Intelligent detection system for stainless steel pipe production based on machine vision

Abstract

The invention relates to the technical field of machine vision, in particular to a stainless steel pipe production intelligent detection system based on machine vision, which comprises a shape evaluation module, a welding quality detection module, a welding quality analysis module, a pipe surface evaluation module, a cutting evaluation module, a process evaluation module and a management database, the system obtains the shape conformity degree of the stainless steel pipe according to the shape parameters of the stainless steel pipe, obtains the weld quality evaluation coefficient of the stainless steel pipe according to the weld defect parameters of the stainless steel pipe, and obtains the surface smoothness degree of the stainless steel pipe according to the surface parameters of the stainless steel pipe. The cutting evaluation coefficient of the stainless steel pipe is obtained according to the appearance parameters of the cutting face of the stainless steel pipe, then the procedure evaluation index of the stainless steel pipe is obtained through comprehensive analysis, the procedure evaluation index is fed back, unqualified products can be reduced, the yield is increased, and it is guaranteed that the performance of the produced pipe is consistent in all aspects.

Description

Technical Field

[0001] The invention relates to the technical field of machine vision, and in particular to an intelligent detection system for stainless steel pipe production based on machine vision. Background Art

[0002] With the continuous development of modern industry, stainless steel pipes have been widely used in many fields, such as construction, machinery, chemical industry, etc. The quality requirements for stainless steel pipes are also increasing. Traditional testing methods are gradually unable to meet production needs in terms of efficiency, accuracy and reliability.

[0003] In this context, machine vision technology came into being and developed rapidly. Using advanced imaging equipment and intelligent algorithms, machine vision can quickly and accurately acquire and analyze the image information of objects, providing a new way for the inspection of stainless steel pipes.

[0004] For example, the existing Chinese patent with application number 202310821644.6 discloses a method for detecting leak defects in stainless steel pipes. This scheme uses a robot to move the pipe from the storage table and place it in the pipe slot of the detection pool. The pipe placed in the slot and connected to the inflation interface at both ends is inflated, and the water surface image is captured by the camera array above the detection pool. The water surface image is processed to find the bubble image to determine whether there is a leak. If there is a leak, it is a defective pipe. The defective pipe is taken out of the detection pool and placed on the coding table, and the inkjet printer is used to mark the code, which helps to ensure the quality of the stainless steel pipe and screen out defective products.

[0005] However, there are the following shortcomings: First, this solution may have certain limitations in judging leak points through bubble images, and ignores the detection of defect parameters such as scratches. Surface defects such as scratches also have an important impact on the quality and appearance of stainless steel pipes. Some pipes with obvious scratches may be ignored, thus affecting the overall quality of the product.

[0006] Second, this solution does not involve the surface characteristics of the object being tested, such as smoothness, which may affect the detection method. The rough surface may interfere with the sealing effect of the pipe to a certain extent, affecting the accuracy of the detection, which may lead to misjudgment or missed judgment. Summary of the invention

[0007] In order to overcome the shortcomings of the background technology, an embodiment of the present invention provides a stainless steel pipe production intelligent detection system based on machine vision, which can effectively solve the problems involved in the above-mentioned background technology.

[0008] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides a stainless steel pipe production intelligent detection system based on machine vision, including: a shape evaluation module, which is used to detect the shape parameters of the stainless steel pipe, and then analyze the shape compliance degree δ of the stainless steel pipe, and the shape parameters include dimensional accuracy, roundness, and straightness.

[0009] The welding quality detection module is used to detect the weld defect parameters of stainless steel pipes. The weld defect parameters include the degree of non-welding and the degree of defects.

[0010] The welding quality analysis module is used to obtain the weld quality evaluation coefficient λ of the stainless steel pipe according to the weld defect parameter analysis of the stainless steel pipe.

[0011] The pipe surface evaluation module is used to detect the surface parameters of stainless steel pipes and then analyze the surface smoothness of stainless steel pipes. Surface parameters include surface roughness and scratch degree.

[0012] The cutting evaluation module is used to detect the appearance parameters of the cut surface of the stainless steel pipe, and then analyze and obtain the cutting evaluation coefficient ω of the stainless steel pipe. The appearance parameters of the cut surface of the stainless steel pipe include the cutting size accuracy. Cutting surface flatness e′.

[0013] The process evaluation module is used to analyze and obtain the process evaluation index of the stainless steel pipe according to the shape conformity, weld quality evaluation coefficient, surface smoothness and cutting evaluation coefficient of the stainless steel pipe, and provide feedback thereon.

[0014] Management database for storing standard diameters and standard cutting models of stainless steel pipes.

[0015] Preferably, the specific analysis method of the shape parameters of the stainless steel pipe is as follows: the first step is to select a number of measuring points on the stainless steel pipe according to the set interval, and to read the pipe diameter size at each measuring point of the stainless steel pipe by clamping a vernier caliper on the outer circumference of each measuring point of the stainless steel pipe, which is recorded as r i , i represents the number of the ,th measurement point, i=1, 2, ..., n, by substituting it into the formula The diameter accuracy of the stainless steel pipe is obtained, r0 represents the standard diameter of the stainless steel pipe, n represents the number of measuring points, and the wall thickness of each measuring point of the stainless steel pipe is obtained by placing the wall thickness micrometer perpendicular to the surface of each measuring point of the stainless steel pipe, which is recorded as d i , substituting it into the formula The wall thickness accuracy β of the stainless steel pipe is obtained, d0 represents the set standard wall thickness of the stainless steel pipe, and the dimensional accuracy γ of the stainless steel pipe is obtained by the formula γ=α*φ1+β*φ2, φ1 and φ2 represent the weight factors of the set diameter accuracy and wall thickness accuracy of the stainless steel pipe respectively.

