Industrial digital X-ray imaging detection control system

Abstract

The invention relates to the technical field of X-ray imaging detection, and particularly discloses an industrial digital X-ray imaging detection control system, which obtains the size misjudgment degree of a pore in a pore detection image corresponding to a high scattering superposition layer in a plurality of historical scattering detection periods, and performs correlation analysis on the size misjudgment degree and the scattering superposition degree. Whether the scattering superposition influences the pore size or not is evaluated, whether the pore size detection error is related to the scattering superposition phenomenon or not can be determined, the source of the detection error can be accurately found out, the detection sequence can be reasonably arranged, the area with the high misjudged superposition value can be preferentially processed, and if the influence is caused, the scattering control coefficient is obtained. The digital X-ray imaging is regulated and controlled according to the scattering control coefficient, the energy of a ray source can be properly increased, and parameters such as the distance or the angle between the ray source and a workpiece can be adjusted, so that scattering is reduced.

Description

Technical Field

[0001] This invention relates to the field of X-ray imaging inspection technology, and more specifically to an industrial digital X-ray imaging inspection and control system. Background Technology

[0002] In industrial production, accurate detection of internal defects in multi-layered workpieces is crucial. Porosity, a common internal defect, plays a key role in ensuring workpiece quality and performance. Digital X-ray imaging technology, with its non-destructive and high-resolution advantages, has become a commonly used method for detecting porosity in multi-layered workpieces. However, in practical applications, this technology faces several problems affecting detection accuracy and efficiency.

[0003] In existing technologies, when using digital X-rays to detect pores within multi-layered workpieces, the complexity of the workpiece structure and the interaction between X-rays and matter cause varying degrees of scattering in digital X-ray images acquired during different imaging cycles. Currently, there is a lack of effective methods to identify which cycles are severely affected by scattering. This makes it difficult to optimize X-ray source parameters for these severely affected cycles, to identify the root cause of severe scattering, and consequently, to determine the optimal geometric parameter settings to minimize scattering. Summary of the Invention

[0004] The purpose of this invention is to provide an industrial digital X-ray imaging detection and control system to solve the problems mentioned above.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] An industrial digital X-ray imaging inspection and control system, comprising:

[0007] Scattering Period Screening Module: When using digital X-rays to detect pores in multi-layer workpieces, the scattering analysis of digital X-rays in multiple imaging detection periods is performed to screen out the scattering detection periods.

[0008] High-dispersion superposition extraction module: Performs inter-layer scattering superposition analysis on adjacent detection scattering periods within the scattering detection cycle, determines whether scattering superposition occurs between adjacent layers, and extracts adjacent layers with high dispersion superposition.

[0009] Misjudgment and superposition analysis module: Within multiple historical scattering detection cycles, the module obtains the degree of misjudgment of pore size in the pore detection image corresponding to the high scattering superposition layer, and performs correlation analysis with the degree of scattering superposition to evaluate whether the degree of scattering superposition affects the pore size.

[0010] Detection and control adjustment module: If it affects the size of the pores, it obtains the scattering control coefficient and adjusts the digital X-ray imaging according to the scattering control coefficient.

[0011] As a further aspect of the present invention: digital X-ray scattering analysis is performed within multiple imaging detection cycles, as follows:

[0012] One imaging detection cycle is selected from multiple imaging detection cycles as the target detection cycle, and it is equally divided into several target detection time periods. In each target detection time period, each pore in different workpiece layers is detected to obtain pore detection images in different workpiece layers as single-layer pore images.

[0013] The size detection value of each pore in the single-layer pore image and the actual size value of each pore are obtained respectively, and then input into the Euclidean distance formula to output the size difference value.

[0014] As a further aspect of the present invention, the screening process for the scattering detection cycle is as follows:

[0015] If the periodic screening value is greater than the periodic screening threshold, the target detection period is marked as the scattering detection period.

[0016] As a further aspect of the present invention: interlayer scattering superposition analysis is performed on adjacent detection scattering periods within the scattering detection period to obtain the fitted ellipse of the previous target and the fitted ellipse of the subsequent target, as follows:

[0017] Within the scattering detection period, adjacent detection scattering time periods are combined to obtain multiple detection scattering combinations. Within each detection scattering combination, the single-layer pore images corresponding to the previous and subsequent detection scattering time periods are extracted as the previous and subsequent single-layer pore images. The same pore in the previous and subsequent single-layer pore images is extracted as the target pore.

[0018] The contours of the target stomata in the front and back single-layer stomata images are extracted respectively, and ellipse fitting is performed based on the contours of the target stomata to obtain the front target fitted ellipse and the back target fitted ellipse.

[0019] As a further aspect of the present invention: determining whether scattering superposition occurs between adjacent layers, the process is as follows:

[0020] Obtain the lengths of the major and minor axes of the fitted ellipse for the front and rear targets respectively, and calculate the ratios to obtain the ratios of the major and minor axes of the front and rear targets.

[0021] If the aspect ratio of the front target and the aspect ratio of the rear target are both greater than the target aspect ratio threshold, they will be displayed as bar-shaped signals of the front and rear targets.

[0022] Extract columnar pore images corresponding to the target pores in the front and rear single-layer pore images respectively, and obtain the angle between the long axis in the columnar pore image and the Z-axis in the single-layer workpiece 3D model respectively, as the front target scattering angle and the rear target scattering angle.

[0023] According to the method of obtaining the scattering angle of the front target and the scattering angle of the back target, the scattering angle of the remaining pores in the front single-layer pore image and the back single-layer pore image are obtained respectively, and the front scattering angle sequence and the back scattering angle sequence are constructed according to the order of obtaining the scattering angle.

[0024] Extract the maximum and minimum front target scattering angles within the forward scattering angle sequence, and the maximum and minimum front target scattering angles within the backward scattering angle sequence, to construct the front target scattering angle set and the backward target scattering angle set;

[0025] If there is an intersection between the set of scattering angles of the front target and the set of scattering angles of the rear target, the angle overlap range corresponding to the intersection is extracted, and the ratio is calculated with the range corresponding to the set of scattering angles of the front target. The overlap angle ratio is output. If the overlap angle ratio is greater than or equal to the overlap angle ratio threshold, it is displayed as a scattering superposition signal.

