CN-121767362-B - Drain pipe shearing quality detection method and system based on shape recognition

CN121767362BCN 121767362 BCN121767362 BCN 121767362BCN-121767362-B

Abstract

The invention relates to the technical field of industrial detection, in particular to a drain pipe shearing quality detection method and system based on shape recognition. The method comprises the steps of obtaining a gray level image of a shearing section of a drain pipe, extracting sub-pixel edge points, constructing a polar coordinate sub-pixel edge sequence comprising an original radial distance and a shearing phase angle, calculating a mechanical collapse index, calculating an elastic modal weight and a plastic collapse weight based on the shearing phase angle, obtaining a local curvature and a local gradient distribution standard deviation of the sub-pixel edge points, calculating a theoretical radial distance by combining the original radial distance, the mechanical collapse index, the elastic modal weight and the plastic collapse weight, further calculating a true diameter and roundness error of the drain pipe, and generating an equipment adjustment feedback signal. According to the invention, through a photoform decoupling and rheological compensation mechanism, the influence of reflection interference and shearing stress deformation is effectively inhibited, and the high-precision detection of the shearing quality of the drain pipe is realized.

Inventors

ZHANG TAO
WEI WUCHAO
Qu Chuanlong
ZHANG XIAODONG

Assignees

陕西润泽博泽科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260303

Claims (7)

1. The method for detecting the shearing quality of the drain pipe based on shape recognition is characterized by comprising the following steps of: Acquiring a gray level image of a shearing section of a drain pipe, extracting sub-pixel edge points, and constructing a polar coordinate sub-pixel edge sequence based on the sub-pixel edge points, wherein the polar coordinate sub-pixel edge sequence comprises an original radial distance and a shearing phase angle of the sub-pixel edge points; Extracting a gray profile by taking the center away from the center of the drain pipe wall as the outer side and the center pointing to the drain pipe wall as the inner side along the normal direction of the sub-pixel edge point, and calculating the mechanical collapse index of the sub-pixel edge point according to the inner gradient index attenuation rate and the outer gradient index attenuation rate obtained by carrying out index fitting on the gray profile; Calculating the actual diameter and roundness error of the drain pipe based on the theoretical radial distance, and generating a quality judgment result and an equipment adjustment feedback signal according to the roundness error; the mechanical collapse index satisfies the relationship: the polar sub-pixel edge sequence comprises a plurality of sub-pixel edge points, wherein, Is the first Mechanical collapse index of the edge points of the sub-pixels, 、 Respectively the first The inner gradient index attenuation rate and the outer gradient index attenuation rate of each sub-pixel edge point, To prevent a very small positive constant with zero denominator, As a threshold value for the tolerance to noise, Is a step function; the elastic modal weights satisfy the relationship: The plastic collapse weight satisfies the relation: the polar sub-pixel edge sequence comprises a plurality of sub-pixel edge points, wherein, 、 Respectively the first Elastic modal weight and plastic collapse weight of the edge points of the sub-pixels, Is the first The shear phase angle of the edge points of the sub-pixels, As the sidewall rebound enhancement factor, Focusing coefficients for the plastic region; The theoretical radial distance satisfies the relationship: the polar sub-pixel edge sequence comprises a plurality of sub-pixel edge points, wherein, Is the first The theoretical radial distance of the edge points of the sub-pixels, Is the first The original radial distance of the edge points of the sub-pixels, 、 Respectively the first Local curvature and local gradient distribution standard deviation of the edge points of each sub-pixel, As the reference curvature of the curve to be processed, Is the standard deviation of the reference gradient distribution, As a correlation coefficient of the modulus of elasticity, For the compensation coefficient of plastic collapse, As a natural logarithmic function, Is a very small positive constant.
2. The method for detecting the shear quality of the drain pipe based on the shape recognition according to claim 1, wherein the noise tolerance threshold is calibrated comprehensively through experiments.
3. The method for detecting the shearing quality of the drain pipe based on shape recognition according to claim 1, wherein the equipment adjusting feedback signal comprises calibrating a 0-degree phase based on a physical installation position of a tool for shearing the drain pipe, wherein the tool comprises a cutter and a clamp, if a quality judgment result is unqualified, counting distribution conditions of sub-pixel edge points with roundness errors exceeding a preset threshold, generating a cutter abrasion early warning signal if the distribution conditions are concentrated in a 180-degree phase region, and generating a clamp pressure adjusting signal if the distribution conditions are concentrated in a 90-degree phase region.
4. The method for detecting the shear quality of the drain pipe based on the shape recognition according to claim 1, wherein the extraction of the sub-pixel edge points is realized by using an edge detection algorithm.
5. The method for detecting shear quality of a drain pipe based on shape recognition according to claim 1, wherein the sidewall rebound enhancement factor is calibrated synthetically by experiment.
6. The method for detecting the shearing quality of the drain pipe based on the shape recognition according to claim 1, wherein the gray level image is acquired by a high-frequency industrial camera matched with a low-angle annular light source.
7. A shape recognition based drain pipe shear quality detection system comprising a processor and a memory, the memory storing computer program instructions that when executed by the processor implement a shape recognition based drain pipe shear quality detection method according to any of claims 1-6.