[0016] In the second step, the cross section at either end of the stainless steel pipe is taken as the initial cross section, and a new cross section is selected at a set distance along the length direction of the stainless steel pipe. The selected cross sections are recorded as the cross sections of the stainless steel pipe, and a number of equally spaced measuring points are selected on the outer circumference of each cross section, which are recorded as the roundness measuring points. The diameter of each roundness measuring point of each cross section of the stainless steel pipe is measured, which is recorded as the diameter l of each roundness measuring point of each cross section of the stainless steel pipe. mu , m represents the number of the mth section, m = 1, 2, ..., q, u represents the number of the uth roundness measurement point, u = 1, 2, ..., v, and the diameter of each roundness measurement point of the stainless steel pipe is obtained by taking the average value of the diameter of each roundness measurement point of each section of the stainless steel pipe, which is recorded as Substituting this into the formula The roundness o of the stainless steel pipe is obtained, q represents the number of cross sections, and v represents the number of roundness measurement points.

[0017] The third step is to use a laser collimator to emit a horizontal laser beam from one end of the stainless steel pipe to the other end, and record the laser beam as the reference line of the stainless steel pipe. Then measure the vertical distance from each measuring point of the stainless steel pipe to the reference line, which is recorded as d'. i , the average vertical distance from the stainless steel pipe to the laser beam is obtained by taking the average value, which is recorded as Substituting this into the formula Get the straightness l of the stainless steel pipe.

[0018] Preferably, the specific analysis method of the shape conformity of the stainless steel pipe is: respectively read the dimensional accuracy γ, roundness o, and straightness e of the stainless steel pipe, and substitute them into the formula Get the shape conformity degree δ of the stainless steel pipe, They respectively represent the weight factors of the preset dimensional accuracy, roundness, and straightness.

[0019] Preferably, the specific analysis method of the weld defect parameters of the stainless steel pipe is: spraying the penetrant evenly on the surface of the stainless steel pipe, waiting for a set time to allow the penetrant to penetrate into the stainless steel pipe, removing excess penetrant on the surface of the stainless steel pipe, and applying a developer to the surface of the stainless steel pipe, acquiring a weld image of the stainless steel pipe by a high-definition camera, extracting the area displayed by the developer by edge detection technology, and recording it as each unwelded area, extracting the area of ​​each unwelded area, and accumulating to obtain the unwelded area area of ​​the stainless steel pipe image, and recording it as M 未焊合, through the formula Get the unwelded degree ε, M of the stainless steel pipe 总 Indicates the total area of ​​the set weld zone.

[0020] In the second step, the weld is magnetized by the electric current method, and the magnetic powder is evenly sprinkled on the weld surface to obtain an image, which is recorded as the weld magnetization image. The edge detection technology is used to extract the edge contour of the area not covered by the magnetic powder in the weld magnetization image, and its area is obtained and recorded as the area of ​​each defect area M. j , j represents the number of the jth defect area, j = 1, 2, ..., g, through the formula Obtain the defect degree ρ of the stainless steel pipe.

[0021] Preferably, the specific analysis method of the welding quality analysis module is: read the unwelded degree ε and defect degree ρ of the stainless steel pipe, and substitute them into the formula The weld quality evaluation coefficient λ of the stainless steel pipe is obtained, where φ3 and φ4 represent the weight factors of the set unwelded degree and defect degree, respectively.

[0022] Preferably, the specific analysis method of the pipe surface evaluation module is as follows: the first step is to place the stainless steel pipe stably on the measuring table, turn on the profilometer, and move the probe of the profilometer along the pipe surface at a uniform speed to scan the entire surface of the stainless steel pipe to form a stainless steel pipe profile curve, select a number of sampling points in the stainless steel pipe profile curve according to the set interval, obtain the height of each sampling point, and record it as the height h of each sampling point of the stainless steel pipe profile curve a , a represents the number of the ath sampling point, a=1, 2, ..., b, the average height of the stainless steel pipe profile curve is obtained by averaging the heights of the sampling points, and the average height of the stainless steel pipe profile curve is recorded as At the same time, the peak amplitude and trough amplitude in the stainless steel pipe profile curve are extracted and recorded as h max 、h min , through the formula The surface roughness H of the stainless steel pipe is obtained, μ1 and μ2 represent the preset height fluctuation degree of the profile curve and the weight factor of the maximum height difference, respectively, and b represents the number of sampling points.

[0023] In the second step, the surface of the stainless steel pipe is imaged, recorded as the stainless steel pipe surface image, and the edge contour of each scratch is extracted from the stainless steel pipe surface image by edge detection technology, the scratches are counted, and the length of each scratch is extracted, recorded as Lp, where p represents the length of the pth scratch, p = 1, 2, ..., c, and the average length of each scratch is calculated to obtain the average scratch length. By formula The scratch degree ξ of the stainless steel pipe is obtained, κ1 and κ2 represent the weight factors of the preset number of scratches and the scratch length respectively, c represents the number of scratches, and c0 represents the reference value of the set number of scratches.

[0024] Preferably, the specific analysis method of the surface smoothness of the stainless steel pipe is: read the surface roughness H and the scratch degree ξ of the stainless steel pipe, and substitute them into the formula Get the surface smoothness of stainless steel pipe Where η1 and η2 represent the weight factors of the set surface roughness and scratch degree, respectively.

[0025] Preferably, the specific analysis method of the appearance parameters of the cut surface of the stainless steel pipe is as follows: the first step is to obtain an image of the cut stainless steel pipe, and use it as a 3D model of the stainless steel pipe, which is recorded as a stainless steel pipe cutting model.