[0026] As a further aspect of the present invention, the extraction process of adjacent layers with high dispersion is as follows:

[0027] The overlap angle ratios corresponding to each detection and dispersion combination are compared, and the pre-detection scattering time period and post-detection scattering time period corresponding to the detection and dispersion combination with the largest overlap angle ratio are selected and mapped to obtain the high dispersion superposition adjacent layers.

[0028] As a further aspect of the present invention, the process for obtaining the degree of scattering superposition is as follows:

[0029] Extract the single-layer stomatal image corresponding to the highly scattering superposition layer, obtain the size detection value of each stomatal in the single-layer stomatal image, and sort them according to the time series of the obtained size detection values ​​to construct the detection size sequence;

[0030] Obtain the actual size value of each pore in a single-layer pore image, and construct an actual size sequence according to the construction method of the size detection sequence;

[0031] Obtain any one pore in the single-layer pore image as the comparison pore, and extract the size detection value of the comparison pore in the detection size sequence and the actual size value in the actual size sequence respectively;

[0032] According to the method of obtaining the corresponding size detection value and actual size value of the pore, each pore in the single-layer pore image is obtained separately and combined to obtain multiple size comparison groups. Within the size comparison group, the difference between the corresponding size detection value and the actual size value of the pore is calculated, the absolute value is taken, and the ratio with the actual size value is calculated to obtain the size misjudgment degree value.

[0033] Extract the overlap angle ratio corresponding to the pores, and subtract it from the overlap angle ratio threshold. Take the absolute value to obtain the scattering superposition degree value.

[0034] As a further aspect of the present invention, the correlation analysis process is as follows:

[0035] By acquiring the size misjudgment value and scattering superposition value of each pore in a single-layer pore image, respectively, the size misjudgment value and scattering superposition value of each pore are obtained by comparing the size misjudgment value and the scattering superposition value of the corresponding pore.

[0036] The mean and standard deviation of the size misjudgment value of each pore in the single-layer pore image are calculated separately to obtain the mean and standard deviation of size misjudgment.

[0037] The scattering superposition degree value of each pore in the single-layer pore image is averaged and the standard deviation is calculated to obtain the scattering superposition mean and scattering superposition standard deviation.

[0038] The mean value of size misjudgment, the standard deviation of size misjudgment, the mean value of scattering superposition, and the standard deviation of scattering superposition are respectively input into the Pearson correlation coefficient formula to obtain the misjudgment superposition value;

[0039] If the misjudged superposition value is greater than the misjudged superposition threshold, it will be displayed as a misjudged superposition associated signal.

[0040] As a further aspect of the present invention, the process for obtaining the scattering control coefficient is as follows:

[0041] Within the size comparison group, the ratio of the size misjudgment degree value to the scattering superposition degree value is calculated, and the degree coefficient is output.

[0042] The scattering control coefficients are calculated by averaging the degree coefficients corresponding to each size comparison group.

[0043] As a further aspect of the present invention, the process of adjusting digital X-ray imaging based on the scattering control coefficient is as follows:

[0044] The scattering control coefficient is obtained by averaging the degree coefficients corresponding to each size comparison group and then outputting them.

[0045] Obtain the overlap angle ratio of adjacent layers with high dispersion, and multiply the overlap angle ratio with the degree coefficient to calculate the scattering control quantity.

[0047] This invention utilizes digital X-rays to detect pores within multi-layered workpieces. It performs scattering analysis on digital X-rays across multiple imaging detection cycles, identifying scattering detection cycles. This process clarifies which cycles are severely affected by scattering, aiding in setting the X-ray source parameters for these cycles, identifying the causes of severe scattering, and finding optimal geometric parameter settings to minimize scattering, thereby improving detection accuracy and efficiency. Furthermore, it performs interlayer scattering superposition analysis on adjacent scattering periods within the scattering detection cycle to determine whether scattering superposition occurs between adjacent layers. Layers with high scattering superposition are identified as high-scattering superposition adjacent layers, clarifying which adjacent layers have severe scattering superposition. Subsequent specialized scattering correction processing can be performed on these high-scattering superposition adjacent layers, effectively reducing scattering interference with the image. Moreover, once the high-scattering superposition adjacent layers are identified, only these layers need to be re-detected or corrected, demonstrating a targeted approach and improving the efficiency of industrial digital X-ray imaging detection.

[0048] This invention acquires the degree of misjudgment of pore size in pore detection images corresponding to highly scattered superposition layers within multiple historical scattering detection cycles, and performs correlation analysis with the degree of scattering superposition to assess whether scattering superposition affects pore size. It can determine whether the error in pore size detection is related to the scattering superposition phenomenon, which helps to accurately find the root cause of detection error. Furthermore, it can rationally arrange the detection sequence, prioritizing the processing of areas with high misjudgment superposition values. If it has an impact, it acquires the scattering control coefficient, and adjusts the digital X-ray imaging based on the scattering control coefficient. It can appropriately increase the energy of the X-ray source, adjust parameters such as the distance or angle between the X-ray source and the workpiece, to reduce scattering, thereby improving image quality and making the edges of defects such as pores clearer, facilitating accurate identification and measurement. Attached Figure Description

[0049] The invention will now be further described with reference to the accompanying drawings.

[0050] Figure 1 This is a schematic diagram of the module structure of an industrial digital X-ray imaging detection and control system according to the present invention. Detailed Implementation

