CN114782451B - Workpiece defect detection method and device, electronic equipment and readable storage medium - Google Patents

Abstract

The application discloses a workpiece defect detection method, a device, an electronic device and a readable storage medium, wherein the workpiece defect detection method comprises the following steps: acquiring polarized light intensity information of a workpiece to be detected, and determining total light intensity information, corresponding polarization degree information and corresponding polarization angle information corresponding to the workpiece to be detected according to the polarized light intensity information; carrying out weighting polymerization on the total light intensity information, the polarization degree information and the polarization angle information to obtain a workpiece polarization characteristic image; carrying out workpiece edge contour detection on the workpiece polarization characteristic image to obtain edge contour characteristic information; and carrying out workpiece defect detection on the workpiece to be detected according to the edge profile characteristic information to obtain a workpiece defect detection result. The technical problem that the workpiece defect detection accuracy is low is solved.

Description

Workpiece defect detection method, device, electronic device and readable storage medium

technical field

The present application relates to the field of industrial inspection, and in particular, to a method, device, electronic device and readable storage medium for workpiece defect detection.

Background technique

With the continuous development of industrial inspection, the application of workpiece defect detection technology has become more and more extensive. Mechanical workpieces will have various defects during processing due to collisions, scratches and the process itself. The defects of workpieces will affect the use of some precision machines. , so it is necessary to detect workpiece defects. At present, the method of detecting these defects in the factory is mainly photoelectric non-contact inspection, such as using industrial cameras and intelligent algorithms to identify workpiece defects, but the cleanliness of the factory is low, and liquids such as processing coolant and oil will contaminate the workpiece, thereby causing the workpiece The stains and defects on the surface cannot be distinguished in the image, resulting in low accuracy of workpiece defect detection.

SUMMARY OF THE INVENTION

The main purpose of the present application is to provide a workpiece defect detection method, device, electronic device and readable storage medium, which aims to solve the technical problem of low workpiece defect detection accuracy.

In order to achieve the above purpose, the present application provides a workpiece defect detection method, the workpiece defect detection method includes:

Obtaining the polarized light intensity information of the workpiece to be measured, and determining the total light intensity information, the corresponding polarization degree information and the corresponding polarization angle information corresponding to the workpiece to be tested according to the polarized light intensity information;

Weighted aggregation is performed on the total light intensity information, the polarization degree information and the polarization angle information to obtain a workpiece polarization characteristic image;

Performing workpiece edge contour detection on the workpiece polarization feature image to obtain edge contour feature information;

According to the edge profile feature information, workpiece defect detection is performed on the workpiece to be tested, and a workpiece defect detection result is obtained.

The application also provides a workpiece defect detection device, the workpiece defect detection device is applied to workpiece defect detection equipment, and the workpiece defect detection device includes:

The light intensity acquisition module is used to obtain the polarized light intensity information of the workpiece to be tested, and according to the polarized light intensity information, determine the total light intensity information, the corresponding polarization degree information and the corresponding polarization angle information corresponding to the workpiece to be tested;

a weighted aggregation module for performing weighted aggregation on the total light intensity information, the polarization degree information and the polarization angle information to obtain a workpiece polarization characteristic image;

a contour detection module, configured to perform contour detection on the feature image matrix in the polarization image information of the workpiece to be tested to obtain edge contour feature information;

The defect detection module is configured to perform workpiece defect detection on the workpiece to be tested according to the edge profile feature information, and obtain a workpiece defect detection result.

The present application also provides an electronic device, the electronic device is a physical device, and the electronic device includes: a memory, a processor, and the workpiece defect detection method stored on the memory and executable on the processor The program of the workpiece defect detection method can realize the steps of the workpiece defect detection method as described above when the program of the workpiece defect detection method is executed by the processor.

The present application also provides a computer-readable storage medium, where a program for implementing the workpiece defect detection method is stored on the computer-readable storage medium, and when the program of the workpiece defect detection method is executed by a processor, the above-mentioned workpiece defect detection is realized steps of the method.

The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned workpiece defect detection method when the computer program is executed by a processor.

The present application provides a workpiece defect detection method, device, electronic device and readable storage medium. First, the polarized light intensity information of the workpiece to be tested is obtained, and the total light intensity information corresponding to the workpiece to be tested is determined according to the polarized light intensity information. , the corresponding polarization degree information and the corresponding polarization angle information, and then weighted aggregation is performed on the total light intensity information, the polarization degree information and the polarization angle information to obtain a workpiece polarization characteristic image, wherein, due to the total light intensity The intensity information, the polarization degree information and the polarization angle information are all optical polarization information, which are not disturbed by stains on the surface of the workpiece, so that the polarization characteristic image of the workpiece can reflect the real surface profile information that is not disturbed by the stains on the surface of the workpiece. The polarization feature image of the workpiece, the workpiece edge contour detection is performed on the workpiece polarization feature image, and the edge contour feature information is obtained. According to the edge contour feature information, the workpiece defect detection is performed on the workpiece to be tested, and the workpiece defect detection result is obtained, which can be To achieve the purpose of detecting workpiece defects based on the real surface profile information that is not disturbed by the surface stains of the workpiece, it overcomes the fact that liquids such as machining coolant and oil will contaminate the workpiece, so that the stains and defects on the surface of the workpiece cannot be distinguished in the image, resulting in workpiece defects. Detecting technical defects with low accuracy improves the accuracy of workpiece defect detection.

Description of drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description serve to explain the principles of the application.

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. In other words, on the premise of no creative labor, other drawings can also be obtained from these drawings.

1 is a schematic flowchart of a first embodiment of a workpiece defect detection method of the present application;

2 is a schematic diagram of the partial derivative operator in step S30 in the first embodiment of the workpiece defect detection method of the present application;

FIG. 3 is a schematic diagram of a device structure of a hardware operating environment involved in a workpiece defect detection method according to an embodiment of the present application.

The realization, functional features and advantages of the present application will be further described with reference to the accompanying drawings in conjunction with the embodiments.

Detailed ways

In order to make the above objects, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, but not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present invention.

