Method and device for non-destructive identification of internal defects of chestnut and medium

Abstract

The application discloses a kind of nondestructive identification chestnut internal defect analysis method, device and medium, the nondestructive identification chestnut internal defect analysis method includes the following steps: obtaining the CT image of chestnut, obtains first layer;Process the first layer, obtain second layer and the first defect information of the chestnut;Process the second layer, obtain third layer and according to preset state value distinguish and extract abnormal state area in the third layer;Process and analyze the abnormal state area, obtain the second defect information of the chestnut;According to the first defect information and the second defect information, the chestnut is screened and classified.The application is processed by multiple times to the CT image of chestnut, respectively obtains the first defect information and the second defect information of chestnut internal, excludes the problem that it is difficult to form unified quality control due to artificial subjective difference, provides an effective, reliable, fast sorting identification method for judging the internal defect of chestnut.

Description

Technical Field

[0001] This invention relates to the field of food quality testing technology, and in particular to an analytical method, apparatus and medium for non-destructive identification of internal defects in chestnuts. Background Technology

[0002] Currently, the quality of chestnuts after harvesting is affected by storage conditions, especially improper temperature or humidity. Chestnuts can rot due to mold, insect infestation, or dehydration and calcification. Internal defects reduce the commercial value of chestnuts during storage. Therefore, related technologies often require manually breaking the chestnut shell and then manually judging the internal defects to screen out chestnuts with internal defects.

[0003] Therefore, mold infection, insect infestation, and dehydration calcification are common causes of chestnut spoilage. These defects usually occur inside the chestnut, and related technologies cannot detect these diseases without damaging the chestnut shell. However, damaging the chestnut shell is not conducive to chestnut storage, and manual judgment is inefficient and difficult to form a unified quality control standard due to large subjective differences. Summary of the Invention

[0004] The main objective of this invention is to provide a non-destructive analytical method, apparatus, and medium for identifying internal defects in chestnuts. This aims to address the existing technical problems that mold infection, insect infestation, and dehydration calcification are common causes of chestnut spoilage, and these defects usually occur inside the chestnut. Related technologies cannot detect these diseases without damaging the chestnut shell. However, damaging the chestnut shell is not conducive to chestnut storage, and manual judgment is inefficient and difficult to establish a unified quality control standard due to large subjective differences among humans.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] In a first aspect, the present invention provides a non-destructive analysis method for identifying internal defects in chestnuts, comprising the following steps:

[0007] Obtain the CT image of the chestnut and get the first layer;

[0008] Process the first layer to obtain the second layer and the first defect information of the chestnut;

[0009] Process the second layer to obtain the third layer, and distinguish and extract the abnormal state areas in the third layer according to preset state values;

[0010] The abnormal state area is processed and analyzed to obtain the second defect information of the chestnut.

[0011] The chestnuts are screened and classified according to the first defect information and the second defect information.

[0012] Optionally, in the above-described non-destructive analysis method for identifying internal defects in chestnuts, the first layer includes an image background layer and multiple consecutive internal cross-sectional layers. The step of processing the first layer to obtain the second layer and the first defect information of the chestnut includes:

[0013] Remove the image background layer;

[0014] The sample area is divided by the outer shell of the chestnut in each of the internal cross-sectional layers, and effective voxels are obtained based on the voxels of the sample area to obtain the second layer;

[0015] The volume average gray value, volume gray standard deviation, and porosity of the first layer are obtained based on the second layer.

[0016] The first defect information is obtained based on the volume average gray value, volume gray standard deviation, and porosity of the first layer.

[0017] Optionally, in the above-described non-destructive analysis method for identifying internal defects in chestnuts, the step of obtaining the volume average gray value, volume gray standard deviation, and porosity of the first layer based on the second layer includes:

[0018] Based on the effective voxels of the sample region, the first voxel gray value of the sample region is obtained, and based on the first voxel gray value and the number of effective voxels, the volume average gray value of the first layer is obtained.

[0019] The volume gray standard deviation is obtained based on the volume average gray value of the first layer and the number of effective voxels;

[0020] The number of voxels in the sample region with a gray value greater than 0 is obtained, and the porosity is obtained based on the number of voxels in the sample region with a gray value greater than 0 and the number of effective voxels.

[0021] Optionally, in the above-described non-destructive analysis method for identifying internal defects in chestnuts, the step of obtaining the first defect information based on the volume average gray value, volume gray standard deviation, and porosity of the first layer includes:

[0022] Determine whether the average gray value of the first layer is greater than or equal to the preset average gray value. If not, the first defect information is that the chestnut is completely rotten. If yes, proceed to the next step.

[0023] Determine whether the standard deviation of the volume grayscale is less than or equal to the preset standard deviation of grayscale. If not, the first defect information is that the chestnut is completely rotten. If yes, proceed to the next step.