[0052] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] Please see Figure 1As shown in the figure, an industrial digital X-ray imaging detection and control system according to an embodiment of the present invention includes the following steps: Scattering Period Screening Module: When using digital X-rays to detect pores in multi-layer workpieces, the scattering analysis of digital X-rays in multiple imaging detection periods is performed to screen out the scattering detection periods. In some embodiments, one imaging detection period is selected from multiple imaging detection periods as the target detection period; The target detection cycle is divided into several target detection time periods, each with an equal duration. It should be noted that only the pores in one layer of workpiece are imaged and detected during each target detection period. Therefore, the number of target detection periods is equal to the total number of layers of multi-layer workpieces. During each target detection period, each pore in different workpiece layers is detected to obtain pore detection images in different workpiece layers, which are used as single-layer pore images. The size detection value of each pore in the single-layer pore image and the actual size value of each pore are obtained respectively; Input the measured size value of each pore and its corresponding actual size value into the Euclidean distance formula in sequence, and output the size difference value. If the size difference value is less than or equal to the size difference threshold, it indicates that the detected pore size deviates little from the actual pore size, and the result is a normal detection signal. If the size difference value is greater than the size difference threshold, it indicates that the detected pore size deviates significantly from the actual pore size, and is displayed as a detection scattering signal. The target detection period that is displayed as the detection scattering signal will be marked as the detection scattering period. The ratio of the number of detection scattering periods within the target detection cycle to the total number of target detection periods is used to obtain the detection scattering number ratio. The difference between the size difference value corresponding to each detection scattering time period and the size difference threshold is calculated, and the ratio is calculated to the size difference threshold. Then, the average value is calculated and the detection scattering degree value is output. The periodic screening value is obtained by multiplying the ratio of detected scattering quantities with the value of detected scattering intensity. If the periodic screening value is greater than the periodic screening threshold, it indicates that there are a large number of detection scattering periods within the target detection period and the degree of detection scattering is large. The target detection period is then marked as the scattering detection period. If the periodic screening value is less than or equal to the periodic screening threshold, it indicates that there are few detection scattering periods within the target detection period and the degree of detection scattering is small, and it is marked as a normal detection period. The purpose of screening out the scattering detection period is that scattering can cause background noise in digital X-ray images, reduce the contrast and clarity of the images, and cause the edges of defects such as pores to become blurred, making them difficult to accurately identify and measure. By screening out the scattering detection period, it is possible to identify which periods of the image are severely affected by scattering, and then special scattering correction processing can be performed on the image. Secondly, since the scattering situation varies in different detection cycles, screening out the scattering detection cycles helps to set the X-ray source parameters for these cycles, find the cause of severe scattering, find the optimal geometric parameter settings, minimize scattering, and improve the accuracy and efficiency of detection. Furthermore, if the scattering detection period is not screened out, images severely affected by scattering may require repeated detections due to the inability to accurately determine the porosity, wasting a lot of time and resources. By screening out the scattering detection period, it is possible to identify which period's detection results are unreliable, and only these specific periods need to be re-detected or corrected, instead of repeating the entire detection process. High-scatter superposition extraction module: Performs interlayer scatter superposition analysis on adjacent detection scattering periods within the scattering detection period, determines whether scattering superposition occurs between adjacent layers, and extracts adjacent layers with high scattering superposition as high-scatter superposition adjacent layers; In some embodiments, adjacent scattering detection periods are combined within a scattering detection period to obtain multiple scattering detection combinations. Among them, adjacent detection scattering periods within a single group of detection and scattering combinations include the preceding detection scattering period and the following detection scattering period; Within the detection and scattering combination, the single-layer pore images corresponding to the pre-detection scattering period and the post-detection scattering period are extracted respectively, and used as the pre-single-layer pore image and the post-single-layer pore image. The same stoma is extracted from the front and back monolayer stoma images respectively and used as the target stoma. It should be noted that the target pores are: because the difference between the front single-layer pore image and the back single-layer pore image is that they are obtained by imaging detection in different workpiece layers. For example, there are five pores in the front single-layer pore image, including pore A1, pore B1, pore C1, pore D1 and pore E1, and there are five pores in the back single-layer pore image, including pore A2, pore B2, pore C2, pore D2 and pore E2. Therefore, the same stoma can be stoma A1 and stoma A2. Stoma A1 and stoma A2 are just not on the same horizontal spatial dimension, but on the same vertical spatial dimension. Ellipse fitting was performed on the target stomata in the front and back monolayer stomatal images, respectively, as follows: For example, the outline of the target stomata in the previous single-layer stomata image is extracted as the previous target stomata outline. Based on the previous target stomata outline, ellipse fitting is performed to obtain the previous target fitted ellipse. Similarly, the contour of the target stomata in the subsequent single-layer stomata image is extracted as the subsequent target stomata contour. Based on the subsequent target stomata contour, ellipse fitting is performed to obtain the subsequent target fitted ellipse. The lengths of the major and minor axes within the fitted ellipse of the target are obtained separately, and their ratios are calculated to obtain the ratio of the major and minor axes of the target. The lengths of the major and minor axes within the fitted ellipse of the target are obtained separately, and their ratios are calculated to obtain the ratio of the major to minor axes of the target. If at least one of the aspect ratios of the front target and the rear target is less than or equal to the target aspect ratio threshold, it means that at least one of the target stomata in the front single-layer stomata image and the rear stomata image is not columnar. If the aspect ratio of the front target and the aspect ratio of the rear target are both greater than the target aspect ratio threshold, it indicates that the outline of the target stomata in both the front single-layer stomatal image and the rear stomatal image is columnar, and is displayed as columnar signals of the front and rear targets. Based on the front and rear target columnar signals, each layer of the multi-layer workpiece is transformed into three-dimensional space, with the X-axis representing length, the Y-axis representing width, and the Z-axis representing height of each layer, to construct a three-dimensional model of a single-layer workpiece. Extract the columnar pore image corresponding to the target pore in the previous single-layer pore image, and obtain the angle between the long axis in the columnar pore image and the Z-axis in the single-layer workpiece 3D model as the front target scattering angle. Extract the columnar pore image corresponding to the target pore in the single-layer pore image, and obtain the angle between the long axis in the columnar pore image and the Z-axis in the single-layer workpiece 3D model as the back target scattering angle. According to the method of obtaining the scattering angle of the front target and the scattering angle of the back target, the scattering angle of the remaining pores in the front single-layer pore image and the back single-layer pore image are obtained respectively, and the front scattering angle sequence and the back scattering angle sequence are constructed according to the order of obtaining the scattering angle. Extract the maximum and minimum forward scattering angles within the forward scattering angle sequence to construct a set of forward scattering angles. Extract the maximum and minimum front target scattering angles within the backscattering angle sequence to construct a set of back target scattering angles. If the set of scattering angles of the front target and the set of scattering angles of the rear target do not intersect, no operation will be performed; If there is an intersection between the set of scattering angles of the front target and the set of scattering angles of the rear target, the angle overlap range corresponding to the intersection is extracted, and the ratio is calculated with the range corresponding to the set of scattering angles of the front target, and the overlap angle ratio is output. If the overlap angle is less than the overlap angle percentage threshold, it indicates that there is a small degree of scattering superposition between pores in the preceding and following single-layer pore images, and it is displayed as a non-scattering superposition signal. If the overlap angle ratio is greater than or equal to the overlap angle ratio threshold, it indicates that there is a large degree of scattering superposition between the pores in the preceding and following single-layer pore images, which is displayed as a scattering superposition signal. The overlap angle ratios of each detection and dispersion combination are compared, and the pre-detection scattering time period and post-detection scattering time period corresponding to the detection and dispersion combination with the largest overlap angle ratio are selected and mapped to obtain the high dispersion superposition adjacent layers. It should be noted that the significance of extracting adjacent layers with high dispersion is that since the pore detection images corresponding to adjacent layers with high dispersion are all images that are severely affected by scattering, it will lead to the inability to accurately determine the pore condition and to perform repeated detections, wasting a lot of time and resources. However, after identifying the adjacent layers with high dispersion, it is only necessary to re-detect or correct the adjacent layers with high dispersion, which is more targeted and improves the efficiency of using industrial digital X-ray imaging detection. By extracting adjacent layers with high dispersion, we can identify which adjacent layers have severe scattering superposition. Subsequently, we can perform special scattering correction processing on adjacent layers with high dispersion, which can effectively reduce the interference of scattering on the image, improve image quality, and make the edges of defects such as pores clearer, making it easier to accurately identify and measure them. The specific solution in this embodiment is as follows: When using digital X-rays to detect pores in multi-layered workpieces, scattering analysis is performed on digital X-rays within multiple imaging detection cycles to screen out scattering detection cycles. This identifies which cycles' images are severely affected by scattering, which helps in setting the X-ray source parameters for these cycles, identifying the causes of severe scattering, finding the optimal geometric parameter settings to minimize scattering, and improving detection accuracy and efficiency. Interlayer scattering superposition analysis is performed on adjacent detection scattering periods within the scattering detection cycle to determine whether scattering superposition occurs between adjacent layers. Adjacent layers with high scattering superposition are extracted as high-scattering superposition adjacent layers, identifying which adjacent layers have severe scattering superposition. Subsequently, specialized scattering correction processing can be performed on high-scattering superposition adjacent layers, effectively reducing scattering interference with the image. Furthermore, after identifying high-scattering superposition adjacent layers, only re-detection or correction processing needs to be performed on these high-scattering superposition adjacent layers, making it targeted and improving the efficiency of industrial digital X-ray imaging detection.