Example 1

Mechanical workpieces will have various defects during processing due to collisions, scratches and the process itself. The defects of the workpiece will affect the use of some precision machines, so the workpiece needs to be inspected. At present, when detecting workpiece defects, industrial cameras and intelligent algorithms are mainly used to identify workpiece defects. However, due to the low cleanliness of the factory, liquids such as machining coolant and oil will contaminate the workpiece, so the stains and defects on the surface of the workpiece cannot be seen in the image. Discrimination, resulting in low accuracy of workpiece defect detection.

An embodiment of the present application provides a workpiece defect detection method. In the first embodiment of the workpiece defect detection method of the present application, referring to FIG. 1 , the workpiece defect detection method includes:

Step S10, obtaining polarized light intensity information of the workpiece to be measured, and determining the total light intensity information, corresponding polarization degree information and corresponding polarization angle information corresponding to the workpiece to be measured according to the polarized light intensity information;

In this embodiment, it should be noted that the polarized light intensity information includes at least one of horizontal, oblique, and vertical light intensity components, and the total light intensity information, the polarization degree information, and the The method of the polarization angle information can be calculated by the Stokes parameter expression and the definition expression corresponding to the polarization degree and the polarization angle, the total light intensity information includes the total light intensity of the workpiece, and the polarization degree information includes the polarization degree of the workpiece, The workpiece polarization angle information includes the workpiece polarization angle.

As an example, step S10 includes: collecting the light intensity in the horizontal polarization direction, the light intensity in the oblique polarization direction, and the light intensity in the vertical polarization direction by using a polarization imaging system; The light intensity in the polarization direction and the light intensity in the vertical polarization direction are calculated to obtain the total light intensity of the workpiece, the light intensity component in the horizontal linear polarization direction, and the light intensity component in the oblique linear polarization direction; The light intensity component in the horizontal linear polarization direction and the light intensity component in the oblique linear polarization direction are calculated to obtain the polarization degree of the workpiece and the polarization angle of the workpiece.

In this embodiment, it should be noted that the light intensity component in the horizontal polarization direction, the light intensity component in the oblique polarization direction, and the light intensity component in the vertical linear polarization direction are the light intensity vectors of each of the linear polarization directions, and the horizontal linear polarization direction The directional light intensity component is the light intensity vector whose vibration plane is limited to the horizontal direction, the light intensity component in the oblique linear polarization direction is the light intensity vector whose vibration plane is limited to the oblique direction, and the vertical linear polarization direction light intensity component The vibration plane of light is limited to the light intensity vector in the vertical direction.

Wherein, the step of determining the total light intensity information corresponding to the workpiece to be measured, the corresponding polarization degree information and the corresponding polarization angle information according to the polarized light intensity information includes:

Step S11, according to the light intensity of each preset polarization direction, determine the total light intensity information, the horizontal linear polarization direction light intensity component and the oblique linear polarization direction light intensity component corresponding to the workpiece to be tested;

Step S12: Determine the polarization degree information and the polarization angle information according to the light intensity component in the horizontal linear polarization direction and the light intensity component in the oblique linear polarization direction.

In this embodiment, as an example, the preset polarization direction may be one or more of four directions of horizontal, oblique, reverse oblique, and vertical, and the light intensity of each preset polarization direction is each The light intensity transmitted by the light transmission axis of the polarizer in the preset polarization direction.

As an example, steps S11 to S12 include: calculating the sum of the light intensity in the horizontal polarization direction and the light intensity in the vertical polarization direction in the light intensity component to obtain the total light intensity of the workpiece; calculating the light intensity in the light intensity component The difference between the light intensity in the horizontal polarization direction and the light intensity in the vertical polarization direction is to obtain the light intensity component in the horizontal linear polarization direction ; by simplifying the expression of the light intensity component in the oblique linear polarization direction, according to the The evaluation of the light intensity in the oblique polarization direction, the light intensity in the horizontal direction, and the light intensity in the vertical polarization direction can be converted into the evaluation of the light intensity in the oblique polarization direction and the light intensity in the reverse oblique polarization direction, and the oblique polarization direction is calculated. The difference between the light intensity and the light intensity in the reverse oblique polarization direction is to obtain the light intensity component in the oblique linear polarization direction; according to the horizontal linear polarization direction light intensity component , the light intensity component in the oblique linear polarization direction and the total light intensity of the workpiece, The degree of polarization of the workpiece is obtained by calculation; the polarization angle of the workpiece is obtained by calculation according to the light intensity component in the horizontal linear polarization direction and the light intensity component in the oblique linear polarization direction.

In this embodiment, it should be noted that the horizontal polarization direction may be a 0° polarization direction, the vertical polarization direction may be a 90° polarization direction, the oblique polarization direction may be a 45° polarization direction, and the reverse polarization direction may be a 45° polarization direction. The oblique polarization direction may be the -45° polarization direction.

As an example, the step of calculating the degree of polarization of the workpiece according to the light intensity component in the horizontal linear polarization direction, the light intensity component in the oblique linear polarization direction, and the total light intensity of the workpiece includes: calculating the horizontal linear polarization The square of the directional light intensity component and the square sum of the light intensity component in the oblique linearly polarized direction, then calculate the arithmetic square root of the square sum, and find the ratio of the arithmetic square root to the total light intensity to obtain the workpiece polarization degree .

As an example, the step of calculating the polarization angle of the workpiece according to the light intensity component in the horizontal linear polarization direction and the light intensity component in the oblique linear polarization direction includes: calculating the light intensity component in the oblique linear polarization direction and the horizontal line For the ratio of the light intensity components in the polarization direction, the arc tangent function value of the ratio is taken, and then half of the arc tangent function value is calculated to obtain the workpiece polarization angle.

As an example, the related expressions for calculating the total light intensity, the light intensity component in the horizontal linear polarization direction, and the light intensity component in the oblique linear polarization direction are as follows:

为水平方向的偏振片透光轴透射光强度，

为倾斜方向的偏振片透光轴透射光强度，

为垂直方向的偏振片透光轴透射光强度，通过化简，上式可以写成：Among them, I is the total light intensity, Q is the light intensity component in the horizontal linear polarization direction, U is the light intensity component in the oblique linear polarization direction,

is the transmitted light intensity along the transmission axis of the polarizer in the horizontal direction,

is the transmitted light intensity of the polarizer transmission axis in the inclined direction,

is the transmitted light intensity of the transmission axis of the polarizer in the vertical direction. By simplification, the above formula can be written as:

As an example, the expression related to calculating the polarization degree and the polarization angle is as follows:

为所述工件偏振度，

为所述工件偏振角。in

is the polarization degree of the workpiece,

is the polarization angle of the workpiece.

