Protective film intelligent positioning method and system based on machine vision

Abstract

The invention relates to the field of image processing, in particular to an intelligent protective film positioning method and system based on machine vision, and the method comprises the steps: obtaining a template image of a protective film and a surface image of the protective film conveyed in real time, and calculating the possibilities that pixel points in the template image and the surface image are located on the protective film; positioning the feature points by using an ORB algorithm and calculating the feature intensity degree of each pixel point around the feature points, and calculating the feature uniqueness of each feature point based on the positions of the feature points, the feature intensity degree of the surrounding pixel points and the gradient information; the feature points with the feature uniqueness smaller than the average feature uniqueness of all the feature points are screened out, and protective film positioning is conducted on the screened feature points. According to the method, by calculating the feature uniqueness of the feature points, the feature points which are crucial to positioning of the protective film are reserved, meanwhile, the unimportant or wrong feature points are removed, and the matching accuracy of the ORB algorithm is improved.

Description

Technical Field

[0001] The present invention relates to the field of image processing, and more specifically, to a method and system for intelligently positioning a protective film based on machine vision. Background Art

[0002] Protective film is a thin film material used to protect the surface of a product from scratches, contamination and damage. It is widely used in the packaging and transportation of various products, especially in the electronics, automotive, construction and medical industries. In the production process of mobile phone lens protective film, the protective film is usually transported to the vacuum nozzle adsorption area by a vibration plate, and then the vacuum nozzle sucks the protective film and places it on the protective sticker. In this process, the conveying of the vibration plate is random, and the protective film may face any direction. However, the production specification requires that the protective film placed on the protective sticker needs to have the same direction, so the protective film in the vacuum nozzle adsorption area needs to be positioned.

[0003] The existing Chinese patent application document with publication number CN118310419A discloses a method for positioning the center of a workpiece hole based on multi-scale image feature matching, which relates to the field of machine vision measurement technology. The method comprises: acquiring a workpiece hole image; cropping the workpiece hole image to obtain a template image retaining hole edge information; extracting template features from the template image; performing mask processing on the template image to remove areas outside the hole area; acquiring a workpiece image to be matched, and using a pyramid template feature matching algorithm to perform super-resolution multi-scale image feature matching on the template features and the workpiece image to be matched to obtain the hole center position.

[0004] This application document solves the problem that the regular geometric shape detection algorithm cannot detect special shape holes, such as straight slot holes and square holes, and solves the problem that the hole position measurement accuracy decreases when the camera lens resolution decreases, thereby reducing the cost of equipment selection. Currently, the ORB (Oriented Fast and Rotated Brief) algorithm is usually used for feature point matching, and the rotation angle of the protective film is calculated and positioned based on the coordinates of the matching point pair. However, the existing technology of the ORB algorithm screens feature points through machine learning algorithms and non-maximum suppression, lacking a detailed evaluation of the quality of feature points, which may result in feature points with obvious features of the protective film being screened out, and thus may not be able to achieve high-quality positioning of the protective film. Summary of the invention

[0005] In order to solve the problem that the feature points with obvious characteristics of the protective film may be screened out due to the lack of detailed evaluation of the quality of the feature points by screening the feature points using the ORB algorithm, and thus high-quality positioning of the protective film may not be achieved, the present invention provides solutions in the following aspects.

[0006] In the first aspect, a method for intelligently positioning a protective film based on machine vision includes: obtaining a template image of a protective film and a surface image of a real-time conveying protective film, respectively calculating the probability that a pixel in the template image and the surface image is on the protective film; using an ORB algorithm to position the surface image, obtaining a plurality of feature points and corresponding directions, and calculating the feature intensity of each pixel around the feature point; based on the position of the feature point, the feature intensity of the surrounding pixel points, and the gradient information, the feature uniqueness of each feature point is calculated, and the feature points whose feature uniqueness is less than the average feature uniqueness of all feature points are screened out, and the screened feature points are positioned on the protective film; wherein, the feature intensity satisfies the following relationship: ; In the formula, Represents the number of The feature intensity of each pixel, Indicates the extreme difference in the probability that all pixels in the neighborhood of any feature point are on the protective film. Indicates the first The number of times the feature value of a pixel point appears in the feature values ​​of the pixels in the neighborhood of the feature point. Represents the number of pixels in the neighborhood of the feature point. Represented by natural numbers The exponential function of base .