[0055] Scattering Period Screening Module: When using digital X-rays to detect pores in multi-layer workpieces, the scattering analysis of digital X-rays in multiple imaging detection periods is performed to screen out the scattering detection periods.

[0056] In some embodiments, one imaging detection period is selected from multiple imaging detection periods as the target detection period;

[0057] The target detection cycle is divided into several target detection time periods, each with an equal duration.

[0058] It should be noted that only the pores in one layer of workpiece are imaged and detected during each target detection period. Therefore, the number of target detection periods is equal to the total number of layers of multi-layer workpieces.

[0059] During each target detection period, each pore in different workpiece layers is detected to obtain pore detection images in different workpiece layers, which are used as single-layer pore images.

[0060] The size detection value of each pore in the single-layer pore image and the actual size value of each pore are obtained respectively;

[0061] Input the measured size value of each pore and its corresponding actual size value into the Euclidean distance formula in sequence, and output the size difference value.

[0062] If the size difference value is less than or equal to the size difference threshold, it indicates that the detected pore size deviates little from the actual pore size, and the result is a normal detection signal.

[0063] If the size difference value is greater than the size difference threshold, it indicates that the detected pore size deviates significantly from the actual pore size, and is displayed as a detection scattering signal. The target detection period that is displayed as the detection scattering signal will be marked as the detection scattering period.

[0064] The ratio of the number of detection scattering periods within the target detection cycle to the total number of target detection periods is used to obtain the detection scattering number ratio.

[0065] The difference between the size difference value corresponding to each detection scattering time period and the size difference threshold is calculated, and the ratio is calculated to the size difference threshold. Then, the average value is calculated and the detection scattering degree value is output.

[0066] The periodic screening value is obtained by multiplying the ratio of detected scattering quantities with the value of detected scattering intensity.

[0067] If the periodic screening value is greater than the periodic screening threshold, it indicates that there are a large number of detection scattering periods within the target detection period and the degree of detection scattering is large. The target detection period is then marked as the scattering detection period.

[0068] If the periodic screening value is less than or equal to the periodic screening threshold, it indicates that there are few detection scattering periods within the target detection period and the degree of detection scattering is small, and it is marked as a normal detection period.

[0069] The purpose of screening out the scattering detection period is that scattering can cause background noise in digital X-ray images, reduce the contrast and clarity of the images, and cause the edges of defects such as pores to become blurred, making them difficult to accurately identify and measure. By screening out the scattering detection period, it is possible to identify which periods of the image are severely affected by scattering, and then special scattering correction processing can be performed on the image.

[0070] Secondly, since the scattering situation varies in different detection cycles, screening out the scattering detection cycles helps to set the X-ray source parameters for these cycles, find the cause of severe scattering, find the optimal geometric parameter settings, minimize scattering, and improve the accuracy and efficiency of detection.

[0071] Furthermore, if the scattering detection period is not screened out, images severely affected by scattering may require repeated detections due to the inability to accurately determine the porosity, wasting a lot of time and resources. By screening out the scattering detection period, it is possible to identify which period's detection results are unreliable, and only these specific periods need to be re-detected or corrected, instead of repeating the entire detection process.

[0072] High-scatter superposition extraction module: Performs interlayer scatter superposition analysis on adjacent detection scattering periods within the scattering detection period, determines whether scattering superposition occurs between adjacent layers, and extracts adjacent layers with high scattering superposition as high-scatter superposition adjacent layers;

[0073] In some embodiments, adjacent scattering detection periods are combined within a scattering detection period to obtain multiple scattering detection combinations.

