Geneva wheel detection method and device, computer equipment and storage medium

Abstract

The invention relates to a sheave detection method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring wire groove morphology data of a to-be-detected grooved wheel; obtaining a statistical characteristic value according to the wire groove morphology data, wherein the statistical characteristic value is used for representing the wire groove depth of the to-be-measured grooved wheel; based on the statistical characteristic value and a preset operation rule, calculating a trunking abnormity threshold value of the to-be-detected sheave; and performing anomaly detection on the to-be-detected sheave according to the trunking anomaly threshold and the trunking morphology data. According to the method, the trunking anomaly threshold value is calculated based on the statistical characteristic value of the trunking morphology data of the sheave, and the trunking anomaly detection efficiency and accuracy are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of semiconductor equipment, and in particular to a groove wheel detection method and device, computer equipment and a storage medium. BACKGROUND

[0002] In the production process of silicon wafers, a single crystal ingot drawn by a Czochralski method is usually cut by a multi-wire saw to obtain silicon wafers. The multi-wire saw usually includes two opposite wire rollers or groove wheels, a single cutting wire is wound in the wire groove of the two wire rollers to obtain a plurality of cutting wire segments parallel to each other and in the same plane, and the rotation of the wire roller drives the movement of the plurality of cutting wire segments to complete the cutting of the single crystal ingot. After the multi-wire saw is used for a period of time, the cutting wire wound on the rotating shaft will cause wear to the wire groove of the rotating shaft, and when the rotating shaft is severely worn, it will cause problems such as wire jumping and wire breaking during wire cutting, and also cause degradation loss of silicon wafers.

[0003] In the prior art, in order to reduce the loss caused by the wire groove problem of the groove wheel, an artificial detection method is used. However, the artificial detection method is low in efficiency, and since there are many wire grooves on the groove wheel, the accuracy of wire groove detection is also low. In order to improve the efficiency, an image sensor is used to collect a wire groove image in the related art, and a groove type anomaly detection is performed based on a comparison result of the wire groove image and a standard image. However, due to the influence of the wire groove processing technology and the use wear, there is a difference between the actual wire groove and the standard wire groove in shape, and the direct use of the standard image comparison method for groove type detection has the problem of low detection accuracy.

[0004] There is no effective solution to the problem of low detection accuracy of the groove type detection in the related art. SUMMARY

[0005] Therefore, it is necessary to provide a groove wheel detection method, device, computer equipment and storage medium capable of improving the detection accuracy of the groove type.

[0006] In a first aspect, the present application provides a groove wheel detection method. The method comprises:

[0007] obtaining wire groove topography data of a to-be-detected groove wheel;

[0008] obtaining a statistical feature value according to the wire groove topography data, the statistical feature value being used to represent a wire groove depth of the to-be-detected groove wheel;

[0009] calculating a wire groove anomaly threshold of the to-be-detected groove wheel based on the statistical feature value and a preset operation rule;

[0010] performing anomaly detection on the to-be-detected groove wheel according to the wire groove anomaly threshold and the wire groove topography data.

[0011] In one of the embodiments, the acquiring the linear slot profile data of the to-be-tested groove wheel comprises: acquiring a detection image of the to-be-tested groove wheel; identifying pixel values of each pixel point in the detection image to obtain pixel data; and calculating the mean value of columns in the pixel matrix to obtain the linear slot profile data.

[0012] In one of the embodiments, the calculating the mean value of columns in the pixel matrix to obtain the linear slot profile data further comprises: acquiring a mapping relationship between the pixel value and the sampling point depth; and obtaining the linear slot profile data according to the mean value of the columns of the pixel matrix and the mapping relationship between the pixel value and the sampling point depth.

[0013] In one of the embodiments, the acquiring the statistical characteristic value according to the linear slot profile data comprises: drawing a linear slot profile curve according to the linear slot profile data, wherein the horizontal position of the linear slot on the groove wheel is represented by the horizontal coordinate of the linear slot profile curve, and the surface height of the linear slot is represented by the vertical coordinate of the linear slot profile curve; and performing sorting statistics on the vertical coordinates of each point on the linear slot profile curve to obtain the statistical characteristic value.

[0014] In one of the embodiments, the drawing the linear slot profile curve according to the linear slot profile data comprises: drawing an initial linear slot curve according to the linear slot profile data; performing peak searching processing on the initial profile curve to obtain a target peak point; performing linear fitting based on the target peak point to obtain a fitting function; and performing vertical correction on the initial profile curve according to the fitting function to obtain the linear slot profile curve.

