A multi-angle defect detection method and system for a ring rail

Abstract

The application discloses a kind of multi-angle defect detection method and system for ring guide rail, it is related to defect detection technical field.The multi-angle defect detection method for ring guide rail includes the following steps: guide rail vibration stability detection, ring guide rail image acquisition optimization determination and defect edge detection accuracy evaluation.The application determines whether there is the demand of ring guide rail image acquisition interference evaluation according to the interference evaluation result of guide rail vibration stability, if it is determined that there is demand, whether ring guide rail image acquisition optimization is carried out according to the interference evaluation result of ring guide rail image acquisition, otherwise, defect degree detection is carried out, whether defect degree detection is carried out after ring guide rail image acquisition interference evaluation is qualified, the effect of improving the data accuracy of high-load ring guide rail defect edge detection is achieved, the problem of low data accuracy of high-load ring guide rail defect edge detection in the prior art is solved.

Description

Technical Field

[0001] This invention relates to the field of defect detection technology, and in particular to a method and system for multi-angle defect detection of ring guide rails. Background Technology

[0002] Multi-angle defect detection of circular guide rails generally includes the following steps: First, preparation: Select appropriate inspection equipment based on the material, shape, and size of the guide rail, such as ultrasonic flaw detectors, laser scanners, and vision inspection systems; then, visual inspection: Use manual or automated vision systems to check the guide rail surface for obvious cracks, scratches, corrosion, or other surface defects; then, defect detection, including various methods such as ultrasonic testing, image detection, magnetic particle testing, laser scanning and 3D imaging, and vibration testing. Image detection first uses a ring light source and a high-resolution industrial camera to continuously capture images of the guide rail surface from multiple angles, obtaining grayscale or color images; then, algorithms such as edge enhancement and texture filtering are used to highlight cracked, scratched, or corroded areas; then, morphological operations and connected component analysis are used to remove noise and mark defect contours; finally, the pixel area, aspect ratio, contrast, and other features of the defects are input into a machine learning model for classification to obtain the defect type, location, and severity.

[0003] A circular guide rail typically involves the following key components: a circular track, a slide (slider), a steering slide, a circular disc, bearings, a pinion, a drive system, and a positioning module. The circular track usually consists of two semi-circular slides, which can be single-edge or double-edge designs, providing a precise path for the slide. The slide is a component mounted on and moving along the track, capable of bearing the workload. Slides are typically equipped with bearings and can run on both edges, usually containing rollers or sliders that mate with the track. The steering slide is a crucial component used to guide the slide smoothly at curves on the circular track, accommodating curves including "S" shaped bends or bends of various radii. The system utilizes tracks, with each steering car's rotating frame equipped with specific, freely movable axial / radial balls; the annular disc is typically a rigid base structure for mounting tracks and drive gear rings; bearings are installed at the contact points between the slide and the track or inside the steering gear to achieve low-friction rolling; the pinion, usually driven by a motor, meshes with the gear ring mounted on the annular disc, providing rotational power for the entire system; the drive system includes motors (servo motors, stepper motors, etc.), reducers, drivers, etc., providing power and control for the pinion; the positioning module includes position sensors (such as encoders, linear scales, proximity switches, etc.) and a control system to achieve precise positioning, indexing, and motion control of the slide.

[0004] There are various types of circular guide rails. High-load double-track circular guide rails are a core type designed specifically for industrial applications requiring extremely high loads, high rigidity, and excellent anti-tipping capabilities. Their key feature is the use of two parallel precision tracks. A high-load double-track circular guide rail typically involves the following key components: a load-bearing base, a rotating shaft, a sprocket assembly, pulleys, gears, a chain, a slide rail, a ring rail assembly, a slide connecting plate, a locking plate, a limit rod, a connector, a fixing plate, fixing holes, a cross-shaped connecting block, and a power unit. The load-bearing base serves as the foundation of the rail system, providing overall support. The power unit is installed on the left side inside the load-bearing base and is connected to the sprocket assembly. Rotating shafts are fixedly installed on both sides of the front and back of the load-bearing base. The sprocket assembly, including pulleys and gears, is fitted onto the rotating shafts. The pulleys are fitted around the outer circumference of the rotating shafts to assist rotation, and the gears are fixed to the pulley surfaces. The two gears are connected by a chain to form a closed-loop transmission. The slide rail is fixed to the outer circumference of the support base, forming a ring track. The ring track assembly includes a slide base connecting plate, a snap-fit ​​plate, a limit rod, and a connector. The slide base connecting plate is the main connecting component. The snap-fit ​​plate is welded to one side of the connecting plate and snaps onto the upper and lower surfaces of the slide rail to achieve vertical limitation. The limit rod is threaded onto the four corners of the snap-fit ​​plate, clamping the side of the slide rail to provide horizontal limitation. The connector is fixed to one side of the snap-fit ​​plate and is fixedly connected to the chain to transmit the chain power to the ring track assembly. The fixing plate is located on both sides of the center of the front or back of the support base and is fixed to the base through a cross-shaped connecting block. Fixing holes are opened at the four corners of the fixing plate for installing and fixing the entire track system. The specific working principle is: the power chain drives the ring track assembly to rotate on the slide rail, thereby realizing the outer circumference rotation of the double-track body.

[0005] The existing technology first acquires an image of the guide rail defect edge; then analyzes the target pixels in the guide rail defect edge image to obtain the edge detection response value of the target pixels, where the target pixels are any pixels in the guide rail defect edge image; finally, when the edge detection response value of the target pixels is greater than a set threshold, the target pixels are taken as defect edge points, thereby obtaining the guide rail defect edge.

[0006] For example, Chinese invention patent CN120275396A discloses a guide rail defect detection system and method based on image recognition, including: an image information acquisition module for acquiring image information of the guide rail surface; an image information processing module for preprocessing the acquired images; an image information transmission module for transmitting the image information processed by the image information processing module; a feature information extraction module for extracting features of the guide rail surface using a deep learning algorithm to identify potential defect areas; a defect identification and classification module for identifying defect information and classifying defects; and a detection result output module for outputting the detection results in image form for operator reference.

[0007] For example, Chinese invention patent CN105044122B discloses a visual inspection method for copper surface defects based on a semi-supervised learning model, which includes: using a conveyor belt and guide rail to move the copper part to four corresponding inspection stations, using an image acquisition system to take pictures to sequentially detect whether there are defects on the upper and lower surfaces and both sides of the copper part, and classifying the copper part according to the judgment results. The system is also equipped with a camera to remotely monitor the operation of the visual inspection system in real time.