[0007] The effect is that the method can more accurately identify the actual position of the protective film by calculating the probability of the pixel points in the template image and the surface image being on the protective film. This possibility-based evaluation is more sophisticated than the traditional positioning method, can reduce errors, and improve the accuracy of positioning. By calculating the uniqueness of the feature points and screening out those feature points with uniqueness lower than the average, noise and irrelevant feature points are removed, and the feature point set is optimized, making the remaining feature points more suitable for the precise positioning of the protective film.

[0008] Preferably, the possibility that the pixel point is on the protective film includes:

[0009] Taking any pixel as the target pixel, normalizing the grayscale value of the target pixel, and then using an exponential function to decay it to obtain the significance of the target pixel;

[0010] Calculate the absolute difference between the target pixel and the average grayscale value of the pixels in the eight neighborhoods of the target pixel, normalize it, and then use the exponential function to decay it to obtain the contribution of the target pixel;

[0011] The contribution of the target pixel is taken as the exponential power of significance to obtain the possibility that the target pixel is on the protective film.

[0012] The effect is: by analyzing the pixels with important visual features in the image, the boundaries and important feature areas of the protective film can be better identified; the absolute difference between the target pixel and the average grayscale value of the pixels in its eight neighborhoods is calculated, and then normalized and decayed using an exponential function, which can effectively enhance the subtle contrast changes in the image and help capture the details and texture features that may exist on the surface of the protective film, thereby improving the accuracy of positioning.

[0013] Preferably, the possibility that the pixel point is on the protective film also includes:

[0014] Taking any pixel as the target pixel, calculate the absolute difference between the target pixel and the average grayscale value of the pixels in the eight neighborhoods of the target pixel, and the ratio between the difference and the grayscale adjustment parameter to obtain the contribution of the target pixel;

[0015] Use an exponential function to exponentially decay the ratio between the gradient value of the target pixel and the gradient adjustment parameter to obtain the edge strength of the target pixel;

[0016] The product of the contribution degree and the edge strength is normalized and processed as the possibility that the target pixel is on the protective film.

[0017] Preferably, the characteristic uniqueness satisfies the following relationship:

[0018] ;

[0019] In the formula, Indicates the uniqueness of any feature point. Represents the number of feature points in the neighborhood. Represents the direction of any feature point and its surrounding The cosine similarity between the gradient directions of pixels, Represents the number of The feature intensity of each pixel, Represents the number of feature points around any feature point, Represents the total number of feature points in the image, Represented by natural numbers The exponential function of base .

[0020] Preferably, the characteristic uniqueness also satisfies the following relationship:

[0021] ;

[0022] In the formula, Indicates the uniqueness of any feature point. Represents the number of feature points in the neighborhood. Represents the direction of any feature point and its surrounding The cosine similarity between the gradient directions of pixels, Represents the number of The feature intensity of each pixel, Indicates The distance between the pixel and the feature point, Indicates the maximum distance within the neighborhood of the feature point. Represents the number of feature points around any feature point, Represents the total number of feature points in the image, Represented by natural numbers The exponential function of base .

[0023] Preferably, positioning the protective film on the selected feature points includes:

[0024] The ORB algorithm is used to extract key feature points and descriptors from the template image and surface image respectively. The descriptors of the feature points are used to find matching pairs between the feature points of the template image and the surface image using the feature matching method. The rotation angle of the protective film in the vacuum nozzle adsorption area relative to the preset position is calculated, and the protective film is positioned based on the feature points.

[0025] The effect is that by screening out those feature points whose uniqueness is lower than the average, noise and irrelevant feature points are removed so that the feature points are matched according to the conditions. By matching the feature points, the rotation angle of the protective film relative to the preset position is calculated, and the position of the protective film can be accurately adjusted to ensure its correct alignment.

[0026] Preferably, the rotation angle includes:

[0027] According to the laminating and alignment angle algorithm, the rotation angle of the protective film in the vacuum nozzle area relative to the preset orientation is calculated, and the protective film is rotated to the preset orientation based on the vacuum nozzle.

[0028] In a second aspect, a protective film intelligent positioning system based on machine vision includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned protective film intelligent positioning method based on machine vision is implemented.

[0029] The present invention has the following effects:

[0030] 1. The present invention calculates the possibility that a pixel is on the protective film, combines the gray value, gradient information and local texture features of the pixel, and enhances the adaptability of the positioning process to different lighting conditions, background interference and changes in the surface characteristics of the protective film. Through multi-dimensional feature evaluation, the positioning system can maintain high robustness even in complex environments.