[0015] In one of the embodiments, the performing vertical correction on the initial profile curve according to the fitting function to obtain the linear slot profile curve comprises: substituting the horizontal coordinates of each point on the initial profile curve into the fitting function to obtain corresponding fitting vertical coordinates; selecting a calibration point from the fitting vertical coordinates, calculating the vertical difference between the other fitting vertical coordinates and the calibration point, and taking the vertical difference as the vertical correction amount; and subtracting the vertical coordinates of each point on the initial profile curve from the corresponding vertical correction amount to obtain target vertical coordinates; and obtaining the linear slot profile curve based on the horizontal coordinates of each point on the initial profile curve and the corresponding target vertical coordinates.

[0016] In one of the embodiments, the determining the abnormality identification threshold according to the statistical characteristic value and the threshold calculation rule comprises: determining the upper limit threshold and the lower limit threshold of the linear slot on the to-be-tested groove wheel according to the statistical characteristic value; acquiring a pre-designed calculation parameter, wherein the pre-designed calculation parameter is determined according to the linear slot depth of the abnormal groove type; and determining the abnormality identification threshold of the to-be-tested groove wheel according to the upper limit threshold, the lower limit threshold, and the pre-designed calculation parameter.

[0017] In a second aspect, the application further provides a groove wheel detection device. The device comprises:

[0018] an acquisition module configured to acquire linear slot topography data of a to-be-tested slot wheel;

[0019] a statistics module configured to acquire a statistics characteristic value according to the linear slot topography data, the statistics characteristic value being used to represent a linear slot depth of the to-be-tested slot wheel;

[0020] a calculation module configured to calculate a linear slot anomaly threshold of the to-be-tested slot wheel based on the statistics characteristic value and a preset operation rule;

[0021] a judgment module configured to perform anomaly detection on the to-be-tested slot wheel according to the linear slot anomaly threshold and the linear slot topography data.

[0022] In a third aspect, the present application further provides a computer device. The computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0023] acquire linear slot topography data of a to-be-tested slot wheel;

[0024] acquire a statistics characteristic value according to the linear slot topography data, the statistics characteristic value being used to represent a linear slot depth of the to-be-tested slot wheel;

[0025] calculate a linear slot anomaly threshold of the to-be-tested slot wheel based on the statistics characteristic value and a preset operation rule;

[0026] perform anomaly detection on the to-be-tested slot wheel according to the linear slot anomaly threshold and the linear slot topography data.

[0027] In a fourth aspect, the present application further provides a computer readable storage medium. The computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the following steps:

[0028] acquire linear slot topography data of a to-be-tested slot wheel;

[0029] acquire a statistics characteristic value according to the linear slot topography data, the statistics characteristic value being used to represent a linear slot depth of the to-be-tested slot wheel;

[0030] calculate a linear slot anomaly threshold of the to-be-tested slot wheel based on the statistics characteristic value and a preset operation rule;

[0031] perform anomaly detection on the to-be-tested slot wheel according to the linear slot anomaly threshold and the linear slot topography data.

[0032] The aforementioned Geneva wheel detection method, apparatus, computer equipment, and storage medium acquire groove morphology data of the Geneva wheel under test, obtain statistical feature values ​​based on the groove morphology data, calculate a groove anomaly threshold for the Geneva wheel under test based on the feature statistical values ​​and preset calculation rules, and perform anomaly detection on the Geneva wheel under test based on the groove anomaly threshold and the groove morphology data. This image-based Geneva wheel detection method improves detection efficiency compared to manual detection techniques; calculating the groove anomaly threshold based on the statistical feature values ​​of the groove morphology data is more targeted than directly setting a standard threshold, thus improving the detection accuracy of Geneva wheel anomalies. Attached Figure Description

[0033] Figure 1 This is a diagram illustrating the application environment of the Geneva wheel detection method in one embodiment;

[0034] Figure 2 This is a flowchart illustrating a Geneva detection method in one embodiment;

[0035] Figure 3 This is a schematic diagram of the Geneva wheel under test in one embodiment;

[0036] Figure 4 This is a schematic diagram of the groove morphology curve in one embodiment;

[0037] Figure 5 This is a schematic diagram of the groove morphology curve before and after correction in one embodiment;