[0024] Determine whether the porosity is less than or equal to a preset porosity. If not, the first defect information is that the chestnut is completely rotten. If yes, the first defect is marked as none.

[0025] Optionally, in the above-described non-destructive analysis method for identifying internal defects in chestnuts, the step of processing the second layer to obtain a third layer and distinguishing and extracting abnormal state areas in the third layer according to preset state values ​​includes:

[0026] The second layer is segmented and processed based on a threshold of the CT image to obtain the third layer;

[0027] Obtain the grayscale value of the second voxel in the third layer, compare the grayscale value of the second voxel with the first preset grayscale value, and obtain the first comparison result;

[0028] The second voxel gray value is compared with the second preset gray value to obtain a second comparison result, and the abnormal state area in the third layer is extracted.

[0029] Based on the first comparison result and the second comparison result, the second defect information of the chestnut is obtained.

[0030] Optionally, in the above-described non-destructive analysis method for identifying internal defects in chestnuts, the preset grayscale value includes a first preset grayscale value and a second preset grayscale value, wherein the first preset grayscale value is greater than the second preset grayscale value, and the step of obtaining the second defect information of the chestnut based on the first comparison result and the second comparison result includes:

[0031] Determine whether the gray value of the second voxel is less than or equal to the first preset gray value. If not, the second defect information is that the chestnut has a calcification defect. If yes, proceed to the next step.

[0032] Determine whether the gray value of the second voxel is greater than or equal to the second preset gray value. If not, extract the abnormal state area in the third layer based on the gray value of the second voxel, and obtain the second defect information based on the abnormal state area.

[0033] If so, then mark the second defect information as none.

[0034] Optionally, in the above-described non-destructive analysis method for identifying internal defects in chestnuts, the step of determining whether the gray value of the second voxel is greater than or equal to the second preset gray value, and if not, extracting the abnormal state area in the third layer based on the gray value of the second voxel, and obtaining the second defect information based on the abnormal state area, includes:

[0035] The effective voxels whose grayscale values ​​of the second voxel are less than the second preset grayscale values ​​are reconstructed, and objects whose effective voxels are less than a preset unit are filtered out to obtain multiple isolated objects.

[0036] Morphological analysis is performed on each isolated object to obtain the anisotropy and sphericity of each isolated object;

[0037] Determine whether the anisotropy of the isolated object is less than a preset anisotropy. If yes, mark the second defect information as none. If no, proceed to the next step.

[0038] Determine whether the sphericity of the isolated object is greater than or equal to a preset sphericity. If yes, the second defect information is that the chestnut has mold spots; if no, the second defect information is that the chestnut has wormholes.

[0039] Optionally, in the above-described non-destructive analysis method for identifying internal defects in chestnuts, the step of determining whether the gray value of the second voxel is greater than or equal to the second preset gray value, and if not, extracting the abnormal state area in the third layer based on the gray value of the second voxel, and obtaining the second defect information based on the abnormal state area, includes:

[0040] The effective voxels whose gray values ​​are less than the second preset gray value are reconstructed to obtain a pore network model inside the chestnut with connected voxels.

[0041] Based on the connecting voxel, the radius of the pore throat inside the chestnut is obtained;

[0042] Determine whether the throat radius is greater than or equal to the throat threshold. If not, the second defect information indicates that the chestnut has a connectivity defect.

[0043] If so, then mark the second defect information as none.

[0044] In a second aspect, the present invention provides a control device for a chestnut screening equipment, characterized in that the control device for the chestnut screening equipment includes: a memory, a processor, and a chestnut screening equipment control program stored in the memory and executable on the processor, wherein when the chestnut screening equipment control program is executed by the processor, it implements the steps of the analysis method for non-destructive identification of internal defects in chestnuts as described above.

[0045] Thirdly, the present invention provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores a chestnut screening equipment control program, which, when executed by a processor, implements the steps of the analysis method for non-destructive identification of internal defects in chestnuts as described above.

[0046] The above-described one or more technical solutions provided by this invention can have the following advantages or at least achieve the following technical effects:

[0047] This invention proposes a non-destructive analysis method, device, and medium for identifying internal defects in chestnuts. By processing CT images of chestnuts multiple times, first and second defect information of the chestnuts are obtained respectively. Based on the first and second defect information, the internal defects of chestnuts are screened and classified, replacing manual judgment, saving more time and effort, and eliminating the problem of difficulty in forming a unified quality control due to large subjective differences of human beings. It provides an effective, reliable, and fast sorting and identification method for judging the internal defects of chestnuts, which can be widely used in actual production chains. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a schematic diagram of the overall process of the analysis method, apparatus and medium for non-destructive identification of internal defects in chestnuts according to the present invention.