[0031] 2. The present invention introduces the calculation of feature intensity and feature uniqueness, which is conducive to more accurate evaluation of the quality of feature points, avoiding the problem of important feature points being erroneously screened due to lack of fine evaluation in the ORB algorithm, retaining feature points that are critical to the positioning of the protective film, while removing unimportant or erroneous feature points, thereby improving the matching accuracy of the ORB algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0033] Figure 1 It is a method flow chart of steps S1 to S3 in the intelligent positioning method of protective film based on machine vision in an embodiment of the present invention.

[0034] Figure 2 It is a schematic diagram of the process of placing a protective film in the protective film intelligent positioning system based on machine vision in an embodiment of the present invention.

[0035] Figure 3 It is a schematic diagram of the protection orientation inside the vibration plate in the protective film intelligent positioning system based on machine vision in an embodiment of the present invention.

[0036] Figure 4 It is a schematic diagram of the standard orientation of protective film production in the protective film intelligent positioning system based on machine vision in an embodiment of the present invention.

[0037] Figure 5 It is a structural block diagram of a protective film intelligent positioning system based on machine vision according to an embodiment of the present invention. DETAILED DESCRIPTION

[0038] 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 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0039] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0040] Specific implementation scenario: In the process of producing mobile phone lens protective film, the protective film is usually transported to the vacuum nozzle adsorption area by a vibration plate, and then the vacuum nozzle absorbs the protective film and places it on the protective sticker. Figure 2In this process, the vibration plate is transported randomly, and the protective film may face any direction. Figure 3 However, the production specification requires that the protective film placed on the protective film must have the same orientation, refer to Figure 4 Therefore, it is necessary to position the protective film in the vacuum nozzle adsorption area to determine the orientation of the protective film in the vacuum nozzle adsorption area, and then unify the orientation of the protective film by using a vacuum nozzle with a rotating function.

[0041] Reference Figure 1 The protective film intelligent positioning method based on machine vision includes steps S1 to S3, which are as follows:

[0042] S1: Acquire a template image of the protective film and a surface image of the protective film being conveyed in real time, and respectively calculate the probability that a pixel point in the template image and the surface image is on the protective film.

[0043] Take any pixel as the target pixel, normalize the gray value of the target pixel, and then use the exponential function to decay it to get the significance of the target pixel.

[0044] Calculate the absolute difference between the target pixel and the average grayscale value of the pixels in the eight neighborhoods of the target pixel, normalize it, and then use the exponential function to decay it to obtain the contribution of the target pixel;

[0045] The contribution of the target pixel is taken as the exponential power of significance to obtain the possibility that the target pixel is on the protective film.

[0046] Specifically, the probability that a pixel point is on the protective film satisfies the following relationship:

[0047] ;

[0048] In the formula, Indicates the image The probability that a pixel is on the protective film, Indicates The gray value of a pixel, Indicates The average gray value of the pixels in the eight neighborhoods of the pixel point, Represented by natural numbers The exponential function of base .

[0049] That is to say, Represents the relative value of the grayscale value of any pixel in the image. Since the protective film border is black and has a low grayscale value, and the image background is bright and has a high grayscale value, usually, The larger the value, the more likely the pixel is to be on the background of the image, and the less likely the pixel is to be on the protective film. The smaller it is, the more likely the pixel is on the protective film in the image.

[0050] It should be noted that the protective film frame has texture lines generated during the production process. The texture lines will reflect light under light and also form highlight areas in the image, affecting the judgment of whether the pixel is on the protective film.

[0051] and then, It indicates the difference between the gray value of any pixel in the image and the gray value of its neighboring pixels. The larger the value, the more likely the pixel is to be a pixel in the reflective area of ​​the texture line on the protective film in the image. The smaller the value, the more likely the pixel is to be a pixel in the normal area inside the protective film or a pixel in the background area. Use the gamma transform Make corrections, The larger the The smaller, right The greater the correction effect, the greater the possibility that the pixel points on the reflective area of ​​the texture lines on the protective film are on the protective film. The smaller the The bigger, right The smaller the correction effect, the possibility of avoiding excessive influence of pixels in the normal area of ​​the protective film on the protective film in the image.