[0008] The above-mentioned technology has at least the following technical problems:

[0009] In high-load-bearing circular guide rails, the chain rigidly drives the slide connecting plate through the connector, and the slide clamping plate holds the upper and lower surfaces of the guide rail. This rigid connection method makes the movement of the slide connecting plate and the chain closely synchronized. Under the high-speed operation of high-load-bearing circular guide rails, the polygonal effect of the chain becomes more significant. When the chain experiences periodic speed fluctuations due to the polygonal effect, the slide connecting plate cannot buffer this vibration through its own elastic deformation and can only be forced to vibrate along with the chain. Moreover, the vibration amplitude and frequency are highly consistent with the chain fluctuations, resulting in more intense and regular slide vibration. This makes the positional jitter of guide rail defects in the image sequence more obvious. During camera acquisition, the vibration of the slide causes the positional jitter of defects in the image sequence. For example, the defect edges that should be continuous are broken into discrete points, and the edge features may be misjudged as noise. This can cover or blur the originally clear defect edge features, thus interfering with the edge feature signals. This results in low accuracy of defect edge detection data for high-load-bearing circular guide rails. Summary of the Invention

[0010] To address the problem of low accuracy in edge detection data for defects in high-load-bearing annular guide rails in existing technologies, this invention provides a multi-angle defect detection method and system for annular guide rails. The technical solution is as follows:

[0011] On the one hand, a multi-angle defect detection method for ring guide rails is provided. This method includes: during defect edge detection of high-load-bearing ring guide rails, determining whether there is a need for ring guide rail image acquisition interference assessment based on the guide rail vibration stability interference assessment results; if a need is determined, determining whether to perform ring guide rail image acquisition optimization based on the ring guide rail image acquisition interference assessment results; otherwise, performing defect degree detection. Ring guide rail image acquisition optimization refers to suppressing the impact of jitter on the overall quality of the image sequence through guide rail image jitter optimization, and improving the clarity of defect edges through guide rail defect edge image blurring optimization; after the ring guide rail image acquisition interference assessment is qualified, determining whether to perform defect degree detection based on the defect edge detection accuracy assessment results.

[0012] On the other hand, a multi-angle defect detection system for ring guide rails is provided. This system applies a multi-angle defect detection method for ring guide rails, including: a guide rail vibration stability detection module, a ring guide rail image acquisition optimization and judgment module, and a defect edge detection accuracy evaluation module. Specifically, the guide rail vibration stability detection module determines whether a ring guide rail image acquisition interference assessment is needed during defect edge detection of high-load-bearing ring guide rails, based on the guide rail vibration stability interference assessment results. The ring guide rail image acquisition optimization and judgment module determines whether to optimize the ring guide rail image acquisition based on the ring guide rail image acquisition interference assessment results if the assessment is deemed necessary; otherwise, it performs defect severity detection. The defect edge detection accuracy evaluation module determines whether to perform defect severity detection after the ring guide rail image acquisition interference assessment is deemed satisfactory, based on the defect edge detection accuracy evaluation results.

[0013] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0014] 1. Based on the vibration stability interference assessment results of the guide rail, determine whether there is a need for interference assessment of the ring guide rail image acquisition. This helps to avoid invalid image processing and resource waste. If it is determined that there is a need, then determine whether to optimize the ring guide rail image acquisition based on the interference assessment results. Otherwise, perform defect severity detection to determine the severity of defect edges of high-load ring guide rails. This helps to quantify the impact of vibration on image quality to support the quantitative analysis of interference in ring guide rail image acquisition. After the ring guide rail image acquisition interference assessment is qualified, determine whether to perform defect severity detection based on the defect edge detection accuracy assessment results. This helps to ensure that the data entering the defect severity detection has high reliability, repeatability, and traceability, further improving the accuracy of defect edge detection data for high-load ring guide rails and solving the problem of low accuracy of defect edge detection data for high-load ring guide rails in the existing technology.

[0015] 2. By optimizing the blurring of the guide rail defect edge image when the blurring value is greater than the preset blurring value, it helps to meet the edge recognition requirements of high-load-bearing ring guide rail defects. By increasing and decreasing the ring light source tilt angle, more precise ring light source tilt angle adjustment can be achieved to reduce ineffective light flux. By increasing and decreasing the ring light source light intensity, more precise ring light intensity adjustment can be achieved, thereby achieving adaptive improvement in the detection accuracy of high-load-bearing ring guide rail defects.

[0016] 3. By combining preset defect edge detection accurate feedback parameters for weighted calculation, accurate defect edge detection data is obtained. Compared with the existing technology that evaluates the accuracy of defect edge detection based on a single parameter, this helps to avoid distortion caused by changes in chain tension or local deformation of the slide rail. By aggregating multiple data and then performing negative correlation quantification to obtain accurate defect edge detection results, it helps to accurately quantify the role of accurate defect edge detection parameters in the accuracy of defect edge detection of high-load-bearing ring guide rails, thereby improving the confidence of defect detection of high-load-bearing ring guide rails. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a multi-angle defect detection method for annular guide rails provided by an embodiment of the present invention;

[0019] Figure 2 This is a general overview diagram of a multi-angle defect detection method for annular guide rails provided in the embodiments of this application;

[0020] Figure 3 This is a schematic diagram of the blurring optimization of the guide rail defect edge image, which is a method for multi-angle defect detection of annular guide rails provided in an embodiment of this application.

[0021] Figure 4 This is a schematic diagram of a multi-angle defect detection system for a ring guide rail provided in an embodiment of the present invention;

[0022] Figure 5 This is a real-time defect detection interface provided in an embodiment of this application;

[0023] Figure 6 The second real-time defect detection interface provided in this application embodiment;

[0024] Figure 7 The third real-time defect detection interface provided in this application embodiment. Detailed Implementation

[0025] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0026] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0027] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0028] This invention provides a method for detecting multi-angle defects in annular guide rails. For example... Figure 1 The flowchart shown is for a multi-angle defect detection method for a ring guide rail. As one embodiment, the processing flow of this method may include the following steps:

[0029] First, guide rail vibration stability detection: During the defect edge detection of high-load-bearing ring guide rails, the need to conduct ring guide rail image acquisition interference assessment is determined based on the guide rail vibration stability interference assessment results. Guide rail vibration stability detection helps to identify the risk of image acquisition interference caused by mechanical resonance or dynamic load fluctuation in advance, and avoids the missed detection of edge feature drift or false defects caused by vibration.

