085] Optionally, the target sample is a piece of meat, and the feature data of any image unit is used to represent the fat data or muscle data at the position corresponding to the image unit in the piece of meat;

[0086] The predicted composition information of any surface region is muscle or fat;

[0087] The composition information represented by the region block is determined according to the fat data or muscle data of each image unit in the region block.

[0088] Optionally, the at least one sample surface image includes an image of the surface of the target sample collected by at least one of a visible light imaging device, a multispectral sensor, and a hyperspectral acquisition device.

[0089] So far, the Figure 3 structural description of the shown device is completed.

[0090] Correspondingly, the embodiment of the present application also provides Figure 3 the hardware structure of the shown device. Refer to Figure 4 , Figure 4 which is the hardware structure diagram provided by the embodiment of the present application. AsFigure 4 As shown, the hardware result includes a processor and a machine-readable storage medium; several computer instructions are stored on the machine-readable storage medium, and when the computer instructions are executed by the processor, the steps in the above method are implemented.

[0091] Based on the same application concept as the above method, an embodiment of the present application further provides a machine-readable storage medium, on which several computer instructions are stored, and when the computer instructions are executed by the processor, the method disclosed in the above examples of the present application can be implemented.

[0092] Exemplarily, the above machine-readable storage medium can be any electronic, magnetic, optical or other physical storage device, which can contain or store information, such as executable instructions, messages, etc. For example, the machine-readable storage medium can be: RAM (Random Access Memory), volatile memory, non-volatile memory, flash memory, storage drive (such as a hard disk drive), solid-state drive, any type of storage disk (such as an optical disk, DVD, etc.), or a similar storage medium, or a combination thereof.

[0093] The systems, devices, modules or units illustrated in the above embodiments can be specifically implemented by a computer processing unit or entity, or by a product with a certain function. A typical implementation device is a computer, and the specific form of the computer can be a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email receiving and sending device, a game console, a tablet computer, a wearable device, or a combination of any several of these devices.

[0094] For the convenience of description, when describing the above devices, they are described separately as various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0095] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0096] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable message processing device to generate a machine, such that the instructions executed by the processor of the computer or other programmable message processing device produce means for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0097] Furthermore, these computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable message processing device to operate in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that implement the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0098] These computer program instructions can also be loaded onto a computer or other programmable message processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0099] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A method for sample fractionation, characterized in that, The method includes: Determining characteristic data of each image unit in the X-ray image; the X-ray image is an image obtained by scanning a target sample with X-rays of at least two energies; the characteristic data of any image unit is used to represent the composition information of the position in the target sample corresponding to the image unit; Performing image registration on at least one sample surface image and the X-ray image, and performing hierarchical evaluation processing on different surface regions of the target sample according to the registration result and the characteristic data of each image unit; the at least one sample surface image is an image obtained by using at least one data acquisition module to acquire the surface of the target sample; wherein, the performing hierarchical evaluation processing on different surface regions of the target sample according to the registration result and the characteristic data of each image unit includes: obtaining the predicted composition information of each different surface region in the upper surface of the target sample; for each surface region, determining the region block corresponding to the surface region from the target sample based on the registration result, and determining the region block characteristic data of the region block according to the characteristic data of each image unit in the region block, the region block characteristic data indicating the composition information of the region block; performing a consistency judgment on the region block characteristic data and the predicted composition information of the surface region; wherein, the hierarchical evaluation result of any surface region indicates the consistency judgment result of the predicted composition information of the surface region and the composition information of the region block corresponding to the surface region; the region block corresponding to any surface region refers to a region block from the surface region to the bottom of the target sample in the target sample; Visually displaying the hierarchical evaluation results of different surface regions of the target sample.

2. The method according to claim 1, wherein The performing image registration on at least one sample surface image and the X-ray image includes: Aligning the sample surface image and the X-ray image according to a set alignment method so that the sample surface image and the X-ray image are spatially aligned.

3. The method according to claim 1, characterized in that The method further includes: Visually displaying the spatial distribution characteristics of the target sample; the spatial distribution characteristics are represented by quantization indexes of each image unit; the quantization index of any image unit includes mass, and / or composition distribution information determined based on the characteristic data of the image unit.

4. The method according to claim 3, wherein The target sample is a piece of meat; The composition distribution information determined based on the characteristic data of the image unit is: fat proportion and / or muscle proportion.

5. The method according to claim 1, characterized in that, The method further includes: Generating a digital identification code based on the hierarchical evaluation results of different surface regions of the target sample; the digital identification code is represented by a quality parameter matrix, and the quality parameter matrix is determined based on the hierarchical evaluation results of different surface regions of the target sample; the digital identification code is attached to the packaging surface of the target sample.

6. The method according to any one of claims 1 to 5, characterized in that The target sample is a piece of meat, and the characteristic data of any image unit is used to represent the fat data or muscle data of the position in the piece of meat corresponding to the image unit; The predicted composition information of any surface region is muscle or fat; The composition information characterized by the region block is determined according to the fat data or muscle data of each image unit in the region block.

7. The method according to any one of claims 1 to 5, characterized in that, The at least one sample surface image includes an image of the surface of the target sample acquired by at least one of a visible light imaging device, a multispectral sensor, and a hyperspectral acquisition device.

8. A sample grading device, characterized in that, The device includes: a determination module configured to determine characteristic data of each image unit in the X-ray image; the X-ray image is an image obtained by scanning the target sample with X-rays of at least two energies; the characteristic data of any image unit is used to represent the composition information of the position corresponding to the image unit in the target sample; a processing module configured to perform image registration on the at least one sample surface image and the X-ray image, and perform hierarchical evaluation processing on different surface regions of the target sample according to the registration result and the characteristic data of each image unit; the at least one sample surface image is an image obtained by using at least one data acquisition module to acquire the surface of the target sample; wherein, the performing hierarchical evaluation processing on different surface regions of the target sample according to the registration result and the characteristic data of each image unit includes: obtaining the predicted composition information of each different surface region in the upper surface of the target sample; for each surface region, determining a corresponding region block of the surface region from the target sample based on the registration result, and determining the region block characteristic data of the region block according to the characteristic data of each image unit in the region block, the region block characteristic data indicating the composition information of the region block; performing a consistency judgment on the region block characteristic data and the predicted composition information of the surface region; wherein, the hierarchical evaluation result of any surface region indicates the consistency judgment result of the predicted composition information of the surface region and the composition information of the region block corresponding to the surface region; the region block corresponding to any surface region refers to a region block from the surface region to the bottom of the target sample in the target sample; a display module configured to visually display the hierarchical evaluation results of different surface regions of the target sample.

9. An electronic device, characterized in that, The electronic device includes a processor and a machine-readable storage medium; a plurality of computer instructions are stored on the machine-readable storage medium, and when the computer instructions are executed by the processor, the steps in any one of the methods according to claims 1 to 7 are implemented.

Images