[0052] In addition, another embodiment further includes:

[0053] Taking any pixel as the target pixel, calculate the absolute difference between the target pixel and the average grayscale value of the pixels in the eight neighborhoods of the target pixel, and the ratio between the difference and the grayscale adjustment parameter to obtain the contribution of the target pixel;

[0054] Use an exponential function to exponentially decay the ratio between the gradient value of the target pixel and the gradient adjustment parameter to obtain the edge strength of the target pixel;

[0055] The product of the contribution degree and the edge strength is normalized and processed as the possibility that the target pixel is on the protective film.

[0056] Specifically, the probability that a pixel point is on the protective film satisfies the following relationship:

[0057] ;

[0058] In the formula, Indicates the image The probability that a pixel is on the protective film, Indicates The gray value of a pixel, Indicates The average gray value of the pixels in the eight neighborhoods of the pixel point, represents the grayscale adjustment parameter, Indicates The gradient value of a pixel, represents the gradient adjustment parameter, Represented by natural numbers The exponential function of base .

[0059] That is to say, , , implementation personnel can make adjustments based on specific circumstances.

[0060] Further analysis shows that although the orientation and position of the protective film in the vacuum nozzle adsorption area is random, the shape of the protective film is fixed and the background of image acquisition is also fixed. Therefore, when the protective film is oriented to different positions, the features around each pixel of the protective film in the image are also relatively fixed, and this feature can be used to screen feature points. Therefore, the present invention calculates the feature intensity of each pixel around the feature point based on the possibility that the pixel around each feature point is on the protective film and the feature values ​​of the pixel around the feature point.

[0061] S2: Use the ORB algorithm to locate the surface image, obtain multiple feature points and corresponding directions, and calculate the feature intensity of each pixel around the feature point.

[0062] The intensity of the features, including:

[0063] Taking any feature point as the center, the preset range is used as the neighborhood range of the feature point, and the local binary pattern algorithm of eight-point sampling is used to obtain the feature value of each pixel in the image. The number of occurrences of the feature value of each pixel in the neighborhood range and the total number of pixels in the range are counted, and the extreme difference of the possibility that all pixels in the range of any feature point are on the protective film is obtained;

[0064] Exemplarily, the preset range is , implementers can make adjustments based on actual conditions.

[0065] The frequency of occurrence of the eigenvalue of each pixel in the neighborhood is calculated, and the specificity of the feature point is obtained using exponential decay. The product of the specificity and the range is calculated to obtain the feature intensity of the pixels around any feature point.

[0066] Specifically, the feature intensity satisfies the following relationship:

[0067] ;

[0068] in, Represents the number of The feature intensity of each pixel, Indicates the extreme difference in the probability that all pixels in the neighborhood of any feature point are on the protective film. Indicates the first The number of times the feature value of a pixel point appears in the feature values ​​of the pixels in the neighborhood of the feature point. Represents the number of pixels in the neighborhood of the feature point. Represented by natural numbers The exponential function of base .

[0069] That is to say, The feature point around The frequency of the feature value of a pixel appearing around the feature point. The larger the value, the higher the frequency of the feature value around the feature point. The more likely a pixel point is to have the same characteristics as more pixels, the more common the feature of the pixel point is, and the more likely the pixels around the feature point are. The smaller the feature intensity of a pixel point, the smaller the value is, indicating that the The more likely a pixel is to have the same characteristics as fewer pixels, the less common the feature of the pixel is, and the more likely the pixels around the feature point are to have the same characteristics as fewer pixels. The stronger the feature of each pixel is, the greater the

[0070] To further illustrate, when all pixels around the feature point are on the protective film, the probability of The larger the value, the more likely it is that the feature point is located at the boundary between the protective film and the image background. Since the outer contour of the protective film has obvious boundary features, the pixels around the feature point are more likely to have unique features. When the larger the The stronger the feature of each pixel, The smaller the value, the more likely it is that the feature point is inside the protective film material or in the image background. Since the pixels around the feature point inside the protective film material or in the image background are similar, the pixels around the feature point may not have unique features. The smaller the feature point, the The smaller the feature intensity of each pixel.

[0071] S3: Based on the position of the feature point, the feature intensity of the surrounding pixels and the gradient information, the feature uniqueness of each feature point is calculated, and the feature points whose feature uniqueness is less than the average feature uniqueness of all feature points are screened out, and the protective film is positioned for the screened feature points.