[0030] Secondly, the image acquisition optimization of the circular guide rail is determined: if it is determined that there is a need, the image acquisition interference assessment results of the circular guide rail determine whether to perform image acquisition optimization. Otherwise, defect degree detection is performed to determine the severity of the defect edges of the high-load circular guide rail. Image acquisition optimization of the circular guide rail means optimizing the image jitter to ensure the relative stability of the defect edge features in the image, thereby suppressing the impact of jitter on the overall quality of the image sequence, and optimizing the image blur of the defect edge to allow the light to better cover the edge area of ​​the guide rail, reducing the edge differences caused by uneven illumination, and improving the clarity of the defect edges. The image acquisition optimization of the circular guide rail helps to specifically suppress the coupled influence of uneven illumination and motion blur on the clarity of the defect edges.

[0031] Finally, the accuracy assessment of defect edge detection: After the interference assessment of the circular guide rail image acquisition is qualified, the determination of whether to perform defect severity detection is based on the result of the accuracy assessment of defect edge detection. The accuracy assessment of defect edge detection helps to quantify edge fidelity and provide a benchmark for the classification of defect severity.

[0032] Before designing the multi-angle defect detection method for ring guide rail provided in this application, a database is established to store various setting data. The database includes, but is not limited to, the set defect edge continuity and integrity value, the preset qualified defect edge offset-jitter response value, and the preset qualified guide rail defect edge image blur degree value, etc. The various values ​​are directly set by technicians.

[0033] like Figure 2 The diagram shown is a general overview of a multi-angle defect detection method for annular guide rails provided in an embodiment of this application; Figure 2 It can be seen that: the vibration stability of the guide rail is detected by obtaining the interference results of the guide rail vibration stability, specifically including the meshing-high-speed operation interference response value and the ring guide rail vibration displacement-stability value. When the detection meets the guide rail vibration stability conditions, that is, the meshing-high-speed operation interference response value is not greater than the preset high-speed operation interference response value, and the ring guide rail vibration displacement-stability value is not greater than the preset high-load ring guide rail vibration stability value, the defect degree detection is performed; otherwise, the defect edge offset-jitter response value and the guide rail defect edge image blur degree value are obtained by ring guide rail image acquisition optimization judgment. When the detected defect edge offset-jitter response value is greater than the preset defect edge offset-jitter response value, the guide rail image jitter optimization is performed, otherwise the guide rail defect edge image blur degree evaluation is performed. When the detected guide rail defect edge image blur degree value is greater than the preset guide rail defect edge image blur degree value, the guide rail defect edge image blur degree optimization is performed; otherwise, the defect edge detection accuracy result is obtained by defect edge detection accuracy evaluation. When the detected defect edge detection accuracy result is greater than the preset defect edge detection accuracy value, the defect degree detection is performed, otherwise a ring guide rail detection failure prompt is sent.

[0034] In this embodiment, the combined effects of guide rail vibration stability detection, ring guide rail image acquisition optimization and judgment, and defect edge detection accuracy evaluation are interconnected, which helps to enhance the dynamic adaptability and reduce the false detection rate of ring guide rail defect edge detection under high load conditions, and helps to improve the accuracy of high load ring guide rail detection under complex working conditions; thus, it improves the accuracy of high load ring guide rail defect edge detection data.

[0035] Specifically, in the complex scenario of high-load double-track circular guide rails, by detecting the vibration stability of the guide rail, optimizing and judging the image acquisition of the circular guide rail, and evaluating the accuracy of defect edge detection, we can avoid the expansion of defects caused by vibration, such as accelerated cracking of fatigue cracks. This helps to support the safety and efficiency of high-load double-track circular guide rails in high-load, continuous operation.

[0036] Furthermore, based on the guide rail vibration stability interference assessment results, it is determined whether there is a need for image acquisition interference assessment of the ring guide rail. The specific process is as follows: Obtain the guide rail vibration stability interference results to reflect the vibration stability of the high-load ring guide rail during a preset vibration stability period. The guide rail vibration stability interference results include the meshing-high-speed operation interference response value and the ring guide rail vibration displacement-stability value. The number of chain and gear meshing times during the preset vibration stability period is monitored by a counter as the meshing-high-speed operation interference response value. The vibration displacement of the high-load ring guide rail at a preset position point during the preset vibration stability period is monitored by a displacement sensor, and the average value is used as the ring guide rail vibration displacement-stability value. Monitoring the meshing-high-speed operation interference response value helps to capture the frequency and intensity changes of chain-gear meshing impact in real time. Monitoring the ring guide rail vibration displacement-stability value helps to directly assess the guide rail structure under dynamic load. Overall rigidity attenuation and local loosening risk; the preset vibration stability time period represents the preset time period corresponding to the guide rail vibration stability interference assessment; judgment is made based on the guide rail vibration stability interference results: if the meshing-high-speed operation interference response value is not greater than the preset high-speed operation interference response value obtained from the database, and the ring guide rail vibration displacement-stability value is not greater than the preset high-load ring guide rail vibration stability value obtained from the database, image detection is performed and the corresponding high-load ring guide rail image is marked as a qualified multi-angle ring guide rail image, and the defect degree is detected based on the qualified multi-angle ring guide rail image; otherwise, a vibration stability failure prompt is sent and a ring guide rail image acquisition interference assessment is performed. Among them, the preset high-speed operation interference response value is represented by the average value of the meshing-high-speed operation interference response value in the historical time period, and the preset high-load ring guide rail vibration stability value is represented by the average value of the ring guide rail vibration displacement-stability value in the historical time period.

[0037] In this embodiment, the meshing-high-speed operation disturbance response value and the ring guide rail vibration displacement-stability value have a mutual influence and are mutually causal. The larger the meshing-high-speed operation disturbance response value, the greater the meshing impact, which may lead to displacement of the guide rail structure, and in turn may lead to an increase in the ring guide rail vibration displacement-stability value. By coordinating the meshing-high-speed operation disturbance response value and the ring guide rail vibration displacement-stability value for judgment and monitoring, a two-way verification mechanism is formed, thereby achieving the calibration and adaptive early warning of dynamic stability boundary under high-speed heavy-load complex working conditions.