Step S20, performing weighted aggregation on the total light intensity information, the polarization degree information and the polarization angle information to obtain a workpiece polarization characteristic image;

In this embodiment, it should be noted that the polarization feature image of the workpiece is a pseudo-color image used to represent the surface feature information of the workpiece to be tested.

As an example, step S20 includes: weighted aggregation of the total light intensity of the workpiece, the polarization degree of the workpiece, and the polarization angle of the workpiece with corresponding weights, respectively, to obtain information contained in the polarization characteristic image of the workpiece.

Wherein, the described total light intensity information, the polarization degree information and the polarization angle information are weighted and aggregated, and the step of obtaining the polarization characteristic image of the workpiece includes;

Step S21, obtaining a weight coefficient, wherein, the weight coefficient is determined according to the material type of the workpiece to be measured;

Step S22, according to the weight coefficient, weighted aggregation is performed on the total light intensity information, the polarization degree information, and the polarization angle information to obtain a polarization characteristic image of the workpiece.

As an example, steps S21 to S22 include: determining a weighting coefficient according to the material type of the workpiece to be tested, where the weighting coefficient includes the total light intensity of the workpiece, the polarization degree of the workpiece, and the polarization angle of the workpiece corresponding to The weighting coefficient of the workpiece, the product of the total light intensity of the workpiece, the polarization degree of the workpiece and the polarization angle of the workpiece and the corresponding weighting coefficient are respectively used as the elements of the feature image matrix to form the workpiece polarization feature image matrix, according to the The polarization characteristic image matrix of the workpiece is obtained, and the polarization characteristic image of the workpiece is obtained.

As an example, the expression for calculating the polarization characteristic image matrix of the workpiece is as follows:

为所述工件偏振特征图像对应的矩阵，

分别为所述总光强

、所述工件偏振度

以及所述工件偏振角

的对应的权重系数，“*”为点乘符号。in,

is the matrix corresponding to the polarization characteristic image of the workpiece,

are the total light intensity

, the workpiece polarization degree

and the workpiece polarization angle

The corresponding weight coefficient of , "*" is the symbol of dot product.

Before the step of performing weighted aggregation on the total light intensity information, the polarization degree information and the polarization angle information to obtain the polarization characteristic image of the workpiece, the method further includes:

Step A10: Obtain the material polarized light intensity information corresponding to the target measurement body, and determine the material total light intensity information, material polarization degree information, and material polarization angle information according to the material polarized light intensity information, the target measurement body and the material to be measured. The material type to which the workpiece belongs is the same;

Step A20, analyzing the sensitivity of the total light intensity information of the material, the polarization degree information of the material, and the polarization angle information of the material to the surface features of the target measurement body, respectively;

Step A30, according to each of the sensitivities, determine the weight coefficients corresponding to the total light intensity information of the material, the polarization degree information of the material, and the polarization angle information of the material respectively.

In this embodiment, it should be noted that the target measurement body may be a flat plate or a workpiece with a certain shape and outline, and the sensitivity is the material, texture, deformation, and smoothness of the surface of the target measurement body. The change of characteristics such as degree is sensitive to the change of the total light intensity information of the material, the polarization degree information of the material, and the polarization angle information of the material, the total light intensity information of the material, the polarization degree information of the material, and the material polarization information. The polarization angle information is used to characterize the surface features of the target measurement body.

As an example, steps A10 to A30 include: collecting polarized light intensity information corresponding to the target measurement body through a polarization imaging system, and determining the corresponding total light intensity information, material polarization degree information, and material polarization angle information. The specific steps are as follows: Step S10 is similar, and will not be repeated here; obtain at least one material polarized light intensity information corresponding to a measurement body of the same material as the target measurement body but with different surface features, and determine the corresponding reference total light intensity information, reference polarization degree information and Refer to polarization angle information. By calculating the changes of the reference total light intensity information, the reference polarization degree information, and the reference polarization angle information relative to the corresponding material total light intensity information, the material polarization degree information, and the material polarization angle information Amplitude, according to calculating the ratio between each of the change amplitudes and the corresponding total light intensity information of the material, the polarization degree information of the material, and the polarization angle information of the material, the change ratio is obtained, and the total light intensity of the material is calculated. Each weight coefficient is obtained from the ratio between the intensity information, the material polarization degree information, and the change ratio corresponding to the material polarization angle information, and the sum of the weight coefficients is 1.

Step S30, performing workpiece edge contour detection on the workpiece polarization feature image to obtain edge contour feature information;

In this embodiment, it should be noted that the method for performing contour detection on the polarization feature image of the workpiece may specifically use a canny edge detection algorithm, and the edge contour feature information may be a grayscale image that has completed edge enhancement.

As an example, step S30 includes: graying the polarization characteristic image of the workpiece to obtain a first polarization integrated image, filtering the first polarization integrated image to obtain a second polarization integrated image; extracting the The gradient magnitude and direction of each pixel point in the second polarization integrated image are obtained, and the gradient value of each pixel point and the direction of each pixel point are obtained; according to the image gradient information corresponding to the second polarization integrated image, the second polarization integrated image Perform non-maximum suppression to eliminate spurious responses brought by edge detection, and obtain a third polarization integrated image; perform double-threshold detection on the third polarization integrated image, and complete the edge in the third polarization integrated image Fitting to obtain the edge contour feature information, so as to complete the polarization integrated image edge enhancement.

Wherein, the step of performing workpiece edge contour detection on the workpiece polarization feature image to obtain edge contour feature information includes:

Step S31, performing grayscale processing on the polarization characteristic image of the workpiece to obtain a first polarization comprehensive image;

Step S32, filtering the first polarization integrated image to obtain a second polarization integrated image;

Step S33, according to the image gradient information corresponding to the second polarization integrated image, non-maximum suppression is performed on the second polarization integrated image to obtain a third polarization integrated image;

Step S34, performing double-threshold detection on the third polarization integrated image to obtain a double-threshold detection result;

Step S35, according to the double-threshold detection result, fit the edge contour of the workpiece to be tested in the third polarization integrated image to obtain the edge contour feature information.