[0072] Specifically, the feature uniqueness satisfies the following relationship:

[0073] ;

[0074] In the formula, Indicates the uniqueness of any feature point. Represents the number of feature points in the neighborhood. Represents the direction of any feature point and its surrounding The cosine similarity between the gradient directions of pixels, Represents the number of The feature intensity of each pixel, Represents the number of feature points around any feature point, Represents the total number of feature points in the image, Represented by natural numbers The exponential function of base .

[0075] That is to say, Indicates the consistency between the pixels around the feature point and the direction of the feature point. The larger the value is, the more consistent the direction of the feature point is with the gradient direction of the surrounding pixels. The more likely the feature point is to have unique features, the greater the uniqueness of the feature point. The smaller the value, the lower the consistency between the direction of the feature point and the gradient direction of the surrounding pixels. The more likely it is that there is no unique feature around the feature point, and the smaller the uniqueness of the feature point. Normalized.

[0076] It should be noted that the first The intensity of the feature of each pixel The larger it is, the more likely the feature points in the neighborhood that contain the pixel point are to have unique features. The smaller it is, the less likely it is that the features of the feature points that contain the pixel in the neighborhood are prominent, and the smaller the uniqueness of the features of the feature points that contain the pixel in the neighborhood is.

[0077] It also needs to be explained. It represents the relative value of the number of feature points around the feature point. Since the shape mutation area of ​​the protective film in the image is more likely to produce feature points, and these feature points have a greater effect on the positioning of the protective film, The larger the value, the more likely the feature is to be in the shape mutation area of ​​the protective film. In order to preserve these feature points well, the gamma transform is used to right Make corrections, The bigger the The smaller, right The greater the correction effect, the easier it is to preserve the feature points in the shape mutation area. The more hours, The bigger, right The smaller the positive effect, the feature points that have little effect on the positioning of the protective film are screened out.

[0078] In addition, another embodiment further includes:

[0079] The uniqueness of the feature also satisfies the following relationship:

[0080] ;

[0081] In the formula, Indicates the uniqueness of any feature point. Represents the number of feature points in the neighborhood. Represents the direction of any feature point and its surrounding The cosine similarity between the gradient directions of pixels, Represents the number of The feature intensity of each pixel, Indicates The distance between the pixel and the feature point, Indicates the maximum distance within the neighborhood of the feature point. Represents the number of feature points around any feature point, Represents the total number of feature points in the image, Represented by natural numbers The exponential function of base .

[0082] That is to say, Indicates the number of The relative size of the distance between the pixel and the feature point, The larger the value, the larger the The greater the influence of the pixel on the feature point, The smaller the feature point, the The influence of the pixel point on the feature point is relatively small, so right Make corrections, The more hours, right The greater the degree of correction, the greater the influence of pixels closer to the feature point on the uniqueness of the feature point. The bigger the right The smaller the degree of correction, the smaller the influence of pixels farther away from the feature point on the characteristic uniqueness of the feature point.

[0083] The ORB algorithm is used to extract key feature points and descriptors from the template image and surface image respectively. The descriptors of the feature points are used to find matching pairs between the feature points of the template image and the surface image using the feature matching method. The rotation angle of the protective film in the vacuum nozzle adsorption area relative to the preset position is calculated, and the protective film is positioned based on the feature points.

[0084] It should be noted that, in this embodiment, the feature matching method is nearest neighbor matching, which is a well-known technology in the art and will not be described in detail.

[0085] Rotation angle, including:

[0086] According to the laminating and alignment angle algorithm, the rotation angle of the protective film in the vacuum nozzle area relative to the preset orientation is calculated, and the protective film is rotated to the preset orientation based on the vacuum nozzle.

[0087] Finally, it should be noted that in the protective film positioning scenario, the height of the vacuum nozzle adsorption area is fixed, the size of the protective film is fixed, and the image acquisition angle is fixed, that is, the size of the protective film in the image is the same. Therefore, in order to improve the efficiency of protective film positioning, the present invention cancels the process of constructing an image pyramid in the existing ORB algorithm to achieve scale invariance of feature points.

[0088] The present invention also provides a protective film intelligent positioning system based on machine vision. Figure 5 As shown, the system includes a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, the protective film intelligent positioning method based on machine vision according to the first aspect of the present invention is implemented.

[0089] The system also includes other components familiar to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0090] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions that may be stored or otherwise maintained by such a computer-readable medium.