[0050] Figure 2 A schematic diagram illustrating the conversion of computed tomography image data into digital data provided by the present invention;

[0051] Figure 3 This is a three-dimensional scatter plot of grayscale values ​​and porosity data of chestnuts of different qualities in the examples of this invention.

[0052] Figure 4 This is a schematic diagram of the reconstructed local calcified tissue in an example of the present invention;

[0053] Figure 5 This is a schematic diagram of the reconstructed low-grayscale abnormal tissue (abnormal state area) in an example of the present invention;

[0054] Figure 6 This is a schematic diagram of the pore network model in an example of the present invention;

[0055] Figure 7 This is a schematic flowchart of an embodiment of step S200 in the non-destructive identification method for internal defects in chestnuts according to the present invention.

[0056] Figure 8 This is a schematic flowchart of step S300 in the analysis method for non-destructive identification of internal defects in chestnuts according to the present invention.

[0057] label name label name 100 Mold 200 wormhole 300 cavity 400 Endothelial

[0058] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0060] It should be noted that in the embodiments of the present invention, all directional indications (such as up, down, left, right, front, back, etc.) are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0061] In this invention, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Without further limitation, an element defined by the phrase "comprising..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element. Furthermore, the meaning of "and / or" throughout the text includes three parallel options; for example, "A and / or B" includes option A, option B, or options where both A and B are satisfied.

[0062] In this invention, unless otherwise explicitly specified and limited, the terms "connection" and "fixed" should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral part; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal connection of two elements or the interaction relationship between two elements.

[0063] In this invention, if there are descriptions involving "first," "second," etc., such descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features.

[0064] In this invention, the use of suffixes such as "mechanism," "component," or "part" to denote elements is merely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "mechanism," "component," or "part" may be used interchangeably.

[0065] For those skilled in the art, the specific meanings of the above terms in this invention can be understood according to the specific circumstances. Furthermore, the technical solutions of the various embodiments can be combined with each other; however, this is based on the premise that those skilled in the art can implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0066] The inventive concept of the present invention will be further explained below with reference to some specific embodiments.

[0067] This invention proposes an analytical method, apparatus, and medium for non-destructive identification of internal defects in chestnuts.

[0068] Reference Figure 1 , Figure 2 and Figure 3 , Figure 1 This is a schematic diagram of the overall process of the analysis method, apparatus and medium for non-destructive identification of internal defects in chestnuts according to the present invention. Figure 2 A schematic diagram illustrating the conversion of computed tomography image data into digital data provided by the present invention; Figure 3 This is a three-dimensional scatter plot of the grayscale values ​​and porosity data of chestnuts of different qualities in an example of the present invention.

[0069] In one embodiment of the present invention, such as Figure 1 As shown, a non-destructive analysis method for identifying internal defects in chestnuts includes the following steps:

[0070] S100: Obtain the CT image of the chestnut and get the first layer;

[0071] S200: Process the first layer to obtain the second layer and the first defect information of the chestnut;

[0072] S300: Process the second layer to obtain the third layer and distinguish and extract the abnormal state area in the third layer according to the preset state value;

[0073] S400: Process and analyze the abnormal state area to obtain the second defect information of the chestnut;

[0074] S500: The chestnuts are screened and classified according to the first defect information and the second defect information.

[0075] For ease of understanding, a specific implementation method is shown below:

[0076] like Figure 2 As shown, a desktop X-ray micro-CT scanner is used to acquire X-ray CT images of chestnuts. The image resolution reaches the micrometer level and above. Chestnuts can be processed independently or in batches. The CT images of chestnuts are acquired by X-ray micro-computed tomography equipment. These image data are represented by the accumulation of images from each cross-sectional plane. The image resolution reaches the micrometer level. The gray value of each voxel is represented by a 16-bit digital signal. After scanning, the image reconstruction and display process can be omitted according to the equipment program. The digital signal can be analyzed directly to speed up the calculation time and save data storage space to obtain the first layer.

[0077] The first layer is processed by removing the image background portion of the first layer. The voxels of the sample area are defined as effective voxels, using the shell on each cross-sectional image as the boundary, to obtain the second layer and the first defect information of the chestnut.

[0078] Understandable Figure 2 This is a cross-sectional image of a chestnut (containing 900×900 voxels). The XY coordinate system represents the position of the voxels. The grayscale value of the voxels is displayed in 16-bit units (grayscale value range: -32768 to 32767). The larger the value, the higher the grayscale, indicating that the tissue at that location attenuates X-rays more. The table shows the conversion of image information to digital information for the grayscale values ​​of each voxel in a 7×7 region. For example, the grayscale value of the voxel in row 425 (X-axis) from bottom to top and column 609 (Y-axis) from left to right is 33454.

