CN116958134B - Plastic film extrusion quality evaluation method based on image processing - Google Patents
Plastic film extrusion quality evaluation method based on image processing Download PDFInfo
- Publication number
- CN116958134B CN116958134B CN202311204130.2A CN202311204130A CN116958134B CN 116958134 B CN116958134 B CN 116958134B CN 202311204130 A CN202311204130 A CN 202311204130A CN 116958134 B CN116958134 B CN 116958134B
- Authority
- CN
- China
- Prior art keywords
- path
- point
- plastic
- starting point
- pixel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 63
- 239000002985 plastic film Substances 0.000 title claims abstract description 54
- 229920006255 plastic film Polymers 0.000 title claims abstract description 54
- 238000001125 extrusion Methods 0.000 title claims abstract description 22
- 238000013441 quality evaluation Methods 0.000 title claims abstract description 16
- 239000004033 plastic Substances 0.000 claims abstract description 212
- 238000001303 quality assessment method Methods 0.000 claims abstract description 7
- 230000008569 process Effects 0.000 claims description 9
- 238000010606 normalization Methods 0.000 claims description 6
- 238000005286 illumination Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of image processing, in particular to a plastic film extrusion quality evaluation method based on image processing, which comprises the following steps: collecting plastic gray images of a plurality of plastic films; obtaining a plurality of plastic areas according to the plastic gray level image; obtaining a local neighborhood difference value, and obtaining local difference probability according to the local neighborhood difference value; obtaining a brightness difference parameter according to the local difference probability; obtaining a path termination point and a path starting point according to the brightness difference parameters; obtaining a path direction according to the path ending point and the path starting point; obtaining an external interference path according to the path direction and the brightness difference parameter; obtaining a brightness threshold according to the external interference path and the brightness parameter; obtaining a reduced plastic region according to the brightness threshold; quality assessment was performed based on the reduced plastic area. The invention ensures that the restored image more accurately represents the real gray distribution of the plastic film, and improves the accuracy of quality assessment.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a plastic film extrusion quality evaluation method based on image processing.
Background
Plastic film extrusion is an important industrial stage in the plastic film production process, and since the quality of products in this industrial stage has a great influence on the quality, performance compliance, cost control and product competitiveness of the following products, an evaluation of the quality of products is required in the plastic film extrusion step, so that the extruded film is ensured to reach the expected quality standard, and good performance and reliable use performance are achieved. The uniformity degree after the plastic film is extruded is an important judging standard, and if the image is not clear due to the interference of external illumination, the accuracy of judging the uniformity degree after the plastic film is extruded is also reduced, so that the image of the plastic film needs to be subjected to gray level reduction to obtain the real gray level distribution of the plastic film, and the quality evaluation is conveniently carried out.
The random path Retinex algorithm can eliminate the influence of illumination to restore the real gray value among all pixel points, but because the selection of the comparison point, the path and the threshold value is random, the self-adaptive selection can not be carried out according to the image characteristics of the plastic film, so that the restored image can not accurately represent the real gray distribution of the plastic film.
Disclosure of Invention
The invention provides a plastic film extrusion quality evaluation method based on image processing, which aims to solve the existing problems: the random path Retinex algorithm is random in selection of comparison points, paths and threshold values, and cannot be selected adaptively according to image features of the plastic film, so that the restored image cannot accurately represent the actual gray distribution of the plastic film.
The plastic film extrusion quality evaluation method based on image processing adopts the following technical scheme:
one embodiment of the present invention provides an image processing-based plastic film extrusion quality evaluation method including the steps of:
collecting plastic gray images of a plurality of plastic films;
dividing the plastic gray image into a plurality of areas to obtain a plurality of plastic areas; obtaining a plurality of local neighborhood difference values of each pixel point in each plastic region; obtaining a plurality of local difference probabilities of each pixel point in each plastic region according to the local neighborhood difference values; obtaining a brightness difference parameter of each pixel point in each plastic area according to the local difference probability;
obtaining a path termination point and a plurality of path starting points of each plastic area according to the brightness difference parameters; obtaining a path direction of each path starting point in each plastic area according to the path ending point and the path starting point; obtaining a plurality of external interference paths of the starting point of each path in each plastic area according to the path direction and the brightness difference parameters;
Obtaining a brightness threshold value of each external interference path according to the external interference path and the brightness difference parameter; reducing the starting point of each path in each plastic region according to the brightness threshold value to obtain each reduced plastic region; quality assessment was performed based on the reduced plastic area.