[0038] Figure 6 This is a schematic diagram of the statistical results of statistical feature values ​​in one embodiment;

[0039] Figure 7 This is a flowchart illustrating the Geneva detection method in another embodiment;

[0040] Figure 8 This is a structural block diagram of the Geneva wheel detection device in one embodiment;

[0041] Figure 9 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0043] The Geneva wheel detection method provided in this application embodiment can be applied to, for example... Figure 1The application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data required by the server 104 to process. The data storage system can be integrated on the server 104, or placed on the cloud or other network servers. The terminal 102 includes a data acquisition device and an information display device, the data acquisition device is used to acquire the wire slot topography data, including but not limited to camera, image sensor, photoelectric sensor and the like; the information display device is used to display the detection result of the slot wheel. The terminal 102 can also include a computing device such as a processor, etc., to process the collected wire slot topography data, such as calculating statistical characteristic values according to the wire slot topography data, calculating the wire slot abnormal threshold based on the statistical characteristic values and the preset operation rules. The function of the computing device can also be realized by the server. The data storage system is used to store the wire slot topography data. The server 104 can be realized by an independent server or a server cluster composed of multiple servers.

[0044] In one embodiment, as shown in Figure 2 , a slot wheel detection method is provided, which is applied to the terminal 102 in Figure 1 for example, including the following steps:

[0045] Step S201, acquiring the wire slot topography data of the slot wheel to be measured.

[0046] Among them, the slot wheel to be measured refers to the wire spool or wire roller used to wind the cutting wire. The wire slot topography data is the data representing the depth of the wire slot on the surface of the slot wheel, for example, the wire slot topography data can be the coordinates of each point on the surface of the wire slot. The wire slot topography data can be collected by camera, image sensor, depth camera, photoelectric sensor, three-dimensional scanner and the like.

[0047] Step S202, acquiring statistical characteristic values according to the wire slot topography data, the statistical characteristic values are used to represent the wire slot depth of the slot wheel to be measured.

[0048] Specifically, after acquiring the wire slot topography data, the wire slot depth of each wire slot in the current slot wheel can be obtained based on the wire slot topography data, and the wire slot depth of all wire slots is statistically analyzed to obtain the wire slot depth distribution of the slot wheel to be measured. The statistical characteristic values can be determined by calculating the wire slot depth, and can also be determined by calculating the slot stem height.

[0049] Step S203, calculating the wire slot abnormal threshold of the slot wheel to be measured based on the statistical characteristic values and the preset operation rules.

[0050] Specifically, the preset operation rule is an operation rule of a wire slot abnormality threshold value designed according to production process requirements and configured according to an abnormal working condition of the wire slot. In the related art, the detection of the wire slot abnormality is usually performed before the wire winding of the slot wheel, and the wire slot depth is compared with a standard value set by a human being. If the wire slot depth is not within the required range of the standard value, it is determined to be abnormal. The detection method has the following disadvantages. First, due to the influence of the machining conditions of the slot wheel, the quality of the slot wheel of different batches itself has different morphology differences. A single standard value cannot adapt to the detection scheme of all slot wheels. Second, as the slot wheel is used, the wire slot is inevitably worn. At this time, the main concern in the production requirements is whether there is a rotten slot in the slot wheel. The rotten slot refers to the wire slot depth of one or a plurality of continuous wire slots being significantly lower than the wire slot depth of other wire slots. The rotten slot is easy to cause the problems of wire jumping and wire merging of the cutting wire net, thereby affecting the quality of the silicon wafer cutting and increasing the material loss. The existing standard value detection scheme is easy to judge the wire slot with a reduced wire slot depth but still usable as a rotten slot with the wear of the wire slot, thereby increasing the misjudgment rate. The preset operation rule in the embodiment calculates the threshold value based on the statistical characteristic value distribution of the wire slot depth of the current slot wheel. The wire slot abnormality threshold value can be adaptively adjusted with the wear of the wire slot, thereby reducing the misjudgment and improving the detection accuracy of the rotten slot condition.

[0051] In step S204, the abnormality of the to-be-tested slot wheel is detected according to the wire slot abnormality threshold value and the wire slot morphology data.