[0079] like Figure 3 As shown, the second layer is processed by performing digital data calculations on the CT images of several chestnuts without external damage. Scatter plots are drawn with the volume average gray value, volume gray standard deviation, and porosity data as X, Y, and Z axis coordinates, respectively. The experimental samples are divided into three categories: severely decayed, slightly decayed, and healthy. It is found that the severely decayed chestnuts have good distinguishability. The third layer is obtained and abnormal state areas in the third layer are distinguished and extracted according to preset state values.

[0080] The abnormal state area is processed and analyzed. Local grayscale analysis is performed on the remaining samples. The grayscale value of normal chestnut tissue is between 600 and 1200. Voxels with excessively high or low grayscale values ​​are signs of abnormal tissue. The second defect information of the chestnut is obtained.

[0081] The chestnuts are screened and classified according to the first defect information and the second defect information, and divided into normal chestnuts, chestnuts with the first defect information, chestnuts with the second permission information, and chestnuts with both the first defect information and the second defect information.

[0082] The technical solution of this invention processes chestnut CT images multiple times to obtain first and second defect information inside the chestnuts. Based on the first and second defect information, the internal defects of the chestnuts are screened and classified, replacing manual judgment. This method is more time-saving and labor-saving, and eliminates the problem of difficulty in forming a unified quality control due to large subjective differences in human factors. It provides an effective, reliable and fast sorting and identification method for judging the internal defects of chestnuts, which can be widely used in actual production chains.

[0083] Continue to refer to Figures 1 to 3 and refer to Figure 7 , Figure 7 This is a schematic flowchart of step S200 in the analysis method for non-destructive identification of internal defects in chestnuts according to the present invention.

[0084] like Figures 1 to 3 and Figure 7 As shown, in one embodiment of the present invention, in the above-described non-destructive identification method for internal defects in chestnuts, the first layer includes an image background layer and multiple continuous internal cross-sectional layers. The step of processing the first layer to obtain the second layer and the first defect information of the chestnut includes:

[0085] D100: Remove the background layer of the image;

[0086] D200: Divide the sample area using the outer shell of the chestnut in each of the internal cross-sectional layers as the boundary, and obtain effective voxels based on the voxels of the sample area to obtain the second layer;

[0087] D300: Obtain the volume average gray value, volume gray standard deviation, and porosity of the first layer based on the second layer;

[0088] D400: Obtain the first defect information based on the volume average gray value, volume gray standard deviation, and porosity of the first layer.

[0089] For ease of understanding, a specific implementation method is shown below:

[0090] First, the background portion of the image is removed. Then, using the outer shell on each cross-sectional image as the boundary, the voxels of the sample region are defined as effective voxels. The parameters are then calculated according to the formula:

[0091] Average gray value of the volume:

[0092] Standard deviation of volume gray level:

[0093] Porosity:

[0094] Where Gv is the gray value of the voxel; N is the number of effective voxels; Nl For voxels with a gray value greater than 0, the first defect information is obtained based on the average gray value, standard deviation of gray value, and porosity of the first layer.

[0095] In one embodiment, the above-described non-destructive analysis method for identifying internal defects in chestnuts includes the step of obtaining the volume average gray value, volume gray standard deviation, and porosity of the first layer based on the second layer, comprising:

[0096] F100: Based on the effective voxels of the sample area, obtain the first voxel gray value of the sample area, and based on the first voxel gray value and the number of effective voxels, obtain the volume average gray value of the first layer.

[0097] F200: Obtain the volume gray standard deviation based on the volume average gray value of the first layer and the number of effective voxels;

[0098] F300: Obtain the number of voxels in the sample region with a gray value greater than 0, and obtain the porosity based on the number of voxels in the sample region with a gray value greater than 0 and the number of effective voxels.

[0099] In one embodiment, the above-described non-destructive analysis method for identifying internal defects in chestnuts includes the step of obtaining the first defect information based on the volume average gray value, volume gray standard deviation, and porosity of the first layer, comprising:

[0100] W100: Determine whether the volume average gray value of the first layer is greater than or equal to the preset volume average gray value. If not, the first defect information is that the chestnut is completely rotten. That is, when the volume average gray value is less than 600, the volume gray standard deviation is greater than 240, or the porosity is greater than 20%, the chestnut is completely rotten. If yes, proceed to the next step.

[0101] W200: Determine whether the volume grayscale standard deviation is less than or equal to the preset grayscale standard deviation. If not, the first defect information is that the chestnut is completely rotten. If yes, proceed to the next step.

[0102] W300: Determine whether the porosity is less than or equal to the preset porosity. If not, the first defect information is that the chestnut is completely rotten. If yes, the first defect is marked as none.