In this embodiment, it should be noted that the grayscale processing is to convert a pseudo-color image composed of pixels of multiple colors into a grayscale image in which each pixel has only one value to represent the color depth, and the filter A Gaussian filter can be used to smooth the image and filter out noise. The gradient magnitude of the pixel point refers to the color depth difference between each pixel point and its neighbors in all directions. The non-maximum value Suppression refers to finding the local maximum value of a pixel point, that is, eliminating non-maximum values, for eliminating the non-filterable noise brought by edge detection, and the double-threshold detection result refers to the determined high threshold and low threshold.

As an example, steps S31 to S34 include: obtaining a grayscale value corresponding to each pixel in the feature image matrix pseudo-color image according to the mapping relationship between each pixel and each parameter in the workpiece polarization feature image , to convert the polarization characteristic image of the workpiece into the first polarization integrated image, which is a grayscale image; filter the first polarization integrated image through a Gaussian filter to complete Filtering and denoising to obtain a Gaussian filtering result, and obtaining the second polarization comprehensive image according to the Gaussian filtering result; performing a maximum value on the second polarization comprehensive image according to the gradient amplitude and direction of each pixel point Suppression to obtain the third polarization integrated image; double-threshold detection is performed on the third polarization integrated image, a gradient amplitude is selected from the gradient amplitudes of each pixel point as a high threshold, and a preset ratio is obtained, according to the The preset ratio and the high threshold are used to obtain a low threshold.

As an example, the step of obtaining the gray value corresponding to each pixel in the feature image matrix pseudo-color image according to the mapping relationship between each pixel and each parameter in the workpiece polarization feature image includes: The product of the total light intensity of the workpiece, the polarization degree of the workpiece, and the polarization angle of the workpiece and the corresponding weight coefficients are respectively substituted into the grayscale formula to obtain the grayscale value of each pixel, and then the said First polarized composite image.

，其中“*”为点乘符号，计算各所述像素点灰度值的表达式如下：As an example, each of the pixel points may be

, where "*" is the dot product symbol, and the expression for calculating the gray value of each pixel point is as follows:

为灰度值，其中的

分别为彩色图像的三通道，通过将

分别代入所述灰度化公式中的

，计算得到所述第一偏振综合图像中各像素点的灰度值。in

is the gray value, where

are the three channels of the color image, respectively, by combining

Substitute into the grayscale formula respectively

, and calculate the gray value of each pixel in the first polarization integrated image.

As an example, the filtering process is performed on the first polarization integrated image through a Gaussian filter to complete filtering and denoising, and the steps of obtaining the Gaussian filtering result include: selecting a rectangular area, and in this embodiment, selecting 3*3 pixel point map, take the center point of the pixel point map as the origin, calculate the square value of the difference between the horizontal and vertical coordinates of each pixel point and the horizontal and vertical coordinates of the origin, and then calculate the horizontal and vertical coordinates of each pixel point. The squared double value of the variance, find the ratio of the square value corresponding to the horizontal and vertical coordinates of each pixel and the corresponding double value, obtain the sum of the ratios corresponding to the horizontal and vertical coordinates of each pixel, and then take the sum of the ratios The opposite number of , taking the opposite number as an index of a natural constant, obtaining the weight of each pixel point, calculating the product of the weight value of the pixel point and the gray value of the pixel point, and obtaining the Gaussian filtering result, and then According to the Gaussian filtering result, the second polarization integrated image is obtained.

As an example, the expression of the two-dimensional Gaussian function is as follows:

为像素点灰度值，

为自然常数e为底数的指数函数，

为原点坐标，

和

为像素点方差，

为像素点坐标。in,

is the gray value of the pixel point,

is an exponential function with the natural constant e as the base,

is the origin coordinate,

and

is the pixel variance,

are pixel coordinates.

As an example, the step of performing maximum suppression on the second polarization integrated image according to the gradient magnitude and direction of each pixel point to obtain the third polarization integrated image includes: extracting the second polarization integrated image. The gradient amplitude and direction of each pixel point in the polarization integrated image, the second polarization integrated image can be regarded as a two-dimensional discrete function in the process of finding the gradient amplitude of each pixel point, and the image gradient amplitude is calculated as the Derive the two-dimensional discrete function, get the derivation result, calculate the inverse of the difference between the gray value of the pixel point and the gray value of the next pixel point in the horizontal direction, and the gray value of the pixel point and the gray value of the next pixel point in the vertical direction Calculate the opposite number of the difference value in the horizontal direction and the opposite number of the difference value in the vertical direction, obtain the image gradient amplitude of each pixel point, and take the sum of the image gradient amplitude The arctangent function value is used to obtain the direction of the image gradient amplitude of each pixel point; according to the image gradient amplitude value of the pixel point and the direction of the image gradient amplitude, the maximum value is performed on the second polarization integrated image. Suppression, a third polarization composite image is obtained.

As an example, the expression for calculating the gradient magnitude of a grayscale image is as follows:

是所述图像梯度幅值，即所述二维离散函数的导数，

是图像像素的灰度值

，

为像素点的坐标，

表示像素点坐标为

的灰度值

。in,

is the image gradient magnitude, that is, the derivative of the two-dimensional discrete function,

is the grayscale value of the image pixel

,

is the coordinates of the pixel point,

Indicates that the pixel coordinates are

the gray value of

.

，

为所述

的方向。Therefore, the gradient direction of each pixel point

,

as stated

direction.

邻域窗口的

方向、45°方向、

方向、135°方向四个方向，各像素点的梯度方向定位属于这四个方向之一，计算四个方向偏导数算子与各方向对应的所述像素点梯度幅值的积，得到四个方向上的各所述像素点梯度幅值一阶偏导数，从各所述像素点四个方向的一阶偏导数从选择出绝对值最大的一阶偏导数，得到各像素点的极大梯度幅值，将所述极大梯度幅度值作为所述像素点的梯度幅值，取所述极大梯度幅度值的方向作为所述像素点的梯度方向。Divide the normal direction of the edge of the image into the current pixel

neighborhood window

direction, 45° direction,

direction, 135° direction four directions, the gradient direction positioning of each pixel belongs to one of these four directions, calculate the product of the four direction partial derivative operators and the gradient amplitudes of the pixel points corresponding to each direction, and obtain four The first-order partial derivative of the gradient amplitude of each pixel point in the direction, the first-order partial derivative with the largest absolute value is selected from the first-order partial derivative in the four directions of each pixel point, and the maximum gradient of each pixel point is obtained. The magnitude of the maximum gradient is taken as the gradient magnitude of the pixel point, and the direction of the maximum gradient magnitude value is taken as the gradient direction of the pixel point.