Preferably, the method for obtaining the local neighborhood difference values of each pixel point in each plastic region includes the following specific steps:
marking any pixel point in any plastic region on any plastic gray level image as a target pixel point, taking the target pixel point as a center, acquiring a gray level value of each pixel point in eight adjacent regions of the target pixel point, and marking the gray level value as a local neighborhood gray level value of the target pixel point; for any local neighborhood gray value of the target pixel, marking the absolute value of the difference value between the gray value of the target pixel and the local neighborhood gray value as the local neighborhood difference value of the target pixel; obtaining all local neighborhood difference values of target pixel points in a plastic region;
and obtaining a plurality of local neighborhood difference values of each pixel point in each plastic region.
Preferably, the obtaining the local difference probabilities of each pixel point in each plastic region according to the local neighborhood difference value includes the specific steps of:
For any one plastic region, acquiring the occurrence frequency of each local neighborhood difference value of each pixel point in the plastic region from all local neighborhood difference values of all pixel points in the plastic region, and marking the occurrence frequency as the local difference probability of each pixel point in the plastic region;
and acquiring a plurality of local difference probabilities of each pixel point in each plastic region.
Preferably, the method for obtaining the brightness difference parameter of each pixel point in each plastic area according to the local difference probability includes the following specific steps:
any pixel point in any one plastic area is marked as a first target pixel point, in the formula,a luminance difference parameter representing a first target pixel point; />Representing a first objectThe number of pixel points contained in the plastic region to which the pixel points belong; />A gray value representing a first target pixel point; />Representing the +.>Gray values of the individual pixels; />Representing preset super parameters; />A mean value representing all local difference probabilities of the first target pixel point;
and acquiring the brightness difference parameters of each pixel point in each plastic region.
Preferably, the method for obtaining the path termination point and the path start points of each plastic region according to the brightness parameter includes the following specific steps:
For any one plastic region, marking the pixel point with the largest brightness difference parameter in the plastic region as a path termination point of the plastic region; marking each pixel point except the path ending point in the plastic area as a path starting point of the plastic area;
and acquiring a path ending point and a plurality of path starting points of each plastic area.
Preferably, the method for obtaining the path direction of each path starting point in each plastic area according to the path ending point and the path starting point includes the following specific steps:
a path ending point for any one of the plastic regions and any one of the path starting points;
if the path ending point is positioned in the upper right area of the path starting point, the path direction of the path starting point is the upper right direction;
if the path ending point is positioned in the right lower area of the path starting point, the path direction of the path starting point is the right lower direction;
if the path ending point is positioned in the upper left area of the path starting point, the path direction of the path starting point is the upper left direction;
if the path ending point is positioned in the left lower area of the path starting point, the path direction of the path starting point is the left lower direction;
if the path ending point is positioned in the horizontal right direction of the path starting point, the path direction of the path starting point is horizontal right;
If the path ending point is positioned in the horizontal left direction of the path starting point, the path direction of the path starting point is horizontal left;
if the path ending point is positioned in the vertical upward direction of the path starting point, the path direction of the path starting point is vertical upward;
if the path ending point is positioned in the vertically downward direction of the path starting point, the path direction of the path starting point is vertically downward;
and acquiring the path direction of the starting point of each path in each plastic area.