[0103] Understandably, the identification protocol identifies severely defective chestnuts that are completely decayed by judging the grayscale parameters after image processing. Chestnuts with a volume average grayscale value of less than 600, a volume grayscale standard deviation of more than 240, or a porosity of more than 20% in CT images are defined as completely decayed chestnuts.

[0104] Continue to refer to Figures 1 to 3and refer to Figure 8 , Figure 8 This is a schematic flowchart of step S300 in the analysis method for non-destructive identification of internal defects in chestnuts according to the present invention.

[0105] like Figures 1 to 3 and Figure 8 As shown, in one embodiment of the present invention, the above-described non-destructive analysis method for identifying internal defects in chestnuts includes the following steps: processing the second layer to obtain a third layer and distinguishing and extracting abnormal state areas in the third layer according to preset state values:

[0106] E100: The second layer is segmented and processed according to a threshold of the CT image to obtain the third layer;

[0107] E200: Obtain the gray value of the second voxel in the third layer, compare the gray value of the second voxel with the first preset gray value, and obtain the first comparison result;

[0108] E300: Compare the gray value of the second voxel with the second preset gray value to obtain a second comparison result, and extract the abnormal state area in the third layer;

[0109] E400: Based on the first comparison result and the second comparison result, obtain the second defect information of the chestnut.

[0110] Specifically, when the gray value of the second voxel is less than 600, the region corresponding to the gray value of the second voxel being less than 600 in the third layer is extracted to obtain the abnormal state region; the abnormal state region is analyzed to obtain the second defect information of the chestnut.

[0111] Understandably, the first preset grayscale value is 1200, and the second preset grayscale value is 600.

[0112] Reference Figure 4 and Figure 5 , Figure 4 This is a schematic diagram of the reconstructed local calcified tissue in an example of the present invention. Figure 5 This is a schematic diagram of the reconstructed low-grayscale abnormal tissue (abnormal state area) in an example of the present invention.

[0113] In one embodiment of the present invention, in the above-described non-destructive analysis method for identifying internal defects in chestnuts, the preset grayscale value includes a first preset grayscale value and a second preset grayscale value, wherein the first preset grayscale value is greater than the second preset grayscale value, and the step of obtaining the second defect information of the chestnut based on the first comparison result and the second comparison result includes:

[0114] G100: such as Figure 4As shown, it is determined whether the gray value of the second voxel is less than or equal to the first preset gray value. If not, the second defect information is that the chestnut has a calcification defect. If so, the next step is executed.

[0115] G200: such as Figure 5 As shown, it is determined whether the gray value of the second voxel is greater than or equal to the second preset gray value. If not, the abnormal state area in the third layer is extracted based on the gray value of the second voxel, and the second defect information is obtained based on the abnormal state area.

[0116] G300: If so, mark the second defect information as none.

[0117] Specifically, abnormal grayscale areas are extracted using CT image thresholding and image processing techniques. For CT images of slightly defective chestnuts, thresholding is used to select voxels with grayscale values ​​below 600 or above 1200 as abnormal voxels, and these voxels are then reconstructed by computer. Chestnuts with calcification are identified by recognizing high-grayscale abnormal areas. Voxel regions with grayscale values ​​above 1200 are defined as high-grayscale abnormal areas; if such areas exist in the CT image of a chestnut, it is identified as a calcified chestnut.

[0118] In one embodiment, such as Figure 5 As shown, in the above-described non-destructive analysis method for identifying internal defects in chestnuts, the step of determining whether the grayscale value of the second voxel is greater than or equal to the second preset grayscale value, and if not, extracting the abnormal state area in the third layer based on the grayscale value of the second voxel, and obtaining the second defect information based on the abnormal state area, includes:

[0119] A100: Reconstruct the effective voxels whose second voxel gray value is less than the second preset gray value, filter out objects whose effective voxels are less than a preset unit, and obtain multiple isolated objects;

[0120] A200: Perform morphological analysis on each of the isolated objects to obtain the anisotropy and sphericity of each isolated object;

[0121] A300: Determine whether the anisotropy of the isolated object is less than the preset anisotropy. If yes, mark the second defect information as none. If no, proceed to the next step.

[0122] A400: Determine whether the sphericity of the isolated object is greater than or equal to the preset sphericity. If yes, the second defect information is that the chestnut has mold spots 100. If no, the second defect information is that the chestnut has wormholes 200.

[0123] Specifically, abnormal grayscale areas are extracted using CT image thresholding and image processing techniques. Voxels with extracted grayscale values ​​below 600 are reconstructed, and target objects with fewer than 1000 voxels (preset units) are filtered out. Morphological analysis is performed on each isolated defect. Target objects with anisotropy less than 0.85 (preset anisotropy) are classified as normal features (cavities 300 or endothelium 400); target objects with anisotropy greater than 0.85 and sphericity greater than 25 (preset sphericity) are classified as mold spots 100; and target objects with anisotropy greater than 0.85 and sphericity less than 25 are classified as wormholes 200.