[0052] Specifically, the wire slot morphology data is compared with the wire slot abnormality threshold value. If the wire slot morphology data does not meet the requirements of the wire slot abnormality threshold value, it is determined that the corresponding wire slot of the to-be-tested slot wheel has an abnormal condition. For example, when the wire slot abnormality threshold value is a slot stem height threshold value, if the slot stem height of the current wire slot is detected to be less than the wire slot abnormality threshold value, it is determined that the current wire slot has a rotten slot condition. For another example, when the wire slot abnormality threshold value is a wire slot depth threshold value, if the wire slot depth of the current wire slot is detected to be less than the wire slot abnormality threshold value, it is determined that the current wire slot has a rotten slot condition. When the wire slot abnormality of the to-be-tested slot wheel is detected, an alarm processing can be performed and the abnormal wire slot position can be output. In one of the embodiments, in order to reduce the misjudgment, a supplementary judgment condition can be added. When a plurality of adjacent wire slots are detected to be abnormal, it is determined that the current slot wheel has a wire slot abnormality.

[0053] In the slot wheel detection method, the statistical characteristic value about the wire slot depth is obtained through statistical analysis of the topographic detection data of the current slot wheel, the real-time wire slot anomaly threshold is calculated according to the statistical characteristic value and a preset operation rule, compared with the scheme of directly setting a standard value for detection in the related art, all the wire slot topographic data of the to-be-detected slot wheel is introduced into the calculation of the wire slot anomaly threshold, the detection adaptability and pertinence of the wire slot anomaly threshold to the current to-be-detected slot wheel are improved, the detection accuracy of the burnt slot anomaly is higher, and the accuracy of the slot wheel detection is improved.

[0054] In one of the embodiments, the obtaining of the wire slot topographic data of the to-be-detected slot wheel includes: obtaining a detection image of the to-be-detected slot wheel; identifying pixel values of each pixel point in the detection image to obtain pixel data; and obtaining the wire slot topographic data by averaging columns in the pixel matrix.

[0055] Specifically, Figure 3 For a to-be-detected slot wheel image in one of the embodiments, the image of the to-be-detected slot wheel is collected by an image sensor or a camera, and the wire slot topographic data is identified based on the image. The advantage of using the image sensor is that when the wire slot topographic data is collected, the wire mesh area can be shielded in the settings of the image sensor, and only the slot wheel image is left. Then, through pixel value operation, only the slot wheel information is included in the collected wire slot topographic data. In the related art, in order to reduce the interference of the cutting line on the wire slot type detection, the wire slot topographic data is basically collected before wiring. The topographic data collection scheme based on the image sensor in the embodiment can shield the cutting line, so that the slot wheel detection can be performed under the wiring condition of the slot wheel, thereby increasing the flexibility and real-time performance of the detection.

[0056] In one of the embodiments, the obtaining of the wire slot topographic data by averaging the columns in the pixel matrix further includes: obtaining a mapping relationship between the pixel value and the sampling point depth; and obtaining the wire slot topographic data according to the average value of the columns of the pixel matrix and the mapping relationship between the pixel value and the sampling point depth.

[0057] Specifically, the sampling point depth is used to represent the spatial depth information of each sampling point on the wire slot surface. The mapping relationship between the pixel value and the sampling point depth can be calibrated and stored based on the data of the actual sample. In subsequent detection of the slot wheel, the mapping relationship can be directly called.

[0058] In one of the embodiments, the obtaining of the statistical characteristic value according to the wire slot topographic data includes: drawing a wire slot topographic curve according to the wire slot topographic data, the horizontal position of the wire slot on the slot wheel is represented by the horizontal coordinate of the wire slot topographic curve, and the surface height of the wire slot is represented by the vertical coordinate of the wire slot topographic curve; and performing sorting statistics on the vertical coordinates of each point on the wire slot topographic curve to obtain the statistical characteristic value.

[0059] Specifically, as Figure 4As shown, after the line groove topography data is obtained, a line groove topography curve is drawn with the horizontal position of the line groove as the horizontal coordinate and the surface height of the line groove as the vertical coordinate, wherein the peak of the line groove topography curve represents the groove stem of the line groove; the peak point is the vertex of the groove stem; the trough point is the groove bottom; the distance between two troughs is the groove stem width; the distance between two peaks is the line groove width; and the vertical distance between adjacent peak and trough is the line groove depth, i.e., the groove stem height. In this embodiment, the vertical coordinates of the points on the line groove topography curve are sorted in descending or ascending order, and the statistical indicators such as mode, median, and average value are calculated, so that the statistical characteristic values are obtained, and the approximate distribution of the line groove topography of the groove wheel surface can be analyzed through the statistical characteristic values.