Preferably, the obtaining a plurality of external interference paths of each path starting point in each plastic area according to the path direction and the brightness parameter includes the following specific methods:
for any path starting point with the path direction being horizontal to the right, taking the pixel points in the eight neighborhood of the path starting point and horizontal to the right as the first path point, stopping traversing until the latest acquired path point is the path ending point, and marking the image area occupied by all the path points as the external interference path of the path starting point; if the path direction of the path starting point is horizontal left, vertical upward or vertical downward, taking pixel points corresponding to the path directions in eight adjacent domains as first path points, and acquiring subsequent path points according to the path directions until traversing to a path ending point, so as to obtain external interference paths of which all the path directions are horizontal right, horizontal left, vertical upward and vertical downward path starting points;
For any one path starting point with the path direction being the right lower direction, in the eight adjacent areas of the path starting point, the pixel point of the path starting point at the horizontal right is marked as the horizontal right side starting point of the path starting point, the pixel point of the path starting point at the horizontal left is marked as the horizontal left side starting point of the path starting point, the pixel point of the path starting point at the vertical upper side of the path starting point is marked as the vertical upper side starting point of the path starting point, and the pixel point of the path starting point at the vertical lower side of the path starting point is marked as the vertical lower side starting point of the path starting point;
for a horizontal left starting point of the path starting point, taking the horizontal left starting point as a first path point, and taking a pixel point vertically downward in eight adjacent areas of the first path point as a second path point; if the brightness difference parameter of the pixel points in the eight neighborhood of the second path point to the right is more than or equal to the brightness difference parameter of the pixel points in the vertical direction, marking the pixel points in the eight neighborhood of the second path point to the right as a third path point; if the brightness difference parameter of the pixel points in the eight neighborhood of the second path point, which is horizontally right, is smaller than the brightness difference parameter of the pixel points in the vertical direction, the pixel points in the eight neighborhood of the second path point, which are vertically downward, are marked as a third path point; if the brightness difference parameter of the pixel points in the eight neighborhood of the third path point to the right is more than or equal to the brightness difference parameter of the pixel points in the vertical direction, marking the pixel points in the eight neighborhood of the third path point to the right as a fourth path point; if the brightness difference parameter of the pixel points in the eight neighborhood of the third path point, which is horizontally right, is smaller than the brightness difference parameter of the pixel points in the vertical direction, the pixel points in the eight neighborhood of the third path point, which are vertically downward, are marked as fourth path points;
In the subsequent path point acquisition process, if the path point acquired by the latest traversal and the path ending point are in the same horizontal line, the subsequently acquired path point is a horizontal rightward pixel point in the eight neighborhood of the previous path point; if the latest traversing acquired path point and the path ending point are in the same vertical line, the subsequently acquired path point is a vertically downward pixel point in the eight neighborhood of the previous path point; if the latest traversed and acquired path point and the path ending point are still not in the same horizontal line or the same vertical line, continuing to acquire the next path point according to the brightness difference parameter until the latest traversed and acquired path point and the path ending point are in the same horizontal line or the same vertical line, acquiring the subsequent path point, stopping traversing until the latest acquired path point is the path ending point, and recording the image area occupied by all the path points as an interference path of a horizontal left starting point;
for the case that a horizontal right starting point, a vertical upper starting point and a vertical lower starting point of the path starting point are respectively used as a first path point, acquiring a next path point according to the brightness difference parameters of two pixel points in the eight adjacent areas corresponding to the path direction and whether the brightness difference parameters are the path starting points or not until interference paths respectively corresponding to the first starting points under the three conditions are finally obtained; respectively marking an interference path of a horizontal left starting point, an interference path of a horizontal right starting point, an interference path of a vertical upper starting point and an interference path of a vertical lower starting point as an external interference path of which the path direction is a path starting point in the lower right direction, and acquiring a plurality of external interference paths of which each path direction is a path starting point in the lower right direction;
For the path starting points of which the path directions are the upper right direction, the upper left direction or the lower left direction, four interference paths of each path starting point are obtained, and the four interference paths of each path starting point are used as external interference paths of each path starting point;
and acquiring a plurality of external interference paths of the starting point of each path in each plastic area.