Description

Drain pipe shearing quality detection method and system based on shape recognition Technical Field The invention relates to the technical field of industrial detection, in particular to a drain pipe shearing quality detection method and system based on shape recognition. Background The drain pipe, especially the drain pipe made of high molecular materials such as PVC-U, PE, is widely applied to construction and municipal engineering, and fixed-length shearing is one of the key procedures in the production process of the drain pipe, however, because the drain pipe has certain plasticity and elasticity, and the shearing process belongs to a strong damage process, the quality of a sheared section is often seriously affected by shearing stress. Traditional quality detection of the shear section of the drain pipe mainly depends on manual caliper measurement or basic machine vision technology. However, the manual measurement efficiency is low and subjective errors exist, but the existing machine vision method mostly adopts pixel-level edge detection operators such as Canny or Sobel to extract the contour of the cut section of the drain pipe, and then the roundness is directly fitted. However, this approach presents two major challenges in practical applications: Firstly, optical calibration is difficult, under the irradiation of an industrial light source, the pipe orifice chamfer angle and the arc surface of the drain pipe are extremely easy to generate strong specular reflection highlight, the edge generated by physical collapse and the facula edge generated by optical reflection cannot be distinguished by the traditional gradient amplitude detection method, the reflection is easily misjudged as serious edge collapse or notch, and the false alarm rate is extremely high; Secondly, the deformation mechanism is complex, a non-uniform stress field is introduced to the shear section of the drain pipe in the shearing process, so that the shear section is subjected to complex nonlinear deformation, such as elastic flattening of the top and plastic inner rolling of the bottom, the conventional detection method generally assumes that the drain pipe is subjected to uniform geometric deformation, and a compensation mechanism for a physical rebound process is lacked, so that the measured size cannot reflect the real geometric property of the drain pipe in a stress-free state, and the optimization of the production process is difficult to guide. Disclosure of Invention The invention provides a drain pipe shearing quality detection method and system based on shape recognition, which aims to solve the problems of low detection precision and high misjudgment rate caused by incapability of distinguishing optical reflection from physical collapse and lack of a compensation mechanism aiming at non-uniform shearing stress in the traditional method. In a first aspect, the invention provides a method for detecting shearing quality of a drain pipe based on shape recognition, which adopts the following technical scheme: A drain pipe shearing quality detection method based on shape recognition comprises the steps of obtaining a gray image of a shearing section of a drain pipe, extracting sub-pixel edge points, constructing a polar coordinate sub-pixel edge sequence based on the sub-pixel edge points, wherein the polar coordinate sub-pixel edge sequence comprises an original radial distance and a shearing phase angle of the sub-pixel edge points, extracting a gray profile taking the center of the wall of the drain pipe as the outer side and the center of the wall of the drain pipe as the inner side along the normal direction of the sub-pixel edge points, calculating a mechanical collapse index of the sub-pixel edge points according to an inner gradient index attenuation rate and an outer gradient index attenuation rate obtained by carrying out index fitting on the gray profile, calculating an elastic mode weight and a plastic collapse weight of the sub-pixel edge points based on the shearing phase angle, calculating a local curvature and a local gradient distribution standard deviation of the sub-pixel edge points, and combining the original radial distance, the mechanical collapse index, the elastic mode weight and the plastic collapse weight, calculating a theoretical radial distance of the sub-pixel edge points, calculating a true diameter and an error of the drain pipe according to the theoretical radial distance, and an error, and a roundness and a feedback signal and a judgment result. According to the invention, by introducing a photoform decoupling mechanism based on bilateral gradient attenuation difference and utilizing the asymmetric difference of Gaussian symmetry and physical collapse of optical reflection, a mechanical collapse index is constructed, so that optical artifacts can be precisely stripped, the real mechanical deformation can be locked, and the misjudgment problem in a strong reflection environment can be solved