[0074] Among them, adjacent detection scattering periods within a single group of detection and scattering combinations include the preceding detection scattering period and the following detection scattering period;

[0075] Within the detection and scattering combination, the single-layer pore images corresponding to the pre-detection scattering period and the post-detection scattering period are extracted respectively, and used as the pre-single-layer pore image and the post-single-layer pore image.

[0076] The same stoma is extracted from the front and back monolayer stoma images respectively and used as the target stoma.

[0077] It should be noted that the target pores are: because the difference between the front single-layer pore image and the back single-layer pore image is that they are obtained by imaging detection in different workpiece layers. For example, there are five pores in the front single-layer pore image, including pore A1, pore B1, pore C1, pore D1 and pore E1, and there are five pores in the back single-layer pore image, including pore A2, pore B2, pore C2, pore D2 and pore E2.

[0078] Therefore, the same stoma can be stoma A1 and stoma A2. Stoma A1 and stoma A2 are just not on the same horizontal spatial dimension, but on the same vertical spatial dimension.

[0079] Ellipse fitting was performed on the target stomata in the front and back monolayer stomatal images, respectively, as follows:

[0080] For example, the outline of the target stomata in the previous single-layer stomata image is extracted as the previous target stomata outline. Based on the previous target stomata outline, ellipse fitting is performed to obtain the previous target fitted ellipse.

[0081] Similarly, the contour of the target stomata in the subsequent single-layer stomata image is extracted as the subsequent target stomata contour. Based on the subsequent target stomata contour, ellipse fitting is performed to obtain the subsequent target fitted ellipse.

[0082] The lengths of the major and minor axes within the fitted ellipse of the target are obtained separately, and their ratios are calculated to obtain the ratio of the major and minor axes of the target.

[0083] The lengths of the major and minor axes within the fitted ellipse of the target are obtained separately, and their ratios are calculated to obtain the ratio of the major to minor axes of the target.

[0084] If at least one of the aspect ratios of the front target and the rear target is less than or equal to the target aspect ratio threshold, it means that at least one of the target stomata in the front single-layer stomata image and the rear stomata image is not columnar.

[0085] If the aspect ratio of the front target and the aspect ratio of the rear target are both greater than the target aspect ratio threshold, it indicates that the outline of the target stomata in both the front single-layer stomatal image and the rear stomatal image is columnar, and is displayed as columnar signals of the front and rear targets.

[0086] Based on the front and rear target columnar signals, each layer of the multi-layer workpiece is transformed into three-dimensional space, with the X-axis representing length, the Y-axis representing width, and the Z-axis representing height of each layer, to construct a three-dimensional model of a single-layer workpiece.

[0087] Extract the columnar pore image corresponding to the target pore in the previous single-layer pore image, and obtain the angle between the long axis in the columnar pore image and the Z-axis in the single-layer workpiece 3D model as the front target scattering angle.

[0088] Extract the columnar pore image corresponding to the target pore in the single-layer pore image, and obtain the angle between the long axis in the columnar pore image and the Z-axis in the single-layer workpiece 3D model as the back target scattering angle.

[0089] According to the method of obtaining the scattering angle of the front target and the scattering angle of the back target, the scattering angle of the remaining pores in the front single-layer pore image and the back single-layer pore image are obtained respectively, and the front scattering angle sequence and the back scattering angle sequence are constructed according to the order of obtaining the scattering angle.

[0090] Extract the maximum and minimum forward scattering angles within the forward scattering angle sequence to construct a set of forward scattering angles.

[0091] Extract the maximum and minimum front target scattering angles within the backscattering angle sequence to construct a set of back target scattering angles.

[0092] If the set of scattering angles of the front target and the set of scattering angles of the rear target do not intersect, no operation will be performed;

[0093] If there is an intersection between the set of scattering angles of the front target and the set of scattering angles of the rear target, the angle overlap range corresponding to the intersection is extracted, and the ratio is calculated with the range corresponding to the set of scattering angles of the front target, and the overlap angle ratio is output.

[0094] If the overlap angle is less than the overlap angle percentage threshold, it indicates that there is a small degree of scattering superposition between pores in the preceding and following single-layer pore images, and it is displayed as a non-scattering superposition signal.

[0095] If the overlap angle ratio is greater than or equal to the overlap angle ratio threshold, it indicates that there is a large degree of scattering superposition between the pores in the preceding and following single-layer pore images, which is displayed as a scattering superposition signal.

[0096] The overlap angle ratios of each detection and dispersion combination are compared, and the pre-detection scattering time period and post-detection scattering time period corresponding to the detection and dispersion combination with the largest overlap angle ratio are selected and mapped to obtain the high dispersion superposition adjacent layers.

[0097] It should be noted that the significance of extracting adjacent layers with high dispersion is that since the pore detection images corresponding to adjacent layers with high dispersion are all images that are severely affected by scattering, it will lead to the inability to accurately determine the pore condition and to perform repeated detections, wasting a lot of time and resources. However, after identifying the adjacent layers with high dispersion, it is only necessary to re-detect or correct the adjacent layers with high dispersion, which is more targeted and improves the efficiency of using industrial digital X-ray imaging detection.

[0098] By extracting adjacent layers with high dispersion, we can identify which adjacent layers have severe scattering superposition. Subsequently, we can perform special scattering correction processing on adjacent layers with high dispersion, which can effectively reduce the interference of scattering on the image, improve image quality, and make the edges of defects such as pores clearer, making it easier to accurately identify and measure them.

[0099] The specific solution in this embodiment is as follows: When using digital X-rays to detect pores in multi-layered workpieces, scattering analysis is performed on digital X-rays within multiple imaging detection cycles to screen out scattering detection cycles. This identifies which cycles' images are severely affected by scattering, which helps in setting the X-ray source parameters for these cycles, identifying the causes of severe scattering, finding the optimal geometric parameter settings to minimize scattering, and improving detection accuracy and efficiency. Interlayer scattering superposition analysis is performed on adjacent detection scattering periods within the scattering detection cycle to determine whether scattering superposition occurs between adjacent layers. Adjacent layers with high scattering superposition are extracted as high-scattering superposition adjacent layers, identifying which adjacent layers have severe scattering superposition. Subsequently, specialized scattering correction processing can be performed on high-scattering superposition adjacent layers, effectively reducing scattering interference with the image. Furthermore, after identifying high-scattering superposition adjacent layers, only re-detection or correction processing needs to be performed on these high-scattering superposition adjacent layers, making it targeted and improving the efficiency of industrial digital X-ray imaging detection.