[0026] The second step is to read the design model of the stainless steel pipe from the management database, and overlap the stainless steel pipe cutting model with the design model of the stainless steel pipe to obtain the overlapping volume of the stainless steel pipe cutting model.

[0027] The third step is to extract the volume of the design model of the stainless steel pipe, and obtain the cutting size accuracy of the stainless steel pipe by dividing the overlapping volume of the stainless steel pipe cutting model by the volume of the design model of the stainless steel pipe.

[0028] The fourth step is to select a plane parallel to the cutting surface of the stainless steel pipe as the reference plane, record it as the cutting reference plane, select several equally spaced measuring points on the cutting surface of the stainless steel pipe, record them as key measuring points, and measure the vertical distance from each key measuring point of the stainless steel pipe to the cutting reference plane respectively.

[0029] The fifth step is to analyze the vertical distance from each key measuring point of the stainless steel pipe to the cutting reference plane according to the method of analyzing the straightness of the stainless steel pipe to obtain the flatness of the cutting surface of the stainless steel pipe.

[0030] Preferably, the specific analysis method of the cutting evaluation coefficient of the stainless steel pipe is: reading the cutting size accuracy of the stainless steel pipe The flatness of the cutting surface is calculated by the formula The cutting evaluation coefficient ω of the stainless steel pipe fittings is obtained, l′ represents the flatness of the cutting surface of the stainless steel pipe, and θ1 and θ2 represent the weight factors of the set cutting size accuracy and cutting surface flatness, respectively.

[0031] Preferably, the specific analysis method of the process evaluation module is: respectively reading the shape conformity degree δ, weld quality evaluation coefficient λ, surface smoothness of the stainless steel pipe Cutting evaluation coefficient ω, substitute it into the formula Get the process evaluation index of stainless steel pipes Among them, w1, w2, w3, and w4 represent the weight factors of the preset shape compliance degree, weld quality evaluation coefficient, surface smoothness, and cutting evaluation coefficient, respectively. The process evaluation index of the stainless steel pipe is compared with the preset process evaluation index threshold. If the process evaluation index of the stainless steel pipe is greater than or equal to the preset process evaluation index threshold, it means that the process evaluation of the stainless steel pipe is qualified. If the process evaluation index of the stainless steel pipe is less than the preset process evaluation index threshold, it means that the process evaluation of the stainless steel pipe is unqualified, and feedback is given to the system.

[0032] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention obtains the shape conformity of the stainless steel pipe according to the shape parameter analysis of the stainless steel pipe, and can accurately measure the conformity between the actual shape of the stainless steel pipe and the ideal shape, thereby ensuring the correctness of the product shape.

[0033] 2. The present invention obtains the weld quality evaluation coefficient of the stainless steel pipe according to the weld defect parameter analysis of the stainless steel pipe, which helps to detect the defects in the weld in time so as to take targeted measures to repair or improve it.

[0034] 3. The present invention obtains the surface smoothness of the stainless steel pipe according to the surface parameter analysis of the stainless steel pipe, ensures that the stainless steel pipe has good surface smoothness, helps to produce pipes with consistent surface smoothness, and ensures the consistency of product quality.

[0035] 4. The present invention obtains the cutting evaluation coefficient of the stainless steel pipe according to the appearance parameter analysis of the cut surface of the stainless steel pipe. A good cut surface helps to maintain the mechanical properties and performance of the pipe, can reduce subsequent rework caused by poor cutting quality, and save time and cost.

[0036] 5. The present invention obtains the process evaluation index of the stainless steel pipe according to the shape conformity, weld quality evaluation coefficient, surface smoothness and cutting evaluation coefficient of the stainless steel pipe, and provides feedback thereon, which helps to reduce the production of unqualified products, improve the yield rate and ensure that the produced pipes have consistency in performance in all aspects. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0038] Figure 1This is a system module connection diagram of the present invention.

[0039] Figure 2 for Figure 1 Flowchart of the appearance parameters of the cut surface of stainless steel pipes in the pipe surface evaluation module.

[0040] Figure 3 for Figure 1 The process judgment block diagram of the process evaluation module. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] See also Figure 1 As shown, a machine vision-based intelligent detection system for stainless steel pipe production includes a shape evaluation module, a welding quality detection module, a welding quality analysis module, a pipe surface evaluation module, a cutting evaluation module, a process evaluation module, and a management database.

[0043] The management database is connected to the shape evaluation module, welding quality analysis module, pipe surface evaluation module, cutting evaluation module, and process evaluation module; the shape evaluation module, welding quality analysis module, pipe surface evaluation module, cutting evaluation module, and process evaluation module are connected; and the welding quality detection module is connected to the welding quality analysis module.

[0044] The shape evaluation module is used to detect the shape parameters of the stainless steel pipe, and then analyze and obtain the shape compliance degree δ of the stainless steel pipe. The shape parameters include dimensional accuracy, roundness, and straightness.

[0045] The specific analysis method of the shape parameters of the stainless steel pipe is as follows: the first step is to select a number of measuring points on the stainless steel pipe according to the set interval, and to read the pipe diameter size at each measuring point of the stainless steel pipe by clamping a vernier caliper on the outer circumference of each measuring point of the stainless steel pipe, which is recorded as r i , i represents the number of the i-th measurement point, i = 1, 2, ..., n, by substituting it into the formula The diameter accuracy of the stainless steel pipe is obtained, r0 represents the standard diameter of the stainless steel pipe, n represents the number of measuring points, and the wall thickness of each measuring point of the stainless steel pipe is obtained by placing the wall thickness micrometer perpendicular to the surface of each measuring point of the stainless steel pipe, which is recorded as d i , substituting it into the formula The wall thickness accuracy β of the stainless steel pipe is obtained, d0 represents the set standard wall thickness of the stainless steel pipe, and the dimensional accuracy γ of the stainless steel pipe is obtained by the formula γ=α*φ1+β*φ2, φ1 and φ2 represent the weight factors of the set diameter accuracy and wall thickness accuracy of the stainless steel pipe respectively; it can accurately evaluate whether the stainless steel pipe meets the set standard diameter and wall thickness, which helps to ensure product quality.