[0060] In one of the embodiments, drawing the line groove topography curve according to the line groove topography data includes: drawing an initial line groove curve according to the line groove topography data; performing peak searching on the initial topography curve to obtain target peak points; performing linear fitting based on the target peak points to obtain a fitting function; and performing vertical correction on the initial topography curve according to the fitting function to obtain the line groove topography curve.

[0061] Specifically, in actual production conditions, the image sensor and its installation structure cannot be kept absolutely horizontal, and therefore, the line groove image collected by the image sensor will be inclined to different degrees. In order to reduce the error caused by the inclination to the detection result and improve the accuracy of the detection, in this embodiment, the line groove topography curve is straightened, i.e., the fitting function obtained through the target peak points is used to correct the line groove topography curve. Through the correction of the line groove topography curve, the detection accuracy of the groove wheel detection is improved, and the misjudgment of the line groove anomaly is reduced.

[0062] In one of the embodiments, performing vertical correction on the initial topography curve according to the fitting function to obtain the line groove topography curve includes: substituting the horizontal coordinates of the points on the initial topography curve into the fitting function to obtain corresponding fitting vertical coordinates; selecting a calibration point from the fitting vertical coordinates and calculating the vertical difference between the other fitting vertical coordinates and the calibration point, wherein the vertical difference is used as the vertical correction amount; subtracting the vertical coordinates of the points on the initial topography curve from the corresponding vertical correction amounts to obtain target vertical coordinates; and obtaining the line groove topography curve based on the horizontal coordinates of the points on the initial topography curve and the corresponding target vertical coordinates.

[0063] Specifically, the embodiment provides a longitudinal correction amount calculation method, including: substituting the abscissa of each point on the initial profile curve into a fitting function to obtain a fitted ordinate. A point based on the abscissa and the corresponding fitted ordinate is referred to as a fitting point. The points on the initial profile curve with the same abscissa correspond to the fitting points. Any fitting point is selected as a calibration point, and preferably, the first fitting point is selected as the calibration point. The difference between the ordinate of the calibration point and the ordinate of each fitting point is calculated, that is, a longitudinal difference value, to obtain the longitudinal correction amount. The ordinate of each point on the corresponding initial profile curve is subtracted by the corresponding longitudinal correction amount, to obtain a corrected ordinate. Further, the corrected ordinate and the corresponding abscissa are used to obtain a corrected wire slot profile curve. The wire slot profile curve is statistically analyzed to determine an abnormality identification threshold. Figure 5 For a schematic diagram of the wire slot profile curve before and after correction in an embodiment, as shown in Figure 5 , the abscissa is the horizontal position of the wire slot, and the ordinate is the surface height of the wire slot. The blue line segment is the initial profile curve, the red straight line is the fitting curve corresponding to the fitting function, and the green line segment is the corrected wire slot profile curve. It can be seen from Figure 5 that the initial profile curve is straightened to obtain the wire slot profile curve, and the wave crest of the wire slot profile curve is basically located on a horizontal straight line. The longitudinal correction amount calculation method of the embodiment improves the detection accuracy of the groove wheel detection by correcting the wire slot profile curve.

[0064] In one of the embodiments, the determining of the abnormality identification threshold according to the statistical characteristic value and the threshold calculation rule includes: determining an upper limit threshold and a lower limit threshold of the wire slot of the to-be-detected groove wheel according to the statistical characteristic value; obtaining a pre-design calculation parameter, the pre-design calculation parameter being determined according to the wire slot depth of the abnormal slot type; and determining the abnormality identification threshold of the to-be-detected groove wheel according to the upper limit threshold, the lower limit threshold, and the pre-design calculation parameter.

[0065] Specifically, the statistical characteristic value can be used to determine the wire slot depth distribution of the to-be-detected groove wheel at this time. For example, the maximum slot stem height and the minimum slot bottom height of the to-be-detected groove wheel under the current working condition are counted, the maximum slot stem height is the upper limit threshold, and the minimum slot bottom height is the lower limit threshold. The pre-design calculation parameter is determined based on the production process requirement, and the pre-design calculation parameter is used to represent the size of the allowable error of the wire slot abnormality threshold. Generally, the pre-design calculation parameter is several times of the difference between the upper limit threshold and the lower limit threshold. The larger the pre-design calculation parameter is, the higher the requirement for the slot stem height of the wire slot is. Compared with the scheme of directly setting the threshold without considering the machining properties and wear conditions of the groove wheel in the related art, the abnormality identification threshold calculation method of the embodiment calculates the abnormality identification threshold based on the wire slot profile data of the to-be-detected groove wheel, and can be adaptively adjusted with the machining process, thereby improving the detection accuracy of the groove wheel abnormality.