[0100] Please see Figure 1 As shown in the figure, an industrial digital X-ray imaging detection and control system according to an embodiment of the present invention includes the following steps: Misjudgment and superposition analysis module: Within multiple historical scattering detection cycles, the module obtains the degree of misjudgment of pore size in the pore detection image corresponding to the high scattering superposition layer, and performs correlation analysis with the degree of scattering superposition to evaluate whether the degree of scattering superposition affects the pore size. In some embodiments, the image of a single pore layer corresponding to the highly scattering superposition layer is extracted, and the size detection value of each pore in the single pore layer image is obtained. The size detection values ​​are then sorted according to the time sequence of the obtained size detection values ​​to construct a detection size sequence. Obtain the actual size value of each pore in a single-layer pore image, and construct an actual size sequence according to the construction method of the size detection sequence; Obtain any one pore in the single-layer pore image as the comparison pore, and extract the size detection value of the comparison pore in the detection size sequence and the actual size value in the actual size sequence respectively; According to the method of obtaining the corresponding size detection value and actual size value of the pores, each pore in the single-layer pore image is acquired separately and combined to obtain multiple size comparison groups; Within the size comparison group, the difference between the detected size value and the actual size value of the air hole is calculated, the absolute value is taken, and the ratio with the actual size value is calculated to obtain the size misjudgment degree value; Extract the overlap angle ratio corresponding to the pores and compare it with the overlap angle ratio threshold. Take the absolute value to obtain the scattering superposition degree value. By acquiring the size misjudgment value and scattering superposition value of each pore in a single-layer pore image, respectively, the size misjudgment value and scattering superposition value of each pore are obtained by comparing the size misjudgment value and the scattering superposition value of the corresponding pore. The average value of size misjudgment is calculated by averaging the size misjudgment of each pore in the single-layer pore image. The standard deviation of the size misjudgment degree value of each pore in the single-layer pore image is calculated to obtain the size misjudgment standard deviation; The scattering superposition value of each pore in the single-layer pore image is averaged to obtain the average scattering superposition value; The standard deviation of the scattering superposition value of each pore in the single-layer pore image is calculated to obtain the scattering superposition standard deviation; The mean value of size misjudgment, the standard deviation of size misjudgment, the mean value of scattering superposition, and the standard deviation of scattering superposition are respectively input into the Pearson correlation coefficient formula to obtain the misjudgment superposition value; If the misjudged superposition value is greater than the misjudged superposition threshold, it indicates that the signal is a misjudged superposition associated signal. If the misjudged superposition value is less than or equal to the misjudged superposition threshold, it indicates that the signal is displayed as a misjudged superposition of non-associated signals. It should be noted that the significance of analyzing the correlation between size misjudgment and scattering superposition lies in the following: When using digital X-ray imaging to detect pore size, size misjudgment occurs. Furthermore, when evaluating X-ray scattering during digital X-ray imaging detection, scattering of the detected X-rays is observed. Therefore, from the perspective of imaging detection accuracy, analyzing the correlation between size misjudgment and scattering superposition can determine whether the error in pore size detection is related to scattering superposition. If a misjudgment superposition correlation signal is displayed, it indicates that the degree of scattering superposition has a significant impact on pore size detection, helping to accurately identify the root cause of the detection error. From the perspective of imaging detection efficiency, it allows for the rational arrangement of the detection sequence, prioritizing areas with high misjudgment superposition values, and the rational allocation of human and equipment resources. More effort and advanced equipment can be invested in the detection stages with serious problems, thereby improving the overall efficiency of the detection work. Detection and control adjustment module: If it affects the size of the pores, it obtains the scattering control coefficient and adjusts the digital X-ray imaging according to the scattering control coefficient; In some embodiments, within a size comparison group, the ratio of the size misjudgment degree value to the scattering superposition degree value is calculated, and the degree coefficient is output. The scattering control coefficient is obtained by averaging the degree coefficients corresponding to each size comparison group and then outputting them. Obtain the overlap angle ratio of adjacent layers with high dispersion, and calculate the scattering control quantity by multiplying the overlap angle ratio with the degree coefficient. It should be noted that the significance of obtaining the scattering control value is that it reflects the degree of influence of scattering on the detection and enables targeted adjustment of the X-ray source parameters of digital X-ray imaging. When the scattering control value is large, it indicates that scattering seriously interferes with the detection results. At this time, the energy of the X-ray source can be appropriately increased, and parameters such as the distance or angle between the X-ray source and the workpiece can be adjusted to reduce the generation of scattering, thereby improving the image quality and making the edges of defects such as pores clearer, which is convenient for accurate identification and measurement. The specific solution in this embodiment is as follows: Within multiple historical scattering detection cycles, the degree of misjudgment of pore size in the pore detection image corresponding to the highly scattering superposition layer is obtained, and a correlation analysis is performed with the degree of scattering superposition to assess whether scattering superposition affects pore size. This can determine whether the error in pore size detection is related to the scattering superposition phenomenon, which helps to accurately find the root cause of the detection error and can reasonably arrange the detection sequence, prioritizing the processing of areas with high misjudgment superposition values. If it has an impact, the scattering control coefficient is obtained, and the digital X-ray imaging is adjusted according to the scattering control coefficient. This can appropriately increase the energy of the X-ray source, adjust parameters such as the distance or angle between the X-ray source and the workpiece, reduce the generation of scattering, thereby improving image quality and making the edges of defects such as pores clearer, which is convenient for accurate identification and measurement.

[0102] Misjudgment and superposition analysis module: Within multiple historical scattering detection cycles, the module obtains the degree of misjudgment of pore size in the pore detection image corresponding to the high scattering superposition layer, and performs correlation analysis with the degree of scattering superposition to evaluate whether the degree of scattering superposition affects the pore size.

[0103] In some embodiments, the image of a single pore layer corresponding to the highly scattering superposition layer is extracted, and the size detection value of each pore in the single pore layer image is obtained. The size detection values ​​are then sorted according to the time sequence of the obtained size detection values ​​to construct a detection size sequence.