[0038] Furthermore, the interference assessment for the circular guide rail image acquisition includes evaluating the jitter of the guide rail defect edges and the blurring degree of the guide rail defect edge images. The specific process for evaluating the jitter of the guide rail defect edges is as follows: The defect edge offset-jitter response value is obtained to reflect the positional jitter of the defect edge in the guide rail defect edge image sequence. The offset of the same defect edge feature point in adjacent defect edge image sequences within a preset guide rail image acquisition time period is monitored using a high-resolution industrial camera, and the average value is taken as the defect edge offset-jitter response value. The preset guide rail image acquisition time period represents the preset time period corresponding to the circular guide rail image acquisition interference assessment. The system is based on the defect edge offset-jitter response value. If the defect edge offset-jitter response value is greater than the preset defect edge offset-jitter response value obtained from the database, the guide rail image jitter is optimized. Otherwise, the blurring degree of the guide rail defect edge image is evaluated. The preset defect edge offset-jitter response value is represented by the average value of the defect edge offset-jitter response values ​​over a historical time period. When the detected defect edge offset-jitter response value is greater than the preset defect edge offset-jitter response value, the guide rail image jitter is optimized. This helps to actively suppress image blurring and edge misalignment caused by mechanical vibration or chain movement, and avoids misjudgment of defects.

[0039] Specifically, guide rail image jitter optimization includes setting the vibration image gain and setting the vibration image exposure time. By simultaneously setting the vibration image gain and exposure time, it helps to compress the exposure window while improving the signal-to-noise ratio.

[0040] Specifically, the vibration image gain setting involves progressively increasing the gain of the high-resolution industrial camera by a step size corresponding to the preset vibration image-camera gain mapping value. The preset vibration image-camera gain mapping value is obtained by mapping the defect edge offset-jitter response value and the guide rail defect edge image contrast into a camera gain mapping set in the database. The average contrast of the guide rail defect edge image at a preset angle within the preset guide rail image acquisition time period is monitored by an imaging photometer and used as the guide rail defect edge image contrast. The camera gain mapping set is pre-configured by the preset personnel and is used to reflect the mapping relationship between the defect edge offset-jitter response value and the guide rail defect edge image contrast and the corresponding preset vibration image-camera gain mapping value. When a vibration image gain setting prompt is received, the gain of the high-resolution industrial camera is progressively increased by a step size corresponding to the preset vibration image-camera gain mapping value. This helps to gradually compensate for the photon flux lost due to shortened exposure and maintain edge contrast.

[0041] Furthermore, the vibration image exposure time setting is specifically as follows: the exposure time of the high-resolution industrial camera is gradually reduced by a step size corresponding to the preset vibration image exposure time mapping value. The preset vibration image exposure time mapping value is obtained by mapping the defect edge offset-jitter response value and the guide chain movement speed into the camera exposure time mapping set in the database. The camera exposure time mapping set is pre-configured by the preset personnel and is used to reflect the mapping relationship between the defect edge offset-jitter response value and the guide chain movement speed and the corresponding preset vibration image exposure time mapping value. The average speed of the guide chain at the preset time point and preset position point is monitored by a laser velocimeter as the guide chain movement speed. When a vibration image exposure time setting prompt is received, the exposure time of the high-resolution industrial camera is gradually reduced by a step size corresponding to the preset vibration image exposure time mapping value, which helps to control motion blur and ensure the spatial consistency of edge geometric features.

[0042] In this embodiment, by evaluating the jitter of the guide rail defect edge, it is helpful to quantify the positional drift of the defect edge in the image sequence, providing a precise trigger basis for subsequent guide rail image jitter optimization. Through the mutual support and interaction between the guide rail defect edge jitter evaluation and the guide rail image jitter optimization, it is helpful to achieve high stability and high fidelity acquisition of guide rail defect edge images under high load conditions, ultimately improving the defect recognition accuracy.

[0043] Furthermore, the specific process for evaluating the blurriness of the guide rail defect edge image is as follows: The blurriness value of the guide rail defect edge image is obtained to reflect the blurriness of the defect edge in the image. This blurriness value is represented by the difference between the average width of the defect edge at the preset angle at the end of the preset guide rail image acquisition time period and the average width of the defect edge at the preset angle at the beginning of the time period. Based on the blurriness value, a judgment is made: if the blurriness value is greater than the preset blurriness value obtained from the database, blurriness optimization is performed; otherwise, the accuracy of defect edge detection is evaluated. The preset blurriness value is represented by the average blurriness value of guide rail defect edge images over historical time periods. When the detected blurriness value is greater than the preset blurriness value, blurriness optimization is performed, which helps to actively compensate for edge degradation caused by tilt mismatch or uneven light intensity, and prevents the geometric features of the defect from being submerged by noise.

[0044] like Figure 3 The image shown is a schematic diagram of guide rail defect edge image blurring optimization in a multi-angle defect detection method for annular guide rails provided in an embodiment of this application; by Figure 3It can be seen that when the image blur optimization condition is met, i.e., the blur degree value of the guide rail defect edge image is greater than the preset blur degree value, the guide rail defect edge image blur optimization is performed. The guide rail defect edge image blur optimization means that the ring light source tilt angle and ring light source intensity are set simultaneously. The ring light source tilt angle setting is as follows: when the initial ring light source tilt angle is not greater than the preset average ring light source tilt angle, the ring light source tilt angle is increased; otherwise, the ring light source tilt angle is decreased. The ring light source intensity setting is as follows: when the initial ring light intensity is not greater than the preset average ring light intensity, the ring light intensity is increased; otherwise, the ring light intensity is decreased.

[0045] Specifically, the optimization of the blurring of the guide rail defect edge image includes setting the ring light source tilt angle and setting the ring light source intensity. The ring light source tilt angle setting includes setting to increase the ring light source tilt angle when the initial ring light source tilt angle is not greater than the preset average ring light source tilt angle, and setting to decrease the ring light source tilt angle when the initial ring light source tilt angle is greater than the preset average ring light source tilt angle. By judging the relationship that the initial ring light source tilt angle is not greater than the preset average ring light source tilt angle, it is decided whether to increase or decrease the ring light source tilt angle, which helps to prevent excessive unidirectional adjustment and the introduction of secondary shadows.

[0046] The setting to increase the tilt angle of the ring light source means that the tilt angle of the ring light source is gradually increased within the preset range of increasing the tilt angle of the ring light source by step size corresponding to the magnitude of the preset tilt angle mapping value. By using the magnitude of the preset tilt angle mapping value as the step size to achieve step-by-step adjustment, it helps to reduce the overflow of invalid light flux and maintain the consistency of edge sharpness.