[0066] In one embodiment, based on the ranking result of the statistical characteristic values, the median of the statistical characteristic values in a preset range is taken as the upper threshold and / or the lower threshold, and an abnormal recognition threshold with better recognition effect can be obtained. Figure 6 For a statistical result of the statistical characteristic values in one embodiment, as shown in Figure 6 the horizontal coordinate is the surface height of the wire slot, and the vertical coordinate is the number of wire slots. For example, the slot stem height data is taken at one end of the statistical characteristic value statistical result close to the maximum value, and the slot bottom height data is taken at one end of the statistical result close to the minimum value. The median value of the top 20% of the slot stem height data is taken as the upper limit value, that is, the upper threshold, and the median value of the top 20% of the slot bottom height data is taken as the lower limit value, that is, the lower threshold, and the abnormal recognition threshold is M=(P-(P-Q) / n), to obtain the threshold for judging the rotten slot. Wherein, M is the abnormal recognition threshold, P is the upper threshold, Q is the lower threshold, and n is a preset calculation parameter. In one specific embodiment, n is 3, and the meaning of the abnormal recognition threshold is that if the slot stem height or the slot depth is less than 2 / 3 of the median value of the slot stem height, it is judged that the corresponding wire slot has a rotten slot.

[0067] In one preferred embodiment, as shown in Figure 7 a slot wheel detection method is provided, and the method comprises:

[0068] In step S701, a slot wheel image is obtained.

[0069] The slot wheel image is collected by a sensor. Since the wire mesh is arranged on the slot wheel, the wire mesh area is shielded in the sensor setting, so that only the slot wheel image can be saved.

[0070] In step S702, the collected image is preprocessed to obtain an initial wire slot curve.

[0071] The collected image is preprocessed to obtain the depth value of each coordinate in the image, which is represented by a pixel value, to obtain a two-dimensional array of image depth values. The invalid wire mesh value is replaced with a NaN value to facilitate subsequent calculation. The vertical mean value of the two-dimensional array is obtained, and each mean value is the wire slot state value of the corresponding wire slot. Based on the state value and the wire slot position, an initial topography curve is drawn, and the obtained initial topography curve is a curve graph with periodic fluctuations, wherein the distance between every two peaks is the slot pitch.

[0072] In step S703, the initial topography curve is graphically corrected to obtain a wire slot topography curve.

[0073] Since in reality the camera module and mechanical structure cannot be kept absolutely horizontal, the slot wheel image obtained will be inclined to varying degrees. In order to avoid the influence of such inclination on the detection result, the initial profile curve of the inclination needs to be corrected, which includes: first, performing a peak value finding operation on the initial profile curve. Based on the found peak value and the index corresponding to the peak value, a linear fitting is performed to obtain a fitting straight line. After fitting, the actual data is straightened based on the longitudinal distance of the actual data and the fitting straight line to obtain a relatively horizontal slot wheel state image, i.e., a linear slot profile curve.

[0074] Step S704, calculate the linear slot anomaly threshold.

[0075] After straightening, statistics and calculation are performed to obtain the threshold for detecting rotten slots, i.e., the linear slot anomaly threshold. This includes: obtaining an upper threshold (the median of the maximum value of the first 20% of the data) and a lower threshold (the median of the minimum value of the first 20% of the data) according to the data distribution of the slot wheel state value. According to the upper threshold, the lower threshold, and the threshold calculation formula (threshold = upper limit value - (upper limit value - lower limit value) / n, n = 3), the threshold for judging rotten slots is obtained.

[0076] Step S705, compare the anomaly recognition threshold with the linear slot profile data to obtain the anomaly detection result of the slot wheel.