[0104] Obtain the actual size value of each pore in a single-layer pore image, and construct an actual size sequence according to the construction method of the size detection sequence;

[0105] Obtain any one pore in the single-layer pore image as the comparison pore, and extract the size detection value of the comparison pore in the detection size sequence and the actual size value in the actual size sequence respectively;

[0106] According to the method of obtaining the corresponding size detection value and actual size value of the pores, each pore in the single-layer pore image is acquired separately and combined to obtain multiple size comparison groups;

[0107] Within the size comparison group, the difference between the detected size value and the actual size value of the air hole is calculated, the absolute value is taken, and the ratio with the actual size value is calculated to obtain the size misjudgment degree value;

[0108] Extract the overlap angle ratio corresponding to the pores and compare it with the overlap angle ratio threshold. Take the absolute value to obtain the scattering superposition degree value.

[0109] By acquiring the size misjudgment value and scattering superposition value of each pore in a single-layer pore image, respectively, the size misjudgment value and scattering superposition value of each pore are obtained by comparing the size misjudgment value and the scattering superposition value of the corresponding pore.

[0110] The average value of size misjudgment is calculated by averaging the size misjudgment of each pore in the single-layer pore image.

[0111] The standard deviation of the size misjudgment degree value of each pore in the single-layer pore image is calculated to obtain the size misjudgment standard deviation;

[0112] The scattering superposition value of each pore in the single-layer pore image is averaged to obtain the average scattering superposition value;

[0113] The standard deviation of the scattering superposition value of each pore in the single-layer pore image is calculated to obtain the scattering superposition standard deviation;

[0114] The mean value of size misjudgment, the standard deviation of size misjudgment, the mean value of scattering superposition, and the standard deviation of scattering superposition are respectively input into the Pearson correlation coefficient formula to obtain the misjudgment superposition value;

[0115] If the misjudged superposition value is greater than the misjudged superposition threshold, it indicates that the signal is a misjudged superposition associated signal.

[0116] If the misjudged superposition value is less than or equal to the misjudged superposition threshold, it indicates that the signal is displayed as a misjudged superposition of non-associated signals.

[0117] It should be noted that the significance of analyzing the correlation between size misjudgment and scattering superposition lies in the following: When using digital X-ray imaging to detect pore size, size misjudgment occurs. Furthermore, when evaluating X-ray scattering during digital X-ray imaging detection, scattering of the detected X-rays is observed. Therefore, from the perspective of imaging detection accuracy, analyzing the correlation between size misjudgment and scattering superposition can determine whether the error in pore size detection is related to scattering superposition. If a misjudgment superposition correlation signal is displayed, it indicates that the degree of scattering superposition has a significant impact on pore size detection, helping to accurately identify the root cause of the detection error. From the perspective of imaging detection efficiency, it allows for the rational arrangement of the detection sequence, prioritizing areas with high misjudgment superposition values, and the rational allocation of human and equipment resources. More effort and advanced equipment can be invested in the detection stages with serious problems, thereby improving the overall efficiency of the detection work.

[0118] Detection and control adjustment module: If it affects the size of the pores, it obtains the scattering control coefficient and adjusts the digital X-ray imaging according to the scattering control coefficient;

[0119] In some embodiments, within a size comparison group, the ratio of the size misjudgment degree value to the scattering superposition degree value is calculated, and the degree coefficient is output.

[0120] The scattering control coefficient is obtained by averaging the degree coefficients corresponding to each size comparison group and then outputting them.

[0121] Obtain the overlap angle ratio of adjacent layers with high dispersion, and calculate the scattering control quantity by multiplying the overlap angle ratio with the degree coefficient.

[0122] It should be noted that the significance of obtaining the scattering control value is that it reflects the degree of influence of scattering on the detection and enables targeted adjustment of the X-ray source parameters of digital X-ray imaging. When the scattering control value is large, it indicates that scattering seriously interferes with the detection results. At this time, the energy of the X-ray source can be appropriately increased, and parameters such as the distance or angle between the X-ray source and the workpiece can be adjusted to reduce the generation of scattering, thereby improving the image quality and making the edges of defects such as pores clearer, which is convenient for accurate identification and measurement.

[0123] The specific solution in this embodiment is as follows: Within multiple historical scattering detection cycles, the degree of misjudgment of pore size in the pore detection image corresponding to the highly scattering superposition layer is obtained, and a correlation analysis is performed with the degree of scattering superposition to assess whether scattering superposition affects pore size. This can determine whether the error in pore size detection is related to the scattering superposition phenomenon, which helps to accurately find the root cause of the detection error and can reasonably arrange the detection sequence, prioritizing the processing of areas with high misjudgment superposition values. If it has an impact, the scattering control coefficient is obtained, and the digital X-ray imaging is adjusted according to the scattering control coefficient. This can appropriately increase the energy of the X-ray source, adjust parameters such as the distance or angle between the X-ray source and the workpiece, reduce the generation of scattering, thereby improving image quality and making the edges of defects such as pores clearer, which is convenient for accurate identification and measurement.

[0124] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. An industrial digital X-ray imaging detection and control system, characterized in that: Scattering Period Screening Module: When using digital X-rays to detect pores in multi-layer workpieces, the scattering analysis of digital X-rays in multiple imaging detection periods is performed to screen out the scattering detection periods. High-dispersion superposition extraction module: Performs inter-layer scattering superposition analysis on adjacent detection scattering periods within the scattering detection cycle, determines whether scattering superposition occurs between adjacent layers, and extracts adjacent layers with high dispersion superposition. Misjudgment and superposition analysis module: Within multiple historical scattering detection cycles, the module obtains the degree of misjudgment of pore size in the pore detection image corresponding to the high scattering superposition layer, and performs correlation analysis with the degree of scattering superposition to evaluate whether the degree of scattering superposition affects the pore size. Detection and control adjustment module: If it affects the size of the pores, it obtains the scattering control coefficient and adjusts the digital X-ray imaging according to the scattering control coefficient.

2. The industrial digital X-ray imaging detection and control system according to claim 1, characterized in that: The digital X-ray scattering analysis is performed over multiple imaging detection cycles, as follows: One imaging detection cycle is selected from multiple imaging detection cycles as the target detection cycle, and it is equally divided into several target detection time periods. In each target detection time period, each pore in different workpiece layers is detected to obtain pore detection images in different workpiece layers as single-layer pore images. The size detection value of each pore in the single-layer pore image and the actual size value of each pore are obtained respectively, and then input into the Euclidean distance formula to output the size difference value.