[0047] Furthermore, the ring light source tilt angle reduction setting indicates that the ring light source tilt angle is gradually reduced within a preset range by a step size corresponding to the magnitude of the preset ring light source tilt angle mapping value. By gradually reducing the ring light source tilt angle, the light is incident at a gentler angle, reducing direct reflection, making the illumination distribution more uniform, and making the defect features clearer. The preset ring light source tilt angle mapping value is obtained by mapping the blur degree value of the guide rail defect edge image and the light spot tilt rate into a ring light source tilt angle mapping set in the database. The ring light source tilt angle mapping set is pre-configured by a preset team and is used to reflect the blur degree value of the guide rail defect edge image and the light spot tilt rate. The slope, and its mapping relationship with the corresponding preset ring light source tilt angle, is determined by monitoring the offset of the ring light source spot center position from the preset position point on the high-load ring guide rail and the preset spot position point set by the preset personnel during the preset guide rail image acquisition time period using a spot analyzer. The ratio of this offset to the spot propagation distance monitored by the laser rangefinder is used as the spot tilt rate. The ring light source tilt angle is within the preset ring light source tilt angle range. The preset ring light source tilt angle increase range and preset ring light source tilt angle decrease range are preset by the preset personnel and respectively include the endpoints of the upper and lower limits of these two ranges. The preset ring light source tilt angle average value is preset by the preset personnel.

[0048] It should be added that the ring light source intensity setting includes an increase setting when the initial ring light source intensity is not greater than the preset average ring light source intensity, and a decrease setting when the initial ring light source intensity is greater than the preset average ring light source intensity. By judging the relationship between the initial ring light source intensity and the preset average ring light source intensity, the system decides whether to increase or decrease the ring light source intensity, which helps to balance the signal-to-noise ratio and overexposure risk under constant exposure constraints. The increase setting indicates that the ring light source intensity is increased stepwise within the preset ring light source intensity increase range by a step size corresponding to the magnitude of the preset ring light source intensity mapping value. By using the magnitude of the preset ring light source intensity mapping value as the step size for stepwise adjustment, it helps to avoid image flicker and feature drift caused by step dimming. The decrease setting indicates that the ring light source intensity is increased stepwise within the preset ring light source intensity increase range by a step size corresponding to the magnitude of the preset ring light source intensity mapping value. Within the range of decreasing ring light intensity, the ring light intensity is gradually reduced with a step size corresponding to the magnitude of a preset ring light intensity mapping value. The preset ring light intensity mapping value is obtained by mapping the blur value of the guide rail defect edge image and the total light field intensity into a ring light intensity mapping set in the database. The ring light intensity mapping set is pre-configured by preset personnel and reflects the mapping relationship between the blur value of the guide rail defect edge image and the total light field intensity and the corresponding preset ring light intensity mapping value. The total light power of the light source in the preset detection area during the preset guide rail image acquisition time period is monitored by an optical power meter and used as the total light field intensity. Within the preset ring light intensity range, the preset ring light intensity increase range and preset ring light intensity decrease range are pre-set by preset personnel and respectively include the endpoints of the upper and lower limits of the range. The preset ring light intensity average value is also pre-set by preset personnel.

[0049] In this embodiment, by evaluating the blurring degree of the guide rail defect edge image, it is helpful to quantify the degree of edge dispersion caused by vibration, and provide a traceable monitoring indicator for the blurring optimization of the guide rail defect edge image. Through the interrelation and interaction between the blurring degree evaluation and the blurring optimization of the guide rail defect edge image, it is helpful to achieve higher clarity and higher consistency of the defect edge image under complex lighting and high dynamic load conditions, and ultimately improve the defect recognition accuracy and reduce the false detection rate.

[0050] Furthermore, the interference assessment for the circular guide rail image acquisition also includes a qualification judgment for the interference optimization of the circular guide rail image acquisition. This qualification judgment includes guide rail image jitter optimization judgment and guide rail defect edge image blurring optimization judgment. The specific process for guide rail image jitter optimization judgment is as follows: After optimizing the guide rail image jitter, the defect edge offset-jitter response value is re-acquired for the next adjacent preset guide rail image acquisition time period. If the defect edge offset-jitter response value is not greater than the preset defect edge offset-jitter response value, the blurring degree of the guide rail defect edge image is evaluated; otherwise, a guide rail image jitter alarm is sent. The specific process for guiding rail defect edge image blurring optimization judgment is as follows: After optimizing the guide rail defect edge image blurring, the blurring degree value of the guide rail defect edge image is re-acquired for the next adjacent preset guide rail image acquisition time period. If the blurring degree value of the guide rail defect edge image is not greater than the preset blurring degree value, the defect edge detection accuracy is evaluated; otherwise, a guide rail defect edge image blurring alarm is sent.

[0051] In this embodiment, after optimizing the guide rail image jitter and blurring the guide rail defect edge image, the acceptance of the circular guide rail image acquisition interference optimization is judged. This helps to achieve dynamic monitoring of image acquisition quality using real-time feedback data from the next adjacent time period. By obtaining the defect edge offset-jitter response value for the next adjacent preset guide rail image acquisition time period after optimizing the guide rail image jitter, it helps to quantify the actual effectiveness of vibration reduction, gain, and exposure adjustment in real time, ensuring that mechanical vibration suppression is in place and the spatial consistency of the image sequence meets the preset requirements. By obtaining the blurring degree value of the guide rail defect edge image for the next adjacent preset guide rail image acquisition time period after optimizing the guide rail defect edge image, it helps to ensure the clear measurability of defect geometric features.

[0052] Furthermore, the specific process for evaluating the accuracy of defect edge detection is as follows: First, the result of the convergence quantization of the accurate defect edge detection parameters and the preset accurate defect edge detection parameters is obtained. Convergence quantization means performing a ratio calculation, and then combining the preset accurate defect edge detection feedback parameters for weighted calculation to obtain the accurate defect edge detection data.

[0053] Specifically, the expression for edge continuity - edge detection accuracy is as follows: E = 1, 2, ..., P, where E represents the number of the preset edge detection time period, P represents the total number of preset edge detection time periods, S1(E) represents the edge continuity-edge detection accuracy value of the Eth preset edge detection time period, A(E) represents the defect edge continuity integrity value of the Eth preset edge detection time period, A(0) represents the preset defect edge continuity integrity value, and S1 represents the edge continuity-feedback value. The number of complete defect edges detected by the preset angle is monitored by a high-resolution industrial camera, and the ratio of its average value to the total number of actual defect edges is taken as the defect edge continuity integrity value. Both the defect edge continuity integrity value and the preset defect edge continuity integrity value have no unit.