As an example, the expression of the first-order partial derivative of the gradient magnitude of the pixel point in each direction is calculated as follows:

The expression for calculating the maximum gradient magnitude of each pixel is as follows:

The expression for calculating the gradient direction of each pixel is as follows:

为各方向偏导数算子，其中“*”为点乘符号，max为取各方向梯度幅值的绝对值的最大值，

为像素点梯度幅度，所述偏导数算子选取参见图2，其中（a）为

算子，（b）为45°算子，（c）为

算子，（d）为135°算子。

为max取值所对应的方向。各像素点的梯度方向确定后，将所述梯度幅值

与它梯度方向上相邻像素的梯度幅值比较，若非局部极大值，就把当前像素点的梯度值

设为0；否则，梯度值保留，经过非极大值抑制后的所述第二偏振综合图像为第三偏振综合图像。in,

is the partial derivative operator in each direction, where "*" is the dot product symbol, max is the maximum value of the absolute value of the gradient amplitude in each direction,

is the gradient magnitude of the pixel point, and the selection of the partial derivative operator is shown in Figure 2, where (a) is

operator, (b) is the 45° operator, (c) is

operator, (d) is a 135° operator.

The direction corresponding to the max value. After the gradient direction of each pixel is determined, the gradient magnitude is

Compared with the gradient magnitude of the adjacent pixels in its gradient direction, if it is not a local maximum value, the gradient value of the current pixel is used.

Set to 0; otherwise, the gradient value is retained, and the second polarization integrated image after non-maximum suppression is the third polarization integrated image.

As an example, performing double-threshold detection on the third polarization integrated image, selecting a gradient amplitude as a high threshold, obtaining a preset ratio, and obtaining a low threshold according to the preset ratio and the high threshold The steps include: the second polarization comprehensive image after non-maximum suppression is represented by a gradient amplitude histogram, the histogram will have a peak of the gradient amplitude in the corresponding area near the zero point, and will form on the right side of the zero point. A series of spikes of gradient magnitude.

A gradient amplitude value with the highest proportion is selected from a series of gradient amplitude values on the right side of the zero point of the gradient amplitude value histogram as the high threshold, and a preset ratio is obtained. The preset ratio can be 1/2. The product of the high threshold and the preset ratio obtains the low threshold.

As an example, step S35 includes: dividing each pixel point in the third polarization integrated image into strong edge points and weak edge points by using the high threshold and the low threshold, and dividing each pixel in the third polarization integrated image into strong edge points and weak edge points, The strong edge point and the weak edge point are fitted in the image to enhance the edge of the third polarization integrated image, so as to obtain the edge contour feature information.

Wherein, the step of performing double-threshold detection and fitting on the third polarization integrated image to obtain the edge contour feature information includes:

Step S351, according to the double-threshold detection result, divide each pixel point in the third polarization integrated image into a strong edge point and a weak edge point;

Step S352, by connecting the strong edge point and the continuous weak edge point in the third polarization integrated image, fitting the edge contour of the workpiece to be tested in the third polarization integrated image to obtain the edge contour characteristic information.

In this embodiment, it should be noted that the strong edge point is a point determined as the edge of the third polarization integrated image, the weak edge point is a point that may be the edge of the third polarization integrated image, and the edge The contour feature information is used to characterize the defect feature information of the workpiece to be tested.

As an example, step S351 to step S352 include: judging the relationship between the gradient amplitude of each pixel in the third polarization integrated image and the high threshold and the low threshold, if the gradient amplitude of the pixel is different is less than the high threshold value, the pixel point is marked as a strong edge point; if the gradient amplitude of the pixel point is greater than the low threshold value and less than the high threshold value, the pixel point is marked as a weak edge point; in the third polarization integrated image Connect the strong edge points and the continuous weak edge points in , and discard the isolated weak edge points, so as to fit the edge contour of the workpiece to be tested in the third polarization integrated image, and obtain the edge contour feature information.

Step S40, according to the edge profile feature information, perform workpiece defect detection on the workpiece to be tested, and obtain a workpiece defect detection result.

In this embodiment, it should be noted that the workpiece defect detection may be implemented through a deep learning network, and the workpiece defect detection result is used to represent defect information of the workpiece to be tested.

As an example, step S40 includes: performing feature dimension reduction on the edge contour feature information to obtain edge contour feature information in a one-dimensional vector format, and then inputting the edge contour feature information in the one-dimensional vector format into the workpiece defect detection model , so as to complete the workpiece defect detection of the workpiece to be tested, and obtain the workpiece defect detection result.

Wherein, according to the edge profile feature information, the workpiece defect detection is performed on the workpiece to be tested, and the steps of obtaining the workpiece defect detection result include:

Step S41, performing feature dimension reduction on the edge contour feature information to obtain edge contour feature information samples;

Step S42, by inputting the edge profile feature information sample into a workpiece defect detection model, and performing workpiece defect detection on the workpiece to be tested, to obtain the workpiece defect detection result.

In this embodiment, it should be noted that the workpiece defect detection model may be a deep belief network model.

As an example, steps S41 to S42 include: performing PCA (principal components analysis, principal component analysis) dimension reduction processing on the edge contour feature information. Specifically, the edge contour feature information includes an edge contour image matrix, and obtains The covariance matrix corresponding to the edge contour image matrix; according to the covariance matrix, obtain each eigenvector and each eigenvalue corresponding to the covariance matrix; Arrange to the right into a one-dimensional matrix to obtain the edge contour feature information sample, which is used to represent the edge contour feature information in a one-dimensional vector format; input the edge contour feature information sample into the depth belief In the network model, the workpiece defect detection of the workpiece to be tested is completed, and the workpiece defect detection result is obtained.