[0077] The linear slot profile curve is processed in a sliding window manner, the maximum value in each window is found, and the maximum value is compared with the anomaly recognition threshold. If the maximum value is less than the anomaly recognition threshold, it is diagnosed as a rotten slot, and the diagnosis result and the position of the rotten slot are output for technicians to find and repair. The size of the window is one slot pitch distance to ensure that each sliding window contains a peak value.

[0078] Through the slot type detection method of the embodiment of the present application, slot wheel detection can be performed based on the automatic threshold setting mode, without the need for manual parameter measurement, reducing labor costs and improving accuracy. The detection can be performed in the case of a line network, increasing the flexibility and real-time performance of the line network.

[0079] It should be understood that although the steps in the flowcharts involved in the embodiments described above are shown in sequence according to the arrows, the steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, the execution of the steps is not strictly limited in sequence, and the steps can be executed in other orders. Moreover, at least some of the steps in the flowcharts involved in the embodiments described above can include multiple steps or multiple stages, which are not necessarily executed at the same time but can be executed at different times, and the execution order of the steps or stages is not necessarily sequential but can be alternately or alternately executed with at least part of other steps or steps or stages in other steps.

[0080] Based on the same inventive concept, the embodiments of the present application also provide a groove wheel detection device for implementing the above-mentioned groove wheel detection method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more groove wheel detection device embodiments provided below can refer to the limitations of the groove wheel detection method described above, which will not be repeated here.

[0081] In one embodiment, as shown in Figure 8 a groove wheel detection device is provided, comprising: an acquisition module 10, a statistical module 20, a calculation module 30 and a judgment module 40, wherein:

[0082] The acquisition module 10 is configured to acquire the line groove topography data of the groove wheel to be detected.

[0083] The statistical module 20 is configured to acquire a statistical feature value according to the line groove topography data, wherein the statistical feature value is used to represent the line groove depth of the groove wheel to be detected.

[0084] The calculation module 30 is configured to calculate the line groove anomaly threshold of the groove wheel to be detected based on the statistical feature value and a preset operation rule.

[0085] The judgment module 40 is configured to perform anomaly detection on the groove wheel to be detected according to the line groove anomaly threshold and the line groove topography data.

[0086] The acquisition module 10 is further configured to acquire a detection image of the groove wheel to be detected, identify the pixel value of each pixel point in the detection image to obtain pixel data, and calculate the mean value of the columns in the pixel matrix to obtain the line groove topography data.

[0087] The acquisition module 10 is further configured to acquire a mapping relationship between the pixel value and the sampling point depth, and obtain the line groove topography data according to the mean value of the columns of the pixel matrix and the mapping relationship between the pixel value and the sampling point depth.

[0088] The statistical module 20 is further configured to draw an initial groove profile curve according to the groove profile data, and perform peak searching on the initial groove profile curve to obtain a target peak point, perform linear fitting based on the target peak point to obtain a fitting function, and perform longitudinal correction on the initial groove profile curve according to the fitting function to obtain a groove profile curve.

[0089] The statistical module 20 is further configured to draw an initial groove profile curve according to the groove profile data, and perform peak searching on the initial groove profile curve to obtain a target peak point, perform linear fitting based on the target peak point to obtain a fitting function, and perform longitudinal correction on the initial groove profile curve according to the fitting function to obtain a groove profile curve.

[0090] The statistical module 20 is further configured to draw an initial groove profile curve according to the groove profile data, and perform peak searching on the initial groove profile curve to obtain a target peak point, perform linear fitting based on the target peak point to obtain a fitting function, and perform longitudinal correction on the initial groove profile curve according to the fitting function to obtain a groove profile curve.

[0091] The computing module 30 is further configured to determine an upper threshold value and a lower threshold value of the groove on the to-be-tested groove wheel according to the statistical characteristic value, obtain a pre-design calculation parameter, the pre-design calculation parameter being determined according to a groove depth of an abnormal groove type, and determine an abnormality recognition threshold value of the to-be-tested groove wheel according to the upper threshold value, the lower threshold value and the pre-design calculation parameter.

[0092] The above-described modules in the groove wheel detection device can be all or partially implemented by software, hardware and a combination thereof. The above-described modules can be embedded in or independent of a processor in a computer device in a hardware form, or can be stored in a memory in the computer device in a software form, so as to be called and executed by a processor to perform operations corresponding to the above-described modules.