[0054] Specifically, the expression for edge jitter - edge detection accuracy is: S2(E) represents the edge jitter-edge detection accuracy value for the Eth preset edge detection time period, B(E) represents the qualified defect edge offset-jitter response value for the Eth preset edge detection time period, B(0) represents the preset qualified defect edge offset-jitter response value, S2 represents the edge jitter-feedback value, and the qualified defect edge offset-jitter response value is represented by the qualified defect edge offset-jitter response value that is not greater than the preset defect edge offset-jitter response value.

[0055] Specifically, the expression for edge image blurring - edge detection accuracy value is as follows: S3(E) represents the edge image blur-edge detection accuracy value during the Eth preset edge detection time period, C(E) represents the blur degree value of the qualified guide rail defect edge image during the Eth preset edge detection time period, C(0) represents the preset qualified guide rail defect edge image blur degree value, S3 represents the edge image blur-feedback value, and the qualified guide rail defect edge image blur degree value is represented by the guide rail defect edge image blur degree value that is not greater than the preset guide rail defect edge image blur degree value.

[0056] Then, after multi-data aggregation processing of the accurate defect edge detection data, negative correlation quantification is performed to obtain the accurate defect edge detection results, which reflects the role of the accurate defect edge detection parameters in the accuracy of defect edge detection of high-load-bearing ring rails.

[0057] It should be explained that accurate defect edge detection data includes edge continuity-accurate edge detection value, edge jitter-accurate edge detection value, and edge image blurring-accurate edge detection value. By quantitatively analyzing the accurate defect edge detection data, the impact of these data on the accuracy of defect edge detection in high-load-bearing circular guide rails can be specifically measured, thus yielding precise defect edge detection results. A larger accurate defect edge detection data indicates a greater deviation between the accurate defect edge detection parameters and the preset accurate defect edge detection parameters. This results in a stronger influence of the accurate defect edge detection parameters on the accuracy of defect edge detection in high-load-bearing circular guide rails, thereby reducing the accuracy of the defect edge detection result. It should be noted that in this embodiment, there is a negative correlation between the accurate defect edge detection data and the accurate defect edge detection result.

[0058] The accurate results of defect edge detection are obtained through the following methods:

[0059] ;

[0060] In the formula, S(E) represents the accurate result of defect edge detection during the Eth preset edge detection time period.

[0061] Finally, a judgment is made based on the accurate result of defect edge detection. If the accurate result of defect edge detection is greater than the preset accurate value of defect edge detection obtained from the database, a qualified circular guide rail detection prompt is sent, and the corresponding guide rail defect edge image is marked as a qualified multi-angle circular guide rail image for defect severity detection. Defect severity detection means that the qualified multi-angle circular guide rail image is input into a machine learning model, such as a multi-branch convolutional neural network model, for training to obtain the training multi-branch convolutional neural network model. Then, the newly obtained qualified multi-angle circular guide rail image is input into the training multi-branch convolutional neural network model to output the defect severity result, such as: normal, slight, moderate, severe. Otherwise, a qualified circular guide rail detection prompt is sent. The preset personnel divide the qualified multi-angle circular guide rail images into training qualified multi-angle circular guide rail images and verification qualified multi-angle circular guide rail images.

[0062] It should be explained that accurate parameters for defect edge detection include the defect edge continuity and integrity value, the qualified defect edge offset-jitter response value, and the qualified guide rail defect edge image blur value. A larger qualified defect edge offset-jitter response value means that the defect edge exhibits more significant offset and jitter in the image, causing what should be a continuous and complete defect edge to become broken or discontinuous due to jitter and offset, thus reducing the defect edge continuity and integrity value. A larger qualified guide rail defect edge image blur value means that the defect edge in the image becomes unclear and blurry, making it more difficult to detect the continuity and integrity of the defect edge. Some defect edges that should be identified as continuous may be misjudged as broken edges, thus affecting the accuracy of the defect edge continuity and integrity value and reducing it. A larger qualified defect edge offset-jitter response value means that jitter may cause motion blur in high-load, ring-shaped images captured by high-resolution cameras, resulting in both offset and jitter issues with the defect edge, making it blurry and thus increasing the qualified guide rail defect edge image blur value.

[0063] The preset defect edge detection accuracy parameters include the preset defect edge continuity and integrity value, the preset qualified defect edge offset-jitter response value, and the preset qualified guide rail defect edge image blur degree value. The preset defect edge detection accuracy parameters are represented by the average value of the defect edge detection accuracy parameters over a historical time period. The preset edge detection time period represents the preset time period corresponding to the defect edge detection accuracy evaluation.

[0064] In the embodiments of this application, a set of mapping groups retrieved from a database is provided. These mapping groups are pre-configured by designated personnel and contain several mapping sets. The mapping relationships defined in the mapping groups are flexible, and can be either a one-to-one correspondence between single parameters or a many-to-one relationship where multiple parameters correspond to one parameter. Specifically, a one-to-one or many-to-one mapping association can be established between accurate defect edge detection parameters and preset accurate defect edge detection feedback parameters. By inputting the real-time acquired accurate defect edge detection parameters into the corresponding mapping groups, the corresponding preset accurate defect edge detection feedback parameters are output according to the pre-set mapping relationships. The value range of the preset accurate defect edge detection feedback parameters is limited to the interval between 0 and 1. The preset accurate defect edge detection feedback parameters are determined based on the proportion of the corresponding accurate defect edge detection parameters in the overall data. The preset accurate defect edge detection feedback parameters include edge continuity feedback value, edge jitter feedback value, and edge image blur feedback value, which are used to measure the degree of influence of the accurate defect edge detection parameters on the accurate defect edge detection data.

[0065] As another implementation method, such as Figure 4The diagram shown is a structural schematic of a multi-angle defect detection system for a ring guide rail provided in an embodiment of this application. This application provides a multi-angle defect detection system for ring guide rails, employing a multi-angle defect detection method for ring guide rails. The system is characterized by comprising: a guide rail vibration stability detection module, a ring guide rail image acquisition optimization judgment module, and a defect edge detection accuracy evaluation module. Specifically, the guide rail vibration stability detection module determines whether a ring guide rail image acquisition interference assessment is needed during defect edge detection of high-load-bearing ring guide rails, based on the guide rail vibration stability interference assessment results. The ring guide rail image acquisition optimization judgment module determines whether to perform ring guide rail image acquisition optimization if the assessment indicates a need, otherwise, it performs defect severity detection to determine the severity of the defect edges of the high-load-bearing ring guide rail. The defect edge detection accuracy evaluation module determines whether to perform defect severity detection after the ring guide rail image acquisition interference assessment is deemed satisfactory, based on the defect edge detection accuracy evaluation results.