[0046] It should be noted that φ1 can be set to 0.5 and φ2 can be set to 0.5. Accurate pipe diameter is critical for connection and coordination with other pipe fittings and equipment. Large deviation in pipe diameter may result in loose connection or failure to connect normally. Wall thickness affects the strength and stability of the pipe. Inaccurate wall thickness may cause problems under specific pressure or load conditions, affecting its safety and reliability. Therefore, the weights of pipe diameter accuracy and wall thickness accuracy of stainless steel pipes are equal.

[0047] In the second step, the cross section at either end of the stainless steel pipe is taken as the initial cross section, and a new cross section is selected at a set distance along the length direction of the stainless steel pipe. The selected cross sections are recorded as the cross sections of the stainless steel pipe, and a number of equally spaced measuring points are selected on the outer circumference of each cross section, which are recorded as the roundness measuring points. The diameter of each roundness measuring point of each cross section of the stainless steel pipe is measured, which is recorded as the diameter l of each roundness measuring point of each cross section of the stainless steel pipe. mu , m represents the number of the mth section, m = 1, 2, ..., q, u represents the number of the uth roundness measurement point, u = 1, 2, ..., v, and the diameter of each roundness measurement point of the stainless steel pipe is obtained by taking the average value of the diameter of each roundness measurement point of each section of the stainless steel pipe, which is recorded as Substituting this into the formula The roundness o of the stainless steel pipe is obtained, q represents the number of sections, and v represents the number of roundness measurement points; by measuring multiple sections along the length direction, we can have a more comprehensive understanding of the diameter of the pipe at different positions, avoiding the one-sidedness caused by measuring only a single section, and can monitor the diameter consistency along the length of the pipe to promptly discover possible uneven changes or defects.

[0048] The third step is to use a laser collimator to emit a horizontal laser beam from one end of the stainless steel pipe to the other end, and record the laser beam as the reference line of the stainless steel pipe. Then measure the vertical distance from each measuring point of the stainless steel pipe to the reference line, which is recorded as d'. i , the average vertical distance from the stainless steel pipe to the laser beam is obtained by taking the average value, which is recorded as Substituting this into the formula Obtain the straightness e of the stainless steel pipe; it can fully understand the deviation of the stainless steel pipe in the straight direction and accurately evaluate its straightness, which helps to ensure that the stainless steel pipe meets the standards in terms of straight line shape and improve product quality.

[0049] The specific analysis method of the shape conformity of the stainless steel pipe is as follows: respectively read the dimensional accuracy γ, roundness o, and straightness l of the stainless steel pipe, and substitute them into the formula Get the shape conformity degree δ of the stainless steel pipe, They represent the preset weight factors of dimensional accuracy, roundness and straightness respectively; they help to ensure that the produced stainless steel pipes are more in line with the requirements in shape, thereby improving the overall quality and performance of the product.

[0050] It should be noted that It can be set to 0.4. It can be set to 0.3. It can be set to 0.3. Usually, the impact of dimensional accuracy may be relatively large. Dimensional accuracy is directly related to the fit and installation of pipes and other components. Inaccurate dimensions may lead to abnormal assembly or loose connection, which has a greater impact on actual use. Poor roundness will affect the appearance quality of the pipe and some occasions with high requirements for fluid flow uniformity, such as increased local resistance. Straightness mainly affects the stability of pipe installation and use, especially in long pipe applications. If the straightness deviation is large, it may lead to uneven installation or shaking during use. Therefore, the weight corresponding to dimensional accuracy is relatively high.

[0051] The welding quality detection module is used to detect the weld defect parameters of stainless steel pipes. The weld defect parameters include the degree of non-welding and the degree of defects.

[0052] The specific analysis method of the weld defect parameters of the stainless steel pipe is as follows: in the first step, the penetrant is evenly sprayed on the surface of the stainless steel pipe, and the penetrant is allowed to penetrate into the stainless steel pipe for a set time, and the excess penetrant on the surface of the stainless steel pipe is removed, and a developer is applied to the surface of the stainless steel pipe, and the weld image of the stainless steel pipe is obtained by a high-definition camera, and the area displayed by the developer is extracted by edge detection technology, which is recorded as each unwelded area, and the area of ​​each unwelded area is extracted and accumulated to obtain the unwelded area area of ​​the stainless steel pipe image, which is recorded as M 未焊合 , through the formula Get the unwelded degree ε, M of the stainless steel pipe 总 Represents the total area of ​​the set weld area; it can effectively screen out problematic pixels, so as to fully grasp the quality status of the weld and not miss any potential defects. It helps to discover problems such as unwelded joints as early as possible, so that timely measures can be taken to repair or improve them and reduce subsequent quality risks.

[0053] In the second step, the weld is magnetized by the electric current method, and the magnetic powder is evenly spread on the weld surface to obtain an image, which is recorded as the weld magnetization image. The edge contour of the area not covered by the magnetic powder in the weld magnetization image is extracted by edge detection technology, and its area is obtained and recorded as the area of ​​each defect area M. j , j represents the number of the jth defect area, j = 1, 2, ..., g, through the formula The defect degree ρ of the stainless steel pipe is obtained; it is convenient to intuitively understand and compare the defect conditions of different welds. After clarifying the defect degree, it is conducive to taking targeted measures to improve the welding process and improve product quality.