[0124] Reference Figure 6 , Figure 6 This is a schematic diagram of the pore network model in an example of the present invention.

[0125] In one embodiment of the present invention, in the above-described non-destructive analysis method for identifying internal defects in chestnuts, the step of determining whether the gray value of the second voxel is greater than or equal to the second preset gray value, and if not, extracting the abnormal state area in the third layer based on the gray value of the second voxel, and obtaining the second defect information based on the abnormal state area includes:

[0126] Z100: such as Figure 6 As shown, the effective voxels whose gray values ​​are less than the second preset gray value are reconstructed to obtain a pore network model inside the chestnut with connected voxels.

[0127] Z200: Obtain the pore throat radius inside the chestnut based on the connecting voxel;

[0128] Z300: Determine whether the throat radius is greater than or equal to the throat threshold. If not, the second defect information is that the chestnut has a connectivity defect.

[0129] Z400: If so, mark the second defect information as none.

[0130] Specifically, an internal porosity network model is used to effectively identify connectivity defects in chestnuts. The porosity network model is used for skeleton analysis of the raw data, segmenting the distance image map, refining the network model, preserving connected voxels, and then converting the voxel skeleton into a spatial graphic object. The distance map of the segmented defect image is calculated, where the value of each voxel is equal to the distance to the nearest boundary voxel. The labeled image is then refined to preserve a string of connected voxels to extract the centerline of the filamentary structure from the defect image data. The distance to the nearest boundary is stored as a thickness attribute at each point in the spatial map. A threshold of 300 μm with an average pore throat radius is selected to filter for connected defect chestnuts.

[0131] For ease of understanding, a specific implementation method is shown below:

[0132] First, the background portion of the image is removed. Then, using the outer shell on each cross-sectional image as the boundary, the voxels of the sample region are defined as effective voxels. The parameters are then calculated according to the formula:

[0133] Average gray value of the volume:

[0134] Standard deviation of volume gray level:

[0135] Porosity:

[0136] Where Gv is the gray value of the voxel; N is the number of valid voxels; and N1 is the voxel with a gray value greater than 0.

[0137] The chestnuts were subjected to batch quality analysis, and the chestnut quality evaluation method included the following steps:

[0138] N100: Performs CT scans on chestnuts and preprocesses the images to acquire image data;

[0139] N200: Perform volumetric grayscale and porosity analysis to screen out chestnuts with internal anomalies from the perspective of the global image;

[0140] N300: Perform local high grayscale anomaly analysis to screen out chestnuts with excessively high grayscale voxels;

[0141] N400: Perform morphological analysis on the extracted isolated defects to distinguish abnormal features such as mold spots 100 and insect damage from abnormal features such as cavities 300 and endothelial 400.

[0142] N500: Perform pore network analysis on the remaining features to identify connectivity defects.

[0143] The CT image is a valid image after removing the background.

[0144] Digital data calculations were performed on several CT images of chestnuts without external damage. (See reference...) Figure 3 Scatter plots were created using the volume average gray value, volume gray standard deviation, and porosity data as X, Y, and Z axis coordinates, respectively, to classify the experimental samples into three categories: severely decayed, slightly decayed, and healthy. It was found that the severely decayed chestnuts had a good degree of differentiation.

[0145] Local grayscale analysis was performed on the remaining samples. The grayscale value of normal chestnut tissue is between 600 and 1200. Voxels with excessively high or low grayscale values ​​are signs of abnormal tissue and need to be analyzed separately.

[0146] Calcified tissue is hard and dense, therefore it is more easily attenuated by X-rays than healthy meat, exhibiting a higher grayscale value. (See also...) Figure 4Here, as shown in step N300, local analysis is performed on the image to screen out chestnuts with calcified tissue. The degree of calcification is proportional to the number of high-grayscale abnormal voxels.

[0147] On CT images, the grayscale values ​​of cavities 300, endothelial cells 400, mold spots 100, and wormholes 200 are lower than those of normal tissue. This is because infected fruit pulp gradually deteriorates, becoming loose and porous, leading to a decrease in tissue density. Cavities 300 and wormholes 200 are essentially air-filled internal parts of the chestnut: the lower the tissue density, the less X-ray attenuation. Step N400 establishes each low-grayscale voxel as an independent feature, allowing for the differentiation of internal chestnut defects through morphological analysis.

[0148] In this example, it was found that healthy chestnuts often have an inner skin with 400 markings and cavities. Therefore, the presence of inner skin with 400 markings and cavities cannot be used as a basis for judging internal defects in chestnuts. Since these four characteristics all belong to the low grayscale range, they cannot be directly distinguished by grayscale values, but they generally show great differences in morphology.