[0093] In an embodiment, a computer device is provided, which can be a terminal, and an internal structure diagram of the computer device can be as shown in FIG. 6. Figure 9As shown in the figure. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The communication interface of the computer device is used for wired or wireless communication with external terminals. Wireless communication can be achieved through WIFI, mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by the processor to implement a groove wheel detection method. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0094] Those skilled in the art can understand that, Figure 9 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. A specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0095] In one embodiment, a computer device is also provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps in the above method embodiments.

[0096] In one embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the steps in the above method embodiments.

[0097] In one embodiment, a computer program product is provided, including a computer program, and the computer program is executed by a processor to implement the steps in the above method embodiments.

[0098] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties.

[0099] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric memory (Ferroelectric Random Access Memory, FRAM), phase change memory (Phase Change Memory, PCM), graphene memory, etc. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0100] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0101] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent of the present application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of protection of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A method for detecting Geneva wheels, characterized in that, The method includes: Obtain the groove morphology data of the wheel under test; Statistical feature values ​​are obtained based on the groove morphology data, and these statistical feature values ​​are used to characterize the groove depth of the groove wheel to be tested. The groove anomaly threshold of the groove wheel under test is calculated based on the statistical feature values ​​and preset calculation rules. Anomaly detection is performed on the groove wheel to be tested based on the groove anomaly threshold and the groove morphology data.

2. The method according to claim 1, characterized in that, The acquisition of the groove morphology data of the wheel under test includes: Obtain the detection image of the grooved wheel to be tested; Identify the pixel values ​​of each pixel in the detected image to obtain a pixel matrix; The mean value of each column in the pixel matrix is ​​calculated to obtain the groove morphology data.

3. The method according to claim 2, characterized in that, The step of averaging the columns in the pixel matrix to obtain the groove morphology data further includes: Obtain the mapping relationship between pixel values ​​and sampling point depth; The groove topography data is obtained based on the mean of the columns of the pixel matrix and the mapping relationship between pixel values ​​and sampling point depth.

4. The method according to claim 1, characterized in that, The step of obtaining statistical feature values ​​based on the groove morphology data includes: Based on the groove topography data, a groove topography curve is plotted. The horizontal axis of the groove topography curve represents the horizontal position of the groove on the groove wheel, and the vertical axis of the groove topography curve represents the surface height of the groove. The statistical characteristic values ​​are obtained by sorting and statistically analyzing the ordinates of each point on the groove morphology curve.

5. The method according to claim 4, characterized in that, The step of drawing the groove topography curve based on the groove topography data includes: The initial groove curve is drawn based on the groove morphology data; Peak finding processing is performed on the initial morphology curve to obtain the target peak point; A linear fit is performed based on the target peak point to obtain the fitting function; The initial morphology curve is longitudinally corrected according to the fitting function to obtain the groove morphology curve.

6. The method according to claim 5, characterized in that, The step of longitudinally correcting the initial morphology curve according to the fitting function to obtain the groove morphology curve includes: Substitute the abscissa of each point on the initial morphology curve into the fitting function to obtain the corresponding fitting ordinate; A calibration point is selected from the fitted ordinate, and the longitudinal difference between the other fitted ordinates and the calibration point is calculated. The longitudinal difference is used as the longitudinal correction amount. The target ordinate is obtained by subtracting the ordinate of each point on the initial morphology curve from the corresponding longitudinal correction. The groove morphology curve is obtained based on the abscissa of each point on the initial morphology curve and the corresponding ordinate of the target.

7. The method according to claim 1, characterized in that, The step of determining the anomaly identification threshold based on the statistical feature values ​​and threshold calculation rules includes: The upper and lower threshold values ​​of the groove on the wheel under test are determined based on the statistical characteristic values. Obtain preset calculation parameters, which are determined based on the groove depth of the abnormal groove type; The anomaly identification threshold of the tested grooved wheel is determined based on the upper limit threshold, the lower limit threshold, and the preset calculation parameters.

8. A Geneva wheel detection device, characterized in that, The device includes: The acquisition module is used to acquire the groove morphology data of the grooved wheel under test; The statistics module is used to obtain statistical feature values ​​based on the groove morphology data, and the statistical feature values ​​are used to characterize the groove depth of the groove wheel to be tested. The calculation module is used to calculate the groove anomaly threshold of the groove wheel under test based on the statistical feature values ​​and preset calculation rules; The judgment module is used to perform anomaly detection on the groove wheel to be tested based on the groove anomaly threshold and the groove morphology data.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images