[0054] The welding quality analysis module is used to obtain the weld quality evaluation coefficient λ of the stainless steel pipe according to the weld defect parameter analysis of the stainless steel pipe.

[0055] The specific analysis method of the welding quality analysis module is: read the unwelded degree ε and defect degree ρ of the stainless steel pipe, and substitute them into the formula The weld quality evaluation coefficient λ of the stainless steel pipe is obtained, where φ3 and φ4 represent the weight factors of the set unwelded degree and defect degree respectively; according to the evaluation results, the welding process can be adjusted or remedial measures can be taken to improve the weld quality.

[0056] It should be noted that φ3 can be set to 0.6 and φ4 can be set to 0.4. The degree of non-welding has a great influence on the integrity and connection strength of the weld. If there is serious non-welding, it may directly lead to weld failure and make the pipe prone to serious problems such as breakage during use. This is the key core issue of weld quality, so the weight is relatively large. Although the degree of defects is also very important, relatively speaking, some smaller defects may not immediately have a fatal impact on the overall performance of the weld, and may be acceptable to a certain extent through subsequent treatment. Of course, serious defects will also bring greater hidden dangers, but overall their importance is slightly less than the degree of non-welding, so the weight corresponding to the degree of non-welding is higher.

[0057] The pipe surface evaluation module is used to detect the surface parameters of stainless steel pipes and then analyze the surface smoothness of stainless steel pipes. Surface parameters include surface roughness and scratch degree.

[0058] The specific analysis method of the pipe surface evaluation module is as follows: the first step is to place the stainless steel pipe stably on the measuring table, turn on the profilometer, and let the probe of the profilometer move at a uniform speed along the pipe surface to scan the entire surface of the stainless steel pipe to form a stainless steel pipe profile curve. In the stainless steel pipe profile curve, a number of sampling points are selected according to the set interval, and the height of each sampling point is obtained, which is recorded as the height h of each sampling point of the stainless steel pipe profile curve. a, a represents the number of the ath sampling point, a=1, 2, ..., b, the average height of the stainless steel pipe profile curve is obtained by averaging the heights of the sampling points, and the average height of the stainless steel pipe profile curve is recorded as At the same time, the peak amplitude and trough amplitude in the stainless steel pipe profile curve are extracted and recorded as h max 、h min , through the formula The surface roughness H of the stainless steel pipe is obtained, μ1 and μ2 represent the preset height fluctuation degree of the profile curve and the weight factor of the maximum height difference, respectively, and b represents the number of sampling points; ensure a comprehensive and detailed scan of the surface of the stainless steel pipe to obtain accurate profile information.

[0059] It should be noted that μ1 can be set to 0.5, and μ2 can be set to 0.5. The height fluctuation of the contour curve can more comprehensively reflect the overall fluctuation of the surface of the stainless steel pipe. It comprehensively reflects the uniformity and consistency of the surface roughness, and has a greater impact on the overall roughness level. The maximum height difference directly reflects the extreme case of surface roughness and can highlight the roughest degree of the surface. This is also very critical for some application scenarios that are more sensitive to local roughness. Therefore, its weight cannot be ignored. Therefore, the weights corresponding to the height fluctuation of the contour curve and the maximum height difference are equal.

[0060] In the second step, the surface of the stainless steel pipe is imaged, recorded as the stainless steel pipe surface image, and the edge contour of each scratch is extracted from the stainless steel pipe surface image by edge detection technology, the scratches are counted, and the length of each scratch is extracted, recorded as Lp, where p represents the length of the pth scratch, p = 1, 2, ..., c, and the average length of each scratch is calculated to obtain the average scratch length. By formula The scratch degree ξ of the stainless steel pipe is obtained, κ1 and κ2 represent the weight factors of the preset number of scratches and scratch length respectively, c represents the number of scratches, and c0 represents the reference value of the set number of scratches; through image acquisition and edge detection, the surface of the pipe can be fully covered without missing any scratch, which helps to timely discover the scratch problem on the surface of the pipe so that corresponding measures can be taken.

[0061] It should be noted that κ1 can be set to 0.4 and κ2 can be set to 0.6. The scratch length has a relatively more direct and critical impact on the appearance and performance of stainless steel pipes. Longer scratches may more easily cause greater impact on the strength and corrosion resistance of the pipe, and can more intuitively reflect the severity of the scratches. Although the number of scratches can also reflect the number of scratches, the number of scratches alone may not be as accurate as the length in measuring the impact on the degree of scratches on the pipe. Therefore, the weight corresponding to the scratch length is higher.

[0062] The specific analysis method of the surface smoothness of the stainless steel pipe is as follows: read the surface roughness H and scratch degree ξ of the stainless steel pipe and substitute them into the formula Get the surface smoothness of stainless steel pipe Among them, η1 and η2 represent the weight factors of the set surface roughness and scratch degree respectively; it is helpful to effectively monitor and manage the surface quality of stainless steel pipes, and to a certain extent can predict the relevant performance of stainless steel pipes in practical applications, such as friction performance.

[0063] It should be noted that η1 can be set to 0.6 and η2 can be set to 0.4. Surface roughness is a basic and key indicator for measuring the surface smoothness of stainless steel pipes. It can more comprehensively and meticulously reflect the microscopic undulations of the entire surface and has a great impact on the surface smoothness. Although the degree of scratches will also affect the smoothness, it is relatively more local and discrete. The impact on the overall smoothness is not as extensive and fundamental as the surface roughness. Therefore, the weight corresponding to the surface roughness is higher.

[0064] The cutting evaluation module is used to detect the appearance parameters of the cut surface of the stainless steel pipe, and then analyze and obtain the cutting evaluation coefficient ω of the stainless steel pipe, including the cutting size accuracy Cutting surface flatness l′.