[0091] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0092] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A protective film intelligent positioning method based on machine vision, characterized in that: include: Acquire a template image of the protective film and a surface image of the protective film being transported in real time, and respectively calculate the probability that a pixel point in the template image and the surface image is on the protective film; Use the ORB algorithm to locate the surface image, obtain multiple feature points and corresponding directions, and calculate the feature intensity of each pixel point around the feature point; Based on the position of the feature point, the feature intensity of the surrounding pixels and the gradient information, the feature uniqueness of each feature point is calculated, and the feature points whose feature uniqueness is less than the average feature uniqueness of all feature points are screened out, and the protective film is positioned for the screened feature points; The characteristic intensity satisfies the following relationship: ; In the formula, Represents the number of The feature intensity of each pixel, Indicates the extreme difference in the probability that all pixels in the neighborhood of any feature point are on the protective film. Indicates the first The number of times the feature value of a pixel point appears in the feature values ​​of the pixels in the neighborhood of the feature point. Represents the number of pixels in the neighborhood of the feature point. Represented by natural numbers The exponential function of base .

2. The protective film intelligent positioning method based on machine vision according to claim 1 is characterized in that: The possibility that the pixel point is on the protective film includes: Taking any pixel as the target pixel, normalizing the grayscale value of the target pixel, and then using an exponential function to decay it to obtain the significance of the target pixel; Calculate the absolute difference between the target pixel and the average grayscale value of the pixels in the eight neighborhoods of the target pixel, normalize it, and then use the exponential function to decay it to obtain the contribution of the target pixel; The contribution of the target pixel is taken as the exponential power of significance to obtain the possibility that the target pixel is on the protective film.

3. The protective film intelligent positioning method based on machine vision according to claim 1 is characterized in that: The possibility that the pixel point is on the protective film also includes: Taking any pixel as the target pixel, calculate the absolute difference between the target pixel and the average grayscale value of the pixels in the eight neighborhoods of the target pixel, and the ratio between the difference and the grayscale adjustment parameter to obtain the contribution of the target pixel; Use an exponential function to exponentially decay the ratio between the gradient value of the target pixel and the gradient adjustment parameter to obtain the edge strength of the target pixel; The product of the contribution degree and the edge strength is normalized and processed as the possibility that the target pixel is on the protective film.

4. The protective film intelligent positioning method based on machine vision according to claim 1, characterized in that: The characteristic uniqueness satisfies the following relationship: ； In the formula, Indicates the uniqueness of any feature point. Represents the number of feature points in the neighborhood. Represents the direction of any feature point and its surrounding The cosine similarity between the gradient directions of pixels, Represents the number of The feature intensity of each pixel, Represents the number of feature points around any feature point, Represents the total number of feature points in the image, Represented by natural numbers The exponential function of base .

5. The protective film intelligent positioning method based on machine vision according to claim 1 is characterized in that: The characteristic uniqueness also satisfies the following relationship: ； In the formula, Indicates the uniqueness of any feature point. Represents the number of feature points in the neighborhood. Represents the direction of any feature point and its surrounding The cosine similarity between the gradient directions of pixels, Represents the number of The feature intensity of each pixel, Indicates The distance between the pixel and the feature point, Indicates the maximum distance within the neighborhood of the feature point. Represents the number of feature points around any feature point, Represents the total number of feature points in the image, Represented by natural numbers The exponential function of base .

6. The method for intelligent positioning of protective film based on machine vision according to claim 1, characterized in that: The protective film positioning of the selected feature points includes: The ORB algorithm is used to extract key feature points and descriptors from the template image and surface image respectively. The descriptors of the feature points are used to find matching pairs between the feature points of the template image and the surface image using the feature matching method. The rotation angle of the protective film in the vacuum nozzle adsorption area relative to the preset position is calculated, and the protective film is positioned based on the feature points.

7. The protective film intelligent positioning method based on machine vision according to claim 6, characterized in that: The rotation angle includes: According to the laminating and alignment angle algorithm, the rotation angle of the protective film in the vacuum nozzle area relative to the preset orientation is calculated, and the protective film is rotated to the preset orientation based on the vacuum nozzle.

8. The intelligent positioning system of protective film based on machine vision is characterized by: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the protective film intelligent positioning method based on machine vision according to any one of claims 1 to 7 is implemented.

Images