[0066] In this embodiment, the various defect edge detection accuracy parameters used to evaluate the accuracy of defect edge detection in heavy-duty ring guides are not isolated from each other, but are interconnected and mutually influential. Only by conducting a correlation analysis of these parameters can their combined effect on the accuracy of defect edge detection in heavy-duty ring guides be fully and accurately described. By deeply analyzing the comprehensive influence between parameters, a precise evaluation of the accuracy of defect edge detection in heavy-duty ring guides can be achieved.

[0067] like Figure 5 As shown, this is a real-time defect detection interface provided in an embodiment of this application. Figure 6 As shown, this is the second real-time defect detection interface provided in this application embodiment. Figure 7 The image shows the third real-time defect detection interface provided in this application embodiment. This application is used to implement the real-time defect detection function in the quality inspection module of the guide rail quality inspection management system. The guide rail quality inspection management system also includes a homepage module, an equipment management module, a data analysis module, and a system settings module. The homepage module includes a system overview, real-time monitoring, and performance trends. The quality inspection module includes real-time defect detection, component quality inspection, batch inspection management, and inspection result analysis. The equipment management module includes equipment status monitoring, equipment management, equipment maintenance records, and equipment configuration management. The data analysis module includes historical data analysis, trend analysis, and predictive analysis.

[0068] Depend on Figure 5 As can be seen, the current circular guide rail type is a high-load double-line guide rail. The current guide rail image detection progress is 50%. The image sequence abnormality indicator and image blur indicator are red, while the sequence optimization indicator and blur setting indicator are green, indicating that optimization and adjustment are needed. You can click [here]. Figure 6 The light source and camera settings buttons are adjusted via the back end. An illuminated optimization indicator light signifies successful optimization, and the guide rail image detection progress is 60%. Clicking on the defect image will display an image of the annular guide rail crack defect, such as... Figure 7 The image shown is an image of a crack defect in a ring-shaped guide rail. The crack features are displayed in the image's annotation box.

[0069] In summary, this application's embodiments determine whether there is a need for annular guide rail image acquisition interference assessment based on the guide rail vibration stability interference assessment results. This helps avoid ineffective image processing and resource waste. If it is determined that there is a need, then the image acquisition of the annular guide rail is optimized based on the annular guide rail image acquisition interference assessment results. Otherwise, defect severity detection is performed to determine the severity of defect edges in high-load-bearing annular guide rails. This helps quantify the impact of vibration on image quality to support quantitative analysis of annular guide rail image acquisition interference. After the annular guide rail image acquisition interference assessment is qualified, the defect severity detection is determined based on the defect edge detection accuracy assessment results. This helps ensure that the data entering the defect severity detection has high reliability, is repeatable and traceable, and further improves the accuracy of defect edge detection data for high-load-bearing annular guide rails, solving the problem of low accuracy of defect edge detection data for high-load-bearing annular guide rails in the prior art.

[0070] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0071] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0072] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0073] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0074] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0075] In the several embodiments provided by this invention, it should be understood that the disclosed methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0076] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0077] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0078] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0079] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for detecting multi-angle defects in annular guide rails, characterized in that, The method includes: During the defect edge detection of high-load-bearing ring rails, the need for ring rail image acquisition interference assessment is determined based on the rail vibration stability interference assessment results. If it is determined that there is a need, then the decision on whether to optimize the image acquisition of the circular guide rail is based on the interference evaluation result of the image acquisition of the circular guide rail; otherwise, the defect degree detection is performed. The optimization of the image acquisition of the circular guide rail means to suppress the impact of jitter on the overall quality of the image sequence by optimizing the image jitter of the guide rail and to improve the clarity of the defect edge by optimizing the blurring of the guide rail defect edge image. After the interference assessment of the circular guide rail image acquisition is qualified, the determination of whether to perform defect degree detection is based on the assessment result of the accuracy of defect edge detection. The process for determining whether an image acquisition interference assessment of the circular guide rail is needed based on the guide rail vibration stability interference assessment results is as follows: The vibration stability disturbance results of the guide rail are obtained to reflect the vibration stability of the high-load-bearing ring guide rail during the preset vibration stability time period. The vibration stability interference results of the guide rail include the meshing-high-speed operation interference response value and the ring guide rail vibration displacement-stability value. The meshing-high-speed operation interference response value is represented by the number of meshing times of the chain and gear during a preset vibration stability time period. The ring guide rail vibration displacement-stability value is represented by the average vibration displacement of a preset position point of a high-load ring guide rail during a preset vibration stability time period. If the meshing-high-speed operation interference response value is not greater than the preset high-speed operation interference response value obtained from the database, and the ring rail vibration displacement-stability value is not greater than the preset high-load ring rail vibration stability value obtained from the database, image detection is performed and the corresponding high-load ring rail image is marked as a qualified multi-angle ring rail image. Defect degree detection is performed based on the qualified multi-angle ring rail image. Otherwise, a vibration stability failure prompt is sent and an ring rail image acquisition interference assessment is performed. The image acquisition interference assessment of the circular guide rail includes the assessment of guide rail defect edge jitter and the assessment of guide rail defect edge image blur. If the accurate result of the defect edge detection is greater than the preset accurate value of the defect edge detection obtained from the database, a qualified notification for the ring rail detection is sent and the corresponding guide rail defect edge image is marked as a qualified multi-angle ring rail image for defect degree detection; otherwise, a qualified notification for the ring rail detection is sent.

2. The method for multi-angle defect detection of annular guide rails according to claim 1, characterized in that, The specific process for evaluating the edge jitter of the guide rail defect is as follows: The defect edge offset-jitter response value is obtained to reflect the position jitter of the defect edge in the guide rail defect edge image sequence. The defect edge offset-jitter response value is represented by the average offset of the same defect edge feature point in adjacent defect edge image sequences within a preset guide rail image acquisition time period. If the defect edge offset-jitter response value is greater than the preset defect edge offset-jitter response value obtained from the database, guide rail image jitter optimization is performed; otherwise, guide rail defect edge image blur degree evaluation is performed.