Before the step of performing the workpiece defect detection on the workpiece to be tested by inputting the edge profile feature information sample into the workpiece defect detection model, and obtaining the workpiece defect detection result, the method further includes:

Step B10, obtaining the defect detection model of the workpiece to be trained, and extracting the training edge contour feature information corresponding to the training workpiece and the real defect label corresponding to the training workpiece;

Step B20, based on the training edge contour feature information and the real defect label, perform iterative training optimization on the workpiece defect detection model to be trained to obtain the workpiece defect detection model.

In this embodiment, it should be noted that the workpiece defect detection model to be trained is an untrained workpiece defect detection model, and the real defect label is information such as defect type and defect degree of the training workpiece.

As an example, steps B10 to B20 include: acquiring an untrained workpiece defect detection model, extracting edge contour feature information and corresponding real defect labels corresponding to at least one training workpiece; Train a workpiece defect detection model, perform defect detection on the training workpiece, obtain a defect detection result, and then calculate the model of the workpiece defect detection model to be trained based on the degree of difference between the defect detection result and the real defect label loss, and then determine whether the model loss converges, if the model loss converges, the workpiece defect detection model to be trained is used as the workpiece defect detection model, if the model loss does not converge, based on the model loss Calculate the gradient, update the defect detection model of the workpiece to be trained by a preset model update method, and return to the execution step: extracting the training edge contour feature information corresponding to the training workpiece and the real defect label corresponding to the training workpiece, wherein the preset The model update methods include gradient descent method and gradient ascent method, etc., and then achieve the purpose of constructing a workpiece defect detection model based on edge contour feature information, so that the workpiece defect detection model can accurately detect workpiece defects based on edge contour feature information, and improve workpiece defects. Accuracy of defect detection.

The embodiments of the present application provide a workpiece defect detection method, device, electronic device, and readable storage medium. First, polarized light intensity information of the workpiece to be tested is obtained, and the total light intensity corresponding to the workpiece to be tested is determined according to the polarized light intensity information. intensity information, the corresponding polarization degree information and the corresponding polarization angle information, and then weighted aggregation is performed on the total light intensity information, the polarization degree information, and the polarization angle information to obtain a workpiece polarization characteristic image, wherein, due to the The total light intensity information, the polarization degree information and the polarization angle information are all optical polarization information, which are not disturbed by stains on the surface of the workpiece, so that the polarization characteristic image of the workpiece can reflect the real surface profile information that is not disturbed by the stains on the surface of the workpiece, Therefore, according to the polarization feature image of the workpiece, the workpiece edge contour detection is performed on the workpiece polarization feature image to obtain edge contour feature information, and according to the edge contour feature information, workpiece defect detection is performed on the workpiece to be tested, and a workpiece defect detection result is obtained. , which can realize the purpose of workpiece defect detection based on the real surface profile information that is not disturbed by the stains on the workpiece surface, so it overcomes the fact that liquids such as machining coolant and oil will contaminate the workpiece, so that the stains and defects on the workpiece surface cannot be distinguished in the image, resulting in The technical defect of the workpiece defect detection accuracy is low, and the accuracy of the workpiece defect detection is improved.

Embodiment 2

The embodiment of the present application also provides a workpiece defect detection device, the workpiece defect detection device is applied to workpiece defect detection equipment, and the workpiece defect detection device includes:

The light intensity acquisition module is used to obtain the polarized light intensity information of the workpiece to be tested, and according to the polarized light intensity information, determine the total light intensity information, the corresponding polarization degree information and the corresponding polarization angle information corresponding to the workpiece to be tested;

a weighted aggregation module for performing weighted aggregation on the total light intensity information, the polarization degree information and the polarization angle information to obtain a workpiece polarization characteristic image;

a contour detection module, configured to perform contour detection on the feature image matrix in the polarization image information of the workpiece to be tested to obtain edge contour feature information;

The defect detection module is configured to perform workpiece defect detection on the workpiece to be tested according to the edge profile feature information, and obtain a workpiece defect detection result.

Optionally, the light intensity acquisition module is also used for:

According to the light intensity of each preset polarization direction, determine the total light intensity information, the horizontal linear polarization direction light intensity component and the oblique linear polarization direction light intensity component corresponding to the workpiece to be tested;

According to the light intensity component in the horizontal linear polarization direction and the light intensity component in the oblique linear polarization direction, the workpiece polarization degree information and the workpiece polarization angle information are determined.

Optionally, the light intensity acquisition module is also used for:

obtaining a weight coefficient, wherein the weight coefficient is determined according to the material type of the workpiece to be tested;

According to the weight coefficient, weighted aggregation is performed on the total light intensity information, the polarization degree information and the polarization angle information to obtain a polarization characteristic image of the workpiece.

Optionally, the light intensity acquisition module is also used for:

Obtain the material polarized light intensity information corresponding to the target measurement body, and determine the material total light intensity information, material polarization degree information and material polarization angle information according to the material polarized light intensity information. The target measurement body and the workpiece to be measured belong to Material type is the same;

Analyzing the sensitivity of the total light intensity information of the material, the polarization degree information of the material, and the polarization angle information of the material to the surface features of the target measurement body respectively;

According to each of the sensitivities, weight coefficients corresponding to the total light intensity information of the material, the polarization degree information of the material, and the polarization angle information of the material, respectively, are determined.

Optionally, the contour detection module is also used for:

Performing grayscale processing on the polarization characteristic image of the workpiece to obtain a first polarization comprehensive image;

filtering the first polarization integrated image to obtain a second polarization integrated image;

performing non-maximum suppression on the second polarization integrated image according to the image gradient information corresponding to the second polarization integrated image to obtain a third polarization integrated image;

performing double-threshold detection on the third polarization comprehensive image to obtain a double-threshold detection result;

According to the double-threshold detection result, the edge contour of the workpiece to be tested is fitted in the third polarization integrated image to obtain the edge contour feature information.

Optionally, the contour detection module is also used for;

According to the double-threshold detection result, each pixel point in the third polarization integrated image is divided into strong edge points and weak edge points;

By connecting the strong edge points and continuous weak edge points in the third polarization integrated image, and fitting the edge contour of the workpiece to be tested in the third polarization integrated image, the edge contour feature information is obtained.