[0149] See Figure 5 Mold spots 100 and wormholes 200 are more concentrated and three-dimensional in space, while cavities 300 and inner skin 400 are flat and dispersed. By classifying each low grayscale feature through the parameter of anisotropy, chestnuts with mold spots 100 and wormholes 200 that have low anisotropy features are judged as rotten chestnuts, while those without this feature may be healthy chestnuts.

[0150] For the two abnormal features, mold spots 100 and wormholes 200, their sphericity is used to distinguish them by the difference in their specific surface area, so as to determine the main disease characteristics suffered by chestnuts during the post-harvest storage period.

[0151] For partially connected defects, which contain two or more composite features and cannot be identified by morphological parameters alone, this invention uses a pore network model for analysis.

[0152] Step N500 refers to setting the tube scale factor of the pore network model inside the entire chestnut to 0.1 and developing the pore throats. Thinner defects result in smaller pore throat radii, typically suitable for endothelial cells (400) and cavities (300). Conversely, thicker defects result in larger pore throat radii, suitable for mold and wormholes (200). Pore throats are the mesh-like structures connecting the various nodes. The average length and diameter of stomata are 549 and 49 μm, respectively. Please refer to [link to relevant documentation]. Figure 6The number of nodes within wormhole 200 is relatively small, and they are mostly located at the boundary between wormhole 200 and the inner skin layer. The pore throats are radially connected, with average lengths and diameters of 1216 and 605 μm, respectively. The more three-dimensional defects such as wormhole 200 and mold spots 100 there are, the larger the pore throat length and radius, and the fewer nodes in the pore network, and the fewer coordinating nodes. During comparison, the greatest difference in the average pore throat radius was observed between endothelial 400 and the connected defective tissue.

[0153] Test results showed that the chestnuts selected by the analysis program were healthy and plump, with no visible rotten tissue inside, a small amount of hollow space (300), and a uniform inner skin (400). Chestnuts detected by volume grayscale and porosity analysis were severely deteriorated and unusable. Chestnuts detected by local grayscale analysis showed internal calcification, mold, and insect damage consistent with the program's findings.

[0154] This embodiment uses image processing to extract parameters to analyze and identify internal defects in chestnuts. Specifically, chestnut quality can be graded according to the user's actual needs.

[0155] Furthermore, the present invention also provides a control device for a chestnut screening equipment, characterized in that the control device for the chestnut screening equipment includes: a memory, a processor, and a chestnut screening equipment control program stored in the memory and executable on the processor, wherein when the chestnut screening equipment control program is executed by the processor, it implements the steps of the analysis method for non-destructive identification of internal defects in chestnuts as described above.

[0156] The specific steps of the non-destructive identification method for internal defects in chestnuts are as described in the above embodiments. Since the control device of this chestnut screening equipment adopts all the technical solutions of all the above embodiments, it has at least all the beneficial effects brought about by the technical solutions of the above embodiments, which will not be repeated here.

[0157] In addition, the present invention also provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores a chestnut screening equipment control program, which, when executed by a processor, implements the steps of the analysis method for non-destructive identification of internal defects in chestnuts as described above.

[0158] The specific steps of the non-destructive identification method for internal defects in chestnuts are as described in the above embodiments. Since this computer-readable storage medium adopts all the technical solutions of all the above embodiments, it has at least all the beneficial effects brought about by the technical solutions of the above embodiments, and will not be repeated here.

[0159] It should be noted that the sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. The above embodiments are only optional embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made under the inventive concept of the present invention using the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are all included within the patent protection scope of the present invention.