[0065] See also Figure 2 As shown, the specific analysis method of the appearance parameters of the cut surface of the stainless steel pipe is as follows: the first step is to obtain an image of the cut stainless steel pipe fitting, and use it as a 3D model of the stainless steel pipe fitting, which is recorded as a stainless steel pipe fitting cutting model; it can present the actual shape of the stainless steel pipe fitting more intuitively and in detail.

[0066] The second step is to read the design model of the stainless steel pipe from the management database, overlap the stainless steel pipe cutting model with the design model of the stainless steel pipe to obtain the overlapping volume of the stainless steel pipe cutting model; compare with the design model to find deviations and problems, which is beneficial to quality control.

[0067] The third step is to extract the volume of the design model of the stainless steel pipe, and obtain the cutting size accuracy of the stainless steel pipe by dividing the overlapping volume of the stainless steel pipe cutting model by the volume of the design model of the stainless steel pipe; good cutting size accuracy and straightness can ensure the performance and reliability of the stainless steel pipe in practical applications.

[0068] It should be noted that the specific analysis method of the cutting size accuracy of the stainless steel pipe is: read the overlapping volume of the stainless steel pipe cutting model, recorded as V 切割 , read the volume of the standard cutting model of the stainless steel pipe from the management database, recorded as V标准 , substituting it into the formula Get the cutting size accuracy of stainless steel pipes

[0069] The fourth step is to select a plane parallel to the cutting surface of the stainless steel pipe as the reference plane, record it as the cutting reference plane, select several equally spaced measuring points on the cutting surface of the stainless steel pipe, record them as key measuring points, and measure the vertical distance from each key measuring point of the stainless steel pipe to the cutting reference plane respectively.

[0070] The fifth step is to analyze the vertical distance from each key measuring point of the stainless steel pipe to the cutting reference plane according to the method of analyzing the straightness of the stainless steel pipe to obtain the flatness of the cutting surface of the stainless steel pipe; detecting and analyzing the flatness of the cutting surface to ensure the regularity of the shape of the pipe fittings is of great significance for subsequent use and assembly.

[0071] The specific analysis method of the cutting evaluation coefficient of the stainless steel pipe is: read the cutting size accuracy of the stainless steel pipe fittings The flatness of the cutting surface is calculated by the formula The cutting evaluation coefficient ω of the stainless steel pipe fittings is obtained, l′ represents the flatness of the cutting surface of the stainless steel pipe, θ1 and θ2 represent the weight factors of the set cutting size accuracy and cutting surface flatness, respectively; ensuring that the stainless steel pipe fittings have better adaptability and stability in subsequent use, because good cutting quality is usually related to performance, which is conducive to maintaining the consistency of product quality and reducing batch differences.

[0072] The process evaluation module is used to analyze and obtain the process evaluation index of the stainless steel pipe according to the shape conformity, weld quality evaluation coefficient, surface smoothness and cutting evaluation coefficient of the stainless steel pipe, and provide feedback thereon.

[0073] See also Figure 3 As shown, the specific analysis method of the process evaluation module is: read the shape compliance degree δ, weld quality evaluation coefficient λ, surface smoothness of the stainless steel pipe respectively. Cutting evaluation coefficient ω, substitute it into the formula Get the process evaluation index of stainless steel pipes Among them, w1, w2, w3, and w4 represent the weight factors of the preset shape compliance degree, weld quality evaluation coefficient, surface smoothness, and cutting evaluation coefficient respectively. The process evaluation index of the stainless steel pipe is compared with the preset process evaluation index threshold. If the process evaluation index of the stainless steel pipe is greater than or equal to the preset process evaluation index threshold, it means that the process evaluation of the stainless steel pipe is qualified. If the process evaluation index of the stainless steel pipe is less than the preset process evaluation index threshold, it means that the process evaluation of the stainless steel pipe is unqualified, and feedback is given to the system; comprehensive consideration of multiple key aspects of the pipe, such as shape, weld quality, surface condition and cutting quality, can more comprehensively and accurately evaluate the overall process quality. At the same time, unqualified situations can be fed back to the system in a timely manner, which is conducive to quickly taking improvement measures to improve product quality and production technology.

[0074] It should be noted that w1 can be set to 0.3, w2 can be set to 0.3, w3 can be set to 0.2, and w4 can be set to 0.2. Whether the shape meets the requirements is one of the basic properties of the pipe. Failure to meet the shape requirements may directly lead to the pipe being unusable, but its importance is relatively not the most prominent. The quality of the weld is directly related to key aspects such as the connection strength and sealing performance of the pipe, and has a greater impact on the overall quality. Although the surface smoothness is also important, its impact on the overall performance and use of the pipe is slightly less than the first two. The cutting quality mainly affects the appearance and subsequent processing convenience, but it is relatively not the most core influencing factor. Therefore, the weights corresponding to the shape compliance and weld quality evaluation coefficient are slightly higher.

[0075] Management database for storing standard diameters and standard cutting models of stainless steel pipes.

[0076] The present invention obtains the shape conformity of the stainless steel pipe according to the shape parameter analysis of the stainless steel pipe, obtains the weld quality evaluation coefficient of the stainless steel pipe according to the weld defect parameter analysis of the stainless steel pipe, obtains the surface smoothness of the stainless steel pipe according to the surface parameter analysis of the stainless steel pipe, obtains the cutting evaluation coefficient of the stainless steel pipe according to the appearance parameter analysis of the cut surface of the stainless steel pipe, and then obtains the process evaluation index of the stainless steel pipe through comprehensive analysis, and feeds back the process evaluation index, which helps to reduce the generation of unqualified products, improve the yield rate, and ensure that the produced pipes have consistency in performance in all aspects.