Optionally, the defect detection module is also used for:

Perform feature dimension reduction on the edge contour feature information to obtain edge contour feature information samples;

By inputting the edge profile feature information sample into the workpiece defect detection model, the workpiece defect detection is performed on the workpiece to be tested, and the workpiece defect detection result is obtained.

The workpiece defect detection device provided by the present invention adopts the workpiece defect detection method in the above-mentioned embodiment, and solves the technical problem of low accuracy of workpiece defect detection. Compared with the prior art, the beneficial effects of the workpiece defect detection device provided by the embodiment of the present invention are the same as the beneficial effects of the workpiece defect detection method provided by the above-mentioned embodiments, and other technical features in the workpiece defect detection device are the same as those of the previous implementation. The disclosed features of the example method are the same, and are not repeated here.

Embodiment 3

An embodiment of the present invention provides an electronic device, the electronic device includes: at least one processor; and a memory connected in communication with the at least one processor; wherein, the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor. The processor executes, so that at least one processor can execute the workpiece defect detection method in the first embodiment.

Referring next to FIG. 3 , it shows a schematic structural diagram of an electronic device suitable for implementing an embodiment of the present disclosure. The electronic devices in the embodiments of the present disclosure may include, but are not limited to, such as mobile phones, notebook computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), in-vehicle terminals (eg, mobile terminals such as in-vehicle navigation terminals), etc., and stationary terminals such as digital TVs, desktop computers, and the like. The electronic device shown in FIG. 3 is only an example, and should not impose any limitation on the function and scope of use of the embodiments of the present disclosure.

As shown in FIG. 3, an electronic device may include processing means (eg, a central processing unit, a graphics processor, etc.), which may be loaded into a random access memory (RAM) according to a program stored in a read only memory (ROM) or from a storage device to execute various appropriate actions and processes. In the RAM, various programs and data necessary for the operation of the electronic device are also stored. The processing device, the ROM, and the RAM are connected to each other through a bus. Input/output (I/O) interfaces are also connected to the bus.

Typically, the following systems can be connected to the I/O interface: input devices including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; including, for example, liquid crystal displays (LCDs), speakers, vibrators output devices, etc.; storage devices including, for example, magnetic tapes, hard disks, etc.; and communication devices. Communication means may allow electronic devices to communicate wirelessly or by wire with other devices to exchange data. While the figures show electronic devices having various systems, it should be understood that not all of the systems shown are required to be implemented or available. More or fewer systems may alternatively be implemented or provided.

In particular, according to embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the method illustrated in the flowchart. In such an embodiment, the computer program may be downloaded and installed from a network via a communication device, or from a storage device, or from a ROM. When the computer program is executed by the processing apparatus, the above-mentioned functions defined in the methods of the embodiments of the present disclosure are executed.

The electronic device provided by the present invention adopts the workpiece defect detection method in the above-mentioned embodiment, and solves the technical problem of low accuracy of workpiece defect detection. Compared with the prior art, the beneficial effects of the electronic device provided by the embodiment of the present invention are the same as those of the workpiece defect detection method provided by the above-mentioned first embodiment, and other technical features in the electronic device are disclosed with the method in the previous embodiment. The features are the same and will not be repeated here.

It should be understood that portions of the present disclosure may be implemented in hardware, software, firmware, or a combination thereof. In the foregoing description of the embodiments, the particular features, structures, materials or characteristics may be combined in any suitable manner in any one or more of the embodiments or examples.

The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention. should be included within the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Embodiment 4

This embodiment provides a computer-readable storage medium having computer-readable program instructions stored thereon, where the computer-readable program instructions are used to execute the workpiece defect detection method in the first embodiment.

The computer-readable storage medium provided by the embodiment of the present invention may be, for example, a U disk, but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system or device, or any combination of the above. More specific examples of computer readable storage media may include, but are not limited to, electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable Programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this embodiment, the computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, system, or device. Program code embodied on a computer-readable storage medium may be transmitted using any suitable medium including, but not limited to, electrical wire, optical fiber cable, RF (radio frequency), etc., or any suitable combination of the foregoing.

The above-mentioned computer-readable storage medium may be included in the electronic device; or may exist alone without being assembled into the electronic device.

The above-mentioned computer-readable storage medium carries one or more programs, and when the above-mentioned one or more programs are executed by the electronic device, the electronic device: obtains the polarized light intensity information of the workpiece to be measured, and determines the polarized light intensity information according to the polarized light intensity information. The total light intensity information, the corresponding polarization degree information and the corresponding polarization angle information corresponding to the workpiece to be tested; weighted aggregation is performed on the total light intensity information, the polarization degree information and the polarization angle information to obtain a polarization characteristic image of the workpiece Performing workpiece edge contour detection on the polarization feature image of the workpiece to obtain edge contour feature information; according to the edge contour feature information, performing workpiece defect detection on the workpiece to be tested to obtain a workpiece defect detection result.

Computer program code for carrying out operations of the present disclosure may be written in one or more programming languages, including object-oriented programming languages—such as Java, Smalltalk, C++, but also conventional Procedural programming language - such as the "C" language or similar programming language. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (eg, using an Internet service provider via Internet connection).

The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code that contains one or more logical functions for implementing the specified functions executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the blocks may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented in dedicated hardware-based systems that perform the specified functions or operations , or can be implemented in a combination of dedicated hardware and computer instructions.

The modules involved in the embodiments of the present disclosure may be implemented in software or hardware. Among them, the name of the module does not constitute a limitation of the unit itself under certain circumstances.

The computer-readable storage medium provided by the present invention stores computer-readable program instructions for executing the above-mentioned workpiece defect detection method, and solves the technical problem of low workpiece defect detection accuracy. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the embodiments of the present invention are the same as those of the workpiece defect detection method provided by the above-mentioned embodiments, which will not be repeated here.

Embodiment 5

The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned workpiece defect detection method when the computer program is executed by a processor.

The computer program product provided by the present application solves the technical problem of low workpiece defect detection accuracy. Compared with the prior art, the beneficial effects of the computer program product provided by the embodiments of the present invention are the same as the beneficial effects of the workpiece defect detection method provided by the above-mentioned embodiments, which will not be repeated here.