3. The industrial digital X-ray imaging detection and control system according to claim 1, characterized in that: The process for selecting the scattering detection cycle is as follows: If the periodic screening value is greater than the periodic screening threshold, the target detection period is marked as the scattering detection period.

4. The industrial digital X-ray imaging detection and control system according to claim 1, characterized in that: Interlayer scattering superposition analysis is performed on adjacent detection scattering periods within the scattering detection cycle to obtain the fitted ellipse of the fore-target and the fitted ellipse of the subsequent target. The process is as follows: Within the scattering detection period, adjacent detection scattering time periods are combined to obtain multiple detection scattering combinations. Within each detection scattering combination, the single-layer pore images corresponding to the previous and subsequent detection scattering time periods are extracted as the previous and subsequent single-layer pore images. The same pore in the previous and subsequent single-layer pore images is extracted as the target pore. The contours of the target stomata in the front and back single-layer stomata images are extracted respectively, and ellipse fitting is performed based on the contours of the target stomata to obtain the front target fitted ellipse and the back target fitted ellipse.

5. The industrial digital X-ray imaging detection and control system according to claim 1, characterized in that: The process for determining whether scattering superposition occurs between adjacent layers is as follows: Obtain the lengths of the major and minor axes of the fitted ellipse for the front and rear targets respectively, and calculate the ratios to obtain the ratios of the major and minor axes of the front and rear targets. If the aspect ratio of the front target and the aspect ratio of the rear target are both greater than the target aspect ratio threshold, they will be displayed as bar-shaped signals of the front and rear targets. Extract columnar pore images corresponding to the target pores in the front and rear single-layer pore images respectively, and obtain the angle between the long axis in the columnar pore image and the Z-axis in the single-layer workpiece 3D model respectively, as the front target scattering angle and the rear target scattering angle. According to the method of obtaining the scattering angle of the front target and the scattering angle of the back target, the scattering angle of the remaining pores in the front single-layer pore image and the back single-layer pore image are obtained respectively, and the front scattering angle sequence and the back scattering angle sequence are constructed according to the order of obtaining the scattering angle. Extract the maximum and minimum front target scattering angles within the forward scattering angle sequence, and the maximum and minimum front target scattering angles within the backward scattering angle sequence, to construct the front target scattering angle set and the backward target scattering angle set; If there is an intersection between the set of scattering angles of the front target and the set of scattering angles of the rear target, the angle overlap range corresponding to the intersection is extracted, and the ratio is calculated with the range corresponding to the set of scattering angles of the front target. The overlap angle ratio is output. If the overlap angle ratio is greater than or equal to the overlap angle ratio threshold, it is displayed as a scattering superposition signal.

6. The industrial digital X-ray imaging detection and control system according to claim 1, characterized in that: The extraction process for high-dispersion stacked adjacent layers is as follows: The overlap angle ratios corresponding to each detection and dispersion combination are compared, and the pre-detection scattering time period and post-detection scattering time period corresponding to the detection and dispersion combination with the largest overlap angle ratio are selected and mapped to obtain the high dispersion superposition adjacent layers.

7. The industrial digital X-ray imaging detection and control system according to claim 1, characterized in that: The process for obtaining the degree of scattering superposition is as follows: Extract the single-layer stomatal image corresponding to the highly scattering superposition layer, obtain the size detection value of each stomatal in the single-layer stomatal image, and sort them according to the time series of the obtained size detection values ​​to construct the detection size sequence; Obtain the actual size value of each pore in a single-layer pore image, and construct an actual size sequence according to the construction method of the size detection sequence; Obtain any one pore in the single-layer pore image as the comparison pore, and extract the size detection value of the comparison pore in the detection size sequence and the actual size value in the actual size sequence respectively; According to the method of obtaining the corresponding size detection value and actual size value of the pore, each pore in the single-layer pore image is obtained separately and combined to obtain multiple size comparison groups. Within the size comparison group, the difference between the corresponding size detection value and the actual size value of the pore is calculated, the absolute value is taken, and the ratio with the actual size value is calculated to obtain the size misjudgment degree value. Extract the overlap angle ratio corresponding to the pores and compare it with the overlap angle ratio threshold. Take the absolute value to obtain the scattering superposition degree value.

8. The industrial digital X-ray imaging detection and control system according to claim 1, characterized in that: The process of association analysis is as follows: By acquiring the size misjudgment value and scattering superposition value of each pore in a single-layer pore image, respectively, the size misjudgment value and scattering superposition value of each pore are obtained by comparing the size misjudgment value and the scattering superposition value of the corresponding pore. The mean and standard deviation of the size misjudgment value of each pore in the single-layer pore image are calculated separately to obtain the mean and standard deviation of size misjudgment. The scattering superposition degree value of each pore in the single-layer pore image is averaged and the standard deviation is calculated to obtain the scattering superposition mean and scattering superposition standard deviation. The mean value of size misjudgment, the standard deviation of size misjudgment, the mean value of scattering superposition, and the standard deviation of scattering superposition are respectively input into the Pearson correlation coefficient formula to obtain the misjudgment superposition value; If the misjudged superposition value is greater than the misjudged superposition threshold, it will be displayed as a misjudged superposition associated signal.

9. The industrial digital X-ray imaging detection and control system according to claim 1, characterized in that: The process of obtaining the scattering control coefficients is as follows: Within the size comparison group, the ratio of the size misjudgment degree value to the scattering superposition degree value is calculated, and the degree coefficient is output. The scattering control coefficients are calculated by averaging the degree coefficients corresponding to each size comparison group.

10. An industrial digital X-ray imaging detection and control system according to claim 1, characterized in that: The process of controlling digital X-ray imaging based on the scattering control coefficient is as follows: The scattering control coefficient is obtained by averaging the degree coefficients corresponding to each size comparison group and then outputting them. Obtain the overlap angle ratio of adjacent layers with high dispersion, and multiply the overlap angle ratio with the degree coefficient to calculate the scattering control quantity.

Images