[0077] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations on the present invention. A person skilled in the art may make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention and they are still covered by the protection scope of the present invention.

Claims

1. An intelligent detection system for stainless steel pipe production based on machine vision, characterized in that: The system specifically includes the following modules: The shape evaluation module is used to detect the shape parameters of the stainless steel pipe and then analyze the shape conformity degree δ of the stainless steel pipe. The shape parameters include dimensional accuracy, roundness, and straightness. Welding quality inspection module, used to inspect weld defect parameters of stainless steel pipes, including the degree of unweldedness and defect degree; The welding quality analysis module is used to obtain the weld quality evaluation coefficient λ of the stainless steel pipe according to the weld defect parameter analysis of the stainless steel pipe; The pipe surface evaluation module is used to detect the surface parameters of stainless steel pipes and then analyze the surface smoothness of stainless steel pipes. Surface parameters include surface roughness and scratch degree; The cutting evaluation module is used to detect the appearance parameters of the cut surface of the stainless steel pipe, and then analyze and obtain the cutting evaluation coefficient ω of the stainless steel pipe. The appearance parameters of the cut surface of the stainless steel pipe include the cutting size accuracy. Cutting surface flatness l′; The process evaluation module is used to analyze and obtain the process evaluation index of the stainless steel pipe according to the shape conformity, weld quality evaluation coefficient, surface smoothness, and cutting evaluation coefficient of the stainless steel pipe, and provide feedback; Management database for storing standard diameters and standard cutting models of stainless steel pipes.

2. According to claim 1, a machine vision-based intelligent detection system for stainless steel pipe production is characterized in that: Specific analysis of the shape parameters of the stainless steel pipe The method is: The first step is to select several measuring points on the stainless steel pipe according to the set spacing, and read the pipe diameter size at each measuring point of the stainless steel pipe by clamping the vernier caliper on the outer circumference of each measuring point of the stainless steel pipe, which is recorded as r. i , i represents the number of the i-th measurement point, i = 1, 2, ..., n, by substituting it into the formula The diameter accuracy of the stainless steel pipe is obtained, r0 represents the standard diameter of the stainless steel pipe, n represents the number of measuring points, and the wall thickness of each measuring point of the stainless steel pipe is obtained by placing the wall thickness micrometer perpendicular to the surface of each measuring point of the stainless steel pipe, which is recorded as d i , substituting it into the formula The wall thickness accuracy β of the stainless steel pipe is obtained, d0 represents the set standard wall thickness of the stainless steel pipe, and the dimensional accuracy γ of the stainless steel pipe is obtained by the formula γ=α*φ1+β*φ2, φ1 and φ2 represent the weight factors of the set diameter accuracy and wall thickness accuracy of the stainless steel pipe respectively; In the second step, the cross section at either end of the stainless steel pipe is taken as the initial cross section, and a new cross section is selected at a set distance along the length direction of the stainless steel pipe. The selected cross sections are recorded as the cross sections of the stainless steel pipe, and a number of equally spaced measuring points are selected on the outer circumference of each cross section, which are recorded as the roundness measuring points. The diameter of each roundness measuring point of each cross section of the stainless steel pipe is measured, which is recorded as the diameter l of each roundness measuring point of each cross section of the stainless steel pipe. mu , m represents the number of the mth section, m = 1, 2, ..., q, u represents the number of the uth roundness measurement point, u = 1, 2, ..., v, and the diameter of each roundness measurement point of the stainless steel pipe is obtained by taking the average value of the diameter of each roundness measurement point of each section of the stainless steel pipe, which is recorded as Substituting this into the formula Get the roundness o of the stainless steel pipe, q represents the number of sections, and v represents the number of roundness measurement points; The third step is to use a laser collimator to emit a horizontal laser beam from one end of the stainless steel pipe to the other end, and record the laser beam as the reference line of the stainless steel pipe. Then measure the vertical distance from each measuring point of the stainless steel pipe to the reference line, which is recorded as d'. i , the average vertical distance from the stainless steel pipe to the laser beam is obtained by taking the average value, which is recorded as Substituting this into the formula Get the straightness l of the stainless steel pipe.

3. According to claim 2, a machine vision-based intelligent detection system for stainless steel pipe production is characterized in that: The specific analysis method for the shape compliance of the stainless steel pipe is as follows: Read the dimensional accuracy γ, roundness o, and straightness l of the stainless steel pipe respectively, and substitute them into the formula Get the shape conformity degree δ of the stainless steel pipe, They respectively represent the weight factors of the preset dimensional accuracy, roundness, and straightness.

4. According to claim 1, a machine vision-based intelligent detection system for stainless steel pipe production is characterized in that: The specific analysis method of the weld defect parameters of the stainless steel pipe is: In the first step, the penetrant is sprayed evenly on the surface of the stainless steel pipe, and the penetrant is allowed to penetrate into the stainless steel pipe for a set time, and the excess penetrant on the surface of the stainless steel pipe is removed. The developer is applied to the surface of the stainless steel pipe, and the weld image of the stainless steel pipe is obtained by a high-definition camera. The area displayed by the developer is extracted using edge detection technology, which is recorded as each unwelded area. The area of ​​each unwelded area is extracted and accumulated to obtain the unwelded area of ​​the stainless steel pipe image, which is recorded as M. 未焊合 , through the formula Get the unwelded degree ε, M of the stainless steel pipe 总 Indicates the total area of ​​the set weld zone; In the second step, the weld is magnetized by the electric current method, and the magnetic powder is evenly spread on the weld surface to obtain an image, which is recorded as the weld magnetization image. The edge contour of the area not covered by the magnetic powder in the weld magnetization image is extracted by edge detection technology, and its area is obtained and recorded as the area of ​​each defect area M. j , j represents the number of the jth defect area, j = 1, 2, ..., g, through the formula Obtain the defect degree ρ of the stainless steel pipe.