The above are only the preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the contents of the description and drawings of the present application, or directly or indirectly applied in other related technical fields , are similarly included within the scope of patent processing of this application.

Claims (8)

1. a workpiece defect detection method, is characterized in that, described workpiece defect detection method comprises: Obtaining the polarized light intensity information of the workpiece to be measured, and determining the total light intensity information, the corresponding polarization degree information and the corresponding polarization angle information corresponding to the workpiece to be tested according to the polarized light intensity information; Weighted aggregation is performed on the total light intensity information, the polarization degree information and the polarization angle information to obtain a workpiece polarization characteristic image; Performing workpiece edge contour detection on the workpiece polarization feature image to obtain edge contour feature information; According to the edge profile feature information, the workpiece defect detection is performed on the workpiece to be tested, and the workpiece defect detection result is obtained; The step of performing weighted aggregation on the total light intensity information, the polarization degree information and the polarization angle information to obtain the polarization characteristic image of the workpiece includes: obtaining a weight coefficient, wherein the weight coefficient is determined according to the material type of the workpiece to be tested; According to the weight coefficient, weighted aggregation is performed on the total light intensity information, the polarization degree information and the polarization angle information to obtain a workpiece polarization characteristic image; Before the step of obtaining the weight coefficient, the method further includes: Obtain the material polarized light intensity information corresponding to the target measurement body, and determine the material total light intensity information, material polarization degree information and material polarization angle information according to the material polarized light intensity information. The target measurement body and the workpiece to be measured belong to Material type is the same; Analyzing the sensitivity of the total light intensity information of the material, the polarization degree information of the material and the polarization angle information of the material to the surface features of the target measurement body respectively; According to each of the sensitivities, weight coefficients corresponding to the total light intensity information of the material, the polarization degree information of the material, and the polarization angle information of the material, respectively, are determined. 2. workpiece defect detection method as claimed in claim 1 is characterized in that, described polarized light intensity information comprises the light intensity of at least one preset polarization direction, and described according to described polarized light intensity information, determine the corresponding workpiece to be tested. The steps of total light intensity information, corresponding polarization degree information and corresponding polarization angle information include: According to the light intensity of each preset polarization direction, determine the total light intensity information, the horizontal linear polarization direction light intensity component and the oblique linear polarization direction light intensity component corresponding to the workpiece to be tested; The polarization degree information and the polarization angle information are determined according to the light intensity component in the horizontal linear polarization direction and the light intensity component in the oblique linear polarization direction. 3. workpiece defect detection method as claimed in claim 1 is characterized in that, described workpiece edge contour detection is carried out to described workpiece polarization characteristic image, and the step that obtains edge contour feature information comprises: Performing grayscale processing on the polarization characteristic image of the workpiece to obtain a first polarization comprehensive image; filtering the first polarization integrated image to obtain a second polarization integrated image; performing non-maximum suppression on the second polarization integrated image according to the image gradient information corresponding to the second polarization integrated image to obtain a third polarization integrated image; performing double-threshold detection on the third polarization comprehensive image to obtain a double-threshold detection result; According to the double-threshold detection result, the edge contour of the workpiece to be tested is fitted in the third polarization integrated image to obtain the edge contour feature information. 4. The workpiece defect detection method according to claim 3, characterized in that, according to the double-threshold detection result, the edge contour of the workpiece to be tested is fitted in the third polarization comprehensive image to obtain the described The steps of edge contour feature information include: According to the double-threshold detection result, each pixel point in the third polarization integrated image is divided into strong edge points and weak edge points; By connecting the strong edge points and continuous weak edge points in the third polarization integrated image, and fitting the edge contour of the workpiece to be tested in the third polarization integrated image, the edge contour feature is obtained information. 5. workpiece defect detection method as claimed in claim 1 is characterized in that, described according to described edge contour feature information, described workpiece defect detection is carried out to described workpiece to be tested, and the step that obtains workpiece defect detection result comprises: Perform feature dimension reduction on the edge contour feature information to obtain edge contour feature information samples; By inputting the edge profile feature information sample into the workpiece defect detection model, the workpiece defect detection is performed on the workpiece to be tested, and the workpiece defect detection result is obtained. 6. A workpiece defect detection device, characterized in that the workpiece defect detection device comprises: an acquisition module, configured to acquire the polarized light intensity information of the workpiece to be measured, and to determine the total light intensity information, the corresponding polarization degree information and the corresponding polarization angle information corresponding to the workpiece to be tested according to the polarized light intensity information; a weighted aggregation module for performing weighted aggregation on the total light intensity information, the polarization degree information and the polarization angle information to obtain a workpiece polarization characteristic image; a contour detection module, which is used to detect the edge contour of the workpiece on the polarization feature image of the workpiece to obtain edge contour feature information; a defect detection module, configured to perform workpiece defect detection on the workpiece to be tested according to the edge profile feature information, and obtain a workpiece defect detection result; Wherein, the weighted aggregation module is also used for: obtaining a weight coefficient, wherein the weight coefficient is determined according to the material type of the workpiece to be tested; According to the weight coefficient, weighted aggregation is performed on the total light intensity information, the polarization degree information and the polarization angle information to obtain a workpiece polarization characteristic image; Before the step of obtaining the weight coefficient, the method further includes: Obtain the material polarized light intensity information corresponding to the target measurement body, and determine the material total light intensity information, material polarization degree information and material polarization angle information according to the material polarized light intensity information. The target measurement body and the workpiece to be measured belong to Material type is the same; Analyzing the sensitivity of the total light intensity information of the material, the polarization degree information of the material and the polarization angle information of the material to the surface features of the target measurement body respectively; According to each of the sensitivities, weight coefficients corresponding to the total light intensity information of the material, the polarization degree information of the material, and the polarization angle information of the material, respectively, are determined. 7. An electronic device, characterized in that the electronic device comprises: at least one processor; and, a memory communicatively coupled to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the execution of any one of claims 1 to 5 The steps of the workpiece defect detection method. 8. A computer-readable storage medium, characterized in that, a program for realizing the workpiece defect detection method is stored on the computer-readable storage medium, and the program for realizing the workpiece defect detection method is executed by a processor to realize the method as claimed in the claims. Steps of the workpiece defect detection method described in any one of 1 to 5.

Images