Claims

1. A non-destructive analytical method for identifying internal defects in chestnuts, characterized in that, Includes the following steps: Obtain the CT image of the chestnut and get the first layer; Process the first layer to obtain the second layer and the first defect information of the chestnut; Process the second layer to obtain the third layer, and distinguish and extract the abnormal state areas in the third layer according to preset state values; The abnormal state area is processed and analyzed to obtain the second defect information of the chestnut. The chestnuts are screened and classified according to the first defect information and the second defect information; The first layer includes an image background layer and multiple consecutive internal cross-section layers. The step of processing the first layer to obtain the second layer and the first defect information of the chestnut includes: Remove the image background layer; The sample area is divided by the outer shell of the chestnut in each of the internal cross-sectional layers, and effective voxels are obtained based on the voxels of the sample area to obtain the second layer; The volume average gray value, volume gray standard deviation, and porosity of the first layer are obtained based on the second layer. The first defect information is obtained based on the volume average gray value, volume gray standard deviation, and porosity of the first layer. The steps of processing the second layer to obtain the third layer and distinguishing and extracting abnormal state areas in the third layer according to preset state values ​​include: The second layer is segmented and processed based on a threshold of the CT image to obtain the third layer; Obtain the grayscale value of the second voxel in the third layer, compare the grayscale value of the second voxel with the first preset grayscale value, and obtain the first comparison result; The second voxel gray value is compared with the second preset gray value to obtain a second comparison result, and the abnormal state area in the third layer is extracted. Based on the first comparison result and the second comparison result, the second defect information of the chestnut is obtained; The preset grayscale value includes a first preset grayscale value and a second preset grayscale value, wherein the first preset grayscale value is greater than the second preset grayscale value. The step of obtaining the second defect information of the chestnut based on the first comparison result and the second comparison result includes: Determine whether the gray value of the second voxel is less than or equal to the first preset gray value. If not, the second defect information is that the chestnut has a calcification defect. If yes, proceed to the next step. Determine whether the gray value of the second voxel is greater than or equal to the second preset gray value. If not, extract the abnormal state area in the third layer based on the gray value of the second voxel, and obtain the second defect information based on the abnormal state area. If so, then mark the second defect information as none.

2. The analytical method for non-destructive identification of internal defects in chestnuts as described in claim 1, characterized in that, The steps of obtaining the volume average gray value, volume gray standard deviation, and porosity of the first layer based on the second layer include: Based on the effective voxels of the sample region, the first voxel gray value of the sample region is obtained, and based on the first voxel gray value and the number of effective voxels, the volume average gray value of the first layer is obtained. The volume gray standard deviation is obtained based on the volume average gray value of the first layer and the number of effective voxels; The number of voxels in the sample region with a gray value greater than 0 is obtained, and the porosity is obtained based on the number of voxels in the sample region with a gray value greater than 0 and the number of effective voxels.

3. The analytical method for non-destructive identification of internal defects in chestnuts as described in claim 2, characterized in that, The step of obtaining the first defect information based on the volume average gray value, volume gray standard deviation, and porosity of the first layer includes: Determine whether the average gray value of the first layer is greater than or equal to the preset average gray value. If not, the first defect information is that the chestnut is completely rotten. If yes, proceed to the next step. Determine whether the standard deviation of the volume grayscale is less than or equal to the preset standard deviation of grayscale. If not, the first defect information is that the chestnut is completely rotten. If yes, proceed to the next step. Determine whether the porosity is less than or equal to a preset porosity. If not, the first defect information is that the chestnut is completely rotten. If yes, the first defect is marked as none.

4. The analytical method for non-destructive identification of internal defects in chestnuts as described in claim 1, characterized in that, The step of determining whether the grayscale value of the second voxel is greater than or equal to the second preset grayscale value, and if not, extracting the abnormal state area in the third layer based on the grayscale value of the second voxel, and obtaining the second defect information based on the abnormal state area includes: The effective voxels whose grayscale values ​​of the second voxel are less than the second preset grayscale values ​​are reconstructed, and objects whose effective voxels are less than a preset unit are filtered out to obtain multiple isolated objects. Morphological analysis is performed on each isolated object to obtain the anisotropy and sphericity of each isolated object; Determine whether the anisotropy of the isolated object is less than a preset anisotropy. If yes, mark the second defect information as none. If no, proceed to the next step. Determine whether the sphericity of the isolated object is greater than or equal to a preset sphericity. If yes, the second defect information is that the chestnut has mold spots; if no, the second defect information is that the chestnut has wormholes.

5. The analytical method for non-destructive identification of internal defects in chestnuts as described in claim 1, characterized in that, The step of determining whether the grayscale value of the second voxel is greater than or equal to the second preset grayscale value, and if not, extracting the abnormal state area in the third layer based on the grayscale value of the second voxel, and obtaining the second defect information based on the abnormal state area includes: The effective voxels whose gray values ​​are less than the second preset gray value are reconstructed to obtain a pore network model inside the chestnut with connected voxels. Based on the connecting voxel, the radius of the pore throat inside the chestnut is obtained; Determine whether the throat radius is greater than or equal to the throat threshold. If not, the second defect information indicates that the chestnut has a connectivity defect. If so, then mark the second defect information as none.

6. A control device for a chestnut screening equipment, characterized in that, The control device of the chestnut screening equipment includes: a memory, a processor, and a chestnut screening equipment control program stored in the memory and executable on the processor. When the chestnut screening equipment control program is executed by the processor, it implements the steps of the analysis method for non-destructive identification of internal defects in chestnuts as described in any one of claims 1 to 5.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a chestnut screening equipment control program, which, when executed by a processor, implements the steps of the analysis method for non-destructive identification of internal defects in chestnuts as described in any one of claims 1 to 5.

Images