5. The intelligent detection system for stainless steel pipe production based on machine vision according to claim 4 is characterized in that: The specific analysis method of the welding quality analysis module is: Read the unwelded degree ε and defect degree ρ of the stainless steel pipe and substitute them into the formula The weld quality evaluation coefficient λ of the stainless steel pipe is obtained, where φ3 and φ4 represent the weight factors of the set unwelded degree and defect degree, respectively.

6. The intelligent detection system for stainless steel pipe production based on machine vision according to claim 1 is characterized in that: The specific analysis method of the pipe surface evaluation module is: The first step is to place the stainless steel pipe steadily on the measuring table, turn on the profilometer, and let the probe of the profilometer move at a uniform speed along the surface of the pipe to scan the entire surface of the stainless steel pipe to form a stainless steel pipe profile curve. In the stainless steel pipe profile curve, select several sampling points according to the set interval, and obtain the height of each sampling point, which is recorded as the height h of each sampling point of the stainless steel pipe profile curve. a , a represents the number of the ath sampling point, a=1, 2, ..., b, the average height of the stainless steel pipe profile curve is obtained by averaging the heights of the sampling points, and the average height of the stainless steel pipe profile curve is recorded as At the same time, the peak amplitude and trough amplitude in the stainless steel pipe profile curve are extracted and recorded as h max 、h min , through the formula The surface roughness H of the stainless steel pipe is obtained, where μ1 and μ2 represent the preset height fluctuation degree of the profile curve and the weight factor of the maximum height difference, respectively, and b represents the number of sampling points; In the second step, the surface of the stainless steel pipe is imaged, recorded as the stainless steel pipe surface image, and the edge contour of each scratch is extracted from the stainless steel pipe surface image by edge detection technology, the scratches are counted, and the length of each scratch is extracted, recorded as L. p , p represents the length of the pth scratch, p = 1, 2, ..., c, and the average length of each scratch is obtained by averaging the lengths of the scratches By formula The scratch degree ξ of the stainless steel pipe is obtained, κ1 and κ2 represent the weight factors of the preset number of scratches and the scratch length respectively, c represents the number of scratches, and c0 represents the reference value of the set number of scratches.

7. The intelligent detection system for stainless steel pipe production based on machine vision according to claim 6 is characterized in that: The specific analysis method of the surface smoothness of the stainless steel pipe is: Read the surface roughness H and scratch degree ξ of the stainless steel pipe and substitute them into the formula Get the surface smoothness of stainless steel pipe Where η1 and η2 represent the weight factors of the set surface roughness and scratch degree, respectively.

8. The intelligent detection system for stainless steel pipe production based on machine vision according to claim 2 is characterized in that: The specific analysis method of the appearance parameters of the cut surface of the stainless steel pipe is as follows: The first step is to obtain an image of the cut stainless steel pipe fitting, and use it as a 3D model of the stainless steel pipe fitting, which is recorded as a stainless steel pipe fitting cutting model; The second step is to read the design model of the stainless steel pipe from the management database, and overlap the stainless steel pipe cutting model with the design model of the stainless steel pipe to obtain the overlapping volume of the stainless steel pipe cutting model; The third step is to extract the volume of the design model of the stainless steel pipe, and obtain the cutting size accuracy of the stainless steel pipe by dividing the overlapping volume of the stainless steel pipe cutting model by the volume of the design model of the stainless steel pipe; The fourth step is to select a plane parallel to the cutting surface of the stainless steel pipe as a reference plane, record it as the cutting reference plane, select a number of equally spaced measuring points on the cutting surface of the stainless steel pipe, record them as key measuring points, and measure the vertical distance from each key measuring point of the stainless steel pipe to the cutting reference plane respectively; The fifth step is to analyze the vertical distance from each key measuring point of the stainless steel pipe to the cutting reference plane according to the method of analyzing the straightness of the stainless steel pipe to obtain the flatness of the cutting surface of the stainless steel pipe.

9. The intelligent detection system for stainless steel pipe production based on machine vision according to claim 8 is characterized in that: The specific analysis method of the cutting evaluation coefficient of the stainless steel pipe is: Read the cutting dimension accuracy of stainless steel pipe fittings The flatness of the cutting surface is calculated by the formula The cutting evaluation coefficient ω of the stainless steel pipe fittings is obtained, l′ represents the flatness of the cutting surface of the stainless steel pipe, and θ1 and θ2 represent the weight factors of the set cutting size accuracy and cutting surface flatness, respectively.

10. The intelligent detection system for stainless steel pipe production based on machine vision according to claim 1, characterized in that: The specific analysis method of the process evaluation module is: Read the shape conformity degree δ, weld quality evaluation coefficient λ, and surface smoothness of the stainless steel pipe respectively Cutting evaluation coefficient ω, substitute it into the formula Get the process evaluation index of stainless steel pipes Among them, w1, w2, w3, and w4 represent the weight factors of the preset shape compliance degree, weld quality evaluation coefficient, surface smoothness, and cutting evaluation coefficient, respectively. The process evaluation index of the stainless steel pipe is compared with the preset process evaluation index threshold. If the process evaluation index of the stainless steel pipe is greater than or equal to the preset process evaluation index threshold, it means that the process evaluation of the stainless steel pipe is qualified. If the process evaluation index of the stainless steel pipe is less than the preset process evaluation index threshold, it means that the process evaluation of the stainless steel pipe is unqualified, and feedback is given to the system.

Images