3. The method for multi-angle defect detection of annular guide rails according to claim 2, characterized in that, The guide rail image jitter optimization includes setting the vibration image gain and setting the vibration image exposure time. The specific method for setting the vibration image gain is as follows: the gain of the high-resolution industrial camera is gradually increased by stepping the amplitude corresponding to the preset vibration image-camera gain mapping value. The preset vibration image-camera gain mapping value is obtained by mapping the defect edge offset-jitter response value and the guide rail defect edge image contrast into the camera gain mapping set in the database. The specific method for setting the vibration image exposure time is as follows: the exposure time of the high-resolution industrial camera is gradually reduced by a step size corresponding to the amplitude of the preset vibration image exposure time mapping value. The preset vibration image exposure time mapping value is obtained by mapping the defect edge offset-jitter response value and the guide rail chain movement speed into the camera exposure time mapping set in the database.

4. The method for multi-angle defect detection of annular guide rails according to claim 2, characterized in that, The specific process for evaluating the blurriness of the guide rail defect edge image is as follows: The blur degree value of the guide rail defect edge image is obtained to reflect the blur degree of the defect edge in the guide rail defect edge image. The blur degree value of the guide rail defect edge image is represented by the difference between the average value of the defect edge width at the preset angle at the end of the preset guide rail image acquisition time period and the average value of the defect edge width at the preset angle at the beginning of the time period. If the blur value of the guide rail defect edge image is greater than the preset blur value of the guide rail defect edge image obtained from the database, the blur of the guide rail defect edge image is optimized; otherwise, the accuracy of defect edge detection is evaluated.

5. The method for multi-angle defect detection of annular guide rails according to claim 4, characterized in that, The optimization of the edge blurring image of the guide rail defect includes setting the tilt angle of the ring light source and setting the light intensity of the ring light source; The ring light source tilt angle setting includes a setting to increase the ring light source tilt angle when the initial ring light source tilt angle is not greater than the preset average ring light source tilt angle, and a setting to decrease the ring light source tilt angle when the initial ring light source tilt angle is greater than the preset average ring light source tilt angle. The setting of increasing the tilt angle of the ring light source means that the tilt angle of the ring light source is gradually increased within the preset range of increasing the tilt angle of the ring light source by a step size corresponding to the magnitude of the preset tilt angle mapping value of the ring light source. The ring light source tilt angle reduction setting means that the ring light source tilt angle is gradually reduced within a preset ring light source tilt angle reduction range by a step size corresponding to the magnitude of the preset ring light source tilt angle mapping value. The preset ring light source tilt angle mapping value is obtained by mapping the blur degree value of the guide rail defect edge image and the light spot tilt rate into the ring light source tilt angle mapping set in the database.

6. The method for multi-angle defect detection of annular guide rails according to claim 5, characterized in that, The ring light source intensity setting includes an increase setting for the ring light source intensity when the initial ring light source intensity is not greater than the preset average ring light source intensity, and a decrease setting for the ring light source intensity when the initial ring light source intensity is greater than the preset average ring light source intensity. The ring light source light intensity increase setting means that the ring light source light intensity is gradually increased within a preset ring light source light intensity increase range by a step size corresponding to the amplitude of the preset ring light source light intensity mapping value. The ring light source light intensity reduction setting means that the ring light source light intensity is reduced step by step within a preset ring light source light intensity reduction range with a step size corresponding to the amplitude of the preset ring light source light intensity mapping value. The preset ring light source intensity mapping value is obtained by mapping the blur degree value of the guide rail defect edge image and the total light field intensity into the ring light source intensity mapping set in the database.

7. The method for multi-angle defect detection of annular guide rails according to claim 1, characterized in that, The image acquisition interference assessment of the circular guide rail also includes determining the pass rate of the image acquisition interference optimization of the circular guide rail. The qualification judgment for interference optimization of the circular guide rail image acquisition includes the guide rail image jitter optimization judgment and the guide rail defect edge image blur optimization judgment. The specific process of the guide rail image jitter optimization judgment is as follows: After optimizing the guide rail image jitter, the defect edge offset-jitter response value of the next adjacent preset guide rail image acquisition time period is re-acquired. If the defect edge offset-jitter response value is not greater than the preset defect edge offset-jitter response value, the blur degree of the guide rail defect edge image is evaluated; otherwise, a guide rail image jitter alarm is sent. The specific process for determining the blurring of the guide rail defect edge image is as follows: After performing blurring optimization on the guide rail defect edge image, the blurring degree value of the guide rail defect edge image is re-acquired in the next adjacent preset guide rail image acquisition time period. If the blurring degree value of the guide rail defect edge image is not greater than the preset blurring degree value of the guide rail defect edge image, the accuracy of defect edge detection is evaluated; otherwise, a blurring alarm prompt for the guide rail defect edge image is sent.

8. The method for multi-angle defect detection of annular guide rails according to claim 7, characterized in that, The specific process for evaluating the accuracy of defect edge detection is as follows: The accurate defect edge detection data is obtained by weighting the results of approximation quantization of the accurate defect edge detection parameters and the preset accurate defect edge detection parameters, combined with the preset accurate defect edge detection feedback parameters. After performing multi-data aggregation processing on the accurate defect edge detection data, negative correlation quantification is then performed to obtain the accurate defect edge detection results, which reflects the role of the accurate defect edge detection parameters in the accuracy of defect edge detection for high-load-bearing ring guide rails. The accurate parameters for defect edge detection include the continuous integrity value of the defect edge, the offset-jitter response value of the qualified defect edge, and the blurring degree value of the image of the qualified guide rail defect edge. The accurate data for defect edge detection includes edge continuity - accurate edge detection value, edge jitter - accurate edge detection value, and edge image blur - accurate edge detection value; The preset defect edge detection accuracy feedback parameter is used to measure the degree of influence of the defect edge detection accuracy parameter on the defect edge detection accuracy data.

9. A multi-angle defect detection system for ring guide rails, employing the multi-angle defect detection method for ring guide rails as described in any one of claims 1-8, characterized in that, include: The module includes a guide rail vibration stability detection module, a ring guide rail image acquisition optimization and judgment module, and a defect edge detection accuracy evaluation module. The guide rail vibration stability detection module is used to determine whether there is a need for image acquisition interference assessment of the ring guide rail during the defect edge detection process of high load-bearing ring guide rails, based on the guide rail vibration stability interference assessment results. The circular guide rail image acquisition optimization judgment module is used to determine whether to perform circular guide rail image acquisition optimization based on the circular guide rail image acquisition interference evaluation result if it is determined that there is a need; otherwise, it performs defect degree detection. The defect edge detection accuracy evaluation module is used to determine whether to perform defect degree detection based on the defect edge detection accuracy evaluation result after the image acquisition interference evaluation of the circular guide rail is qualified.

Images