Preferably, the method for obtaining the brightness threshold of each external interference path according to the external interference path and the brightness difference parameter includes the following specific steps:
for any external disturbance path at the start point of any path in any plastic region, in the formula,shading threshold representing external disturbance pathA value; />Representing preset super parameters; />Representing the number of pixels contained within the plastic region; />Indicating the%>A lighting difference parameter of each pixel point; />Representing the maximum value of the brightness difference parameters of all pixel points in the plastic area; />Representing the minimum value of the brightness difference parameters of all pixel points in the plastic area; />Representing preset super parameters; />Representing the number of path points contained in the external interference path; />Representing the average value of the brightness difference parameters of all pixel points in the plastic area; />An exponential function that is based on a natural constant;
And acquiring a brightness threshold value of each external interference path.
Preferably, the reducing the starting point of each path in each plastic region according to the brightness threshold to obtain each reduced plastic region includes the following specific steps:
for any external interference path of any path starting point in any plastic region, obtaining the relative brightness relationship of the external interference path by using a Retinex algorithm according to the brightness threshold of the external interference path; acquiring the relative brightness relationship of all external interference paths of the path starting points in the plastic area;
the method comprises the following steps of restoring the path starting point according to the relative brightness relation of all external interference paths of the path starting point in the plastic area:
in the method, in the process of the invention,representing the gray value of the path after the initial point is restored; />Gray values representing the path termination points within the plastic region; />The average value of the relative brightness relationship of all external interference paths at the starting point of the path is represented;
reducing all path starting points in the plastic area, and marking the reduced plastic area as a reduced plastic area;
each reduced plastic region is acquired.
Preferably, the quality evaluation is performed according to the reduced plastic region, and the specific method comprises the following steps:
For any plastic gray level image, the accumulated sum of gray level values of all pixel points in each reduced plastic area is recorded as the initial uniformity degree of each reduced plastic area, the initial uniformity degree of all the reduced plastic areas is subjected to linear normalization, and the normalized initial uniformity degree is recorded as the uniformity degree of each reduced plastic area;
marking a preset uniformity degree threshold as T2, and marking the plastic gray level image as a plastic gray level image with unqualified quality if the uniformity degree of any one of the reduced plastic areas is larger than T2; if any reduced plastic area does not exist, the uniformity degree of the reduced plastic area is larger than T2, and no treatment is carried out on the plastic gray image;
and obtaining all plastic gray-scale images with unqualified quality.
The technical scheme of the invention has the beneficial effects that: obtaining a brightness difference parameter of each pixel point in each plastic region through region division, obtaining a path ending point and a plurality of path starting points according to the brightness difference parameter, obtaining a path direction according to the path ending point and the plurality of path starting points, obtaining a plurality of external interference paths according to the path direction and the brightness difference parameter, obtaining a light and dark threshold according to the external interference paths and the brightness difference parameter, and performing quality assessment according to the light and dark threshold; compared with the prior art, the method has the defects that the selection of the comparison point, the path and the threshold value is random and the self-adaptive selection cannot be carried out according to the image characteristics of the plastic film; the brightness difference parameter of the pixel point better represents the influence degree of external illumination on the pixel point, and the comparison point is more accurately determined; the brightness threshold value better represents the degree of the influence of external interference on the external interference path and the strength to be eliminated; the acquired comparison points, paths and threshold values are more in accordance with the image characteristics of the plastic film, so that the restored image can more accurately represent the real gray distribution of the plastic film, and the accuracy of quality assessment is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart showing the steps of the image processing-based plastic film extrusion quality evaluation method of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description refers to the specific implementation, structure, characteristics and effects of the plastic film extrusion quality evaluation method based on image processing according to the invention, which are provided by the invention, with reference to the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the plastic film extrusion quality evaluation method based on image processing provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for evaluating extrusion quality of a plastic film based on image processing according to an embodiment of the present invention is shown, the method includes the following steps:
step S001: and collecting plastic gray images of a plurality of plastic films.
It should be noted that, the random path Retinex algorithm may eliminate the influence of light to restore the real gray value between the pixel points, but because the selection of the comparison point, the path and the threshold value is random, the self-adaptive selection cannot be performed according to the image characteristics of the plastic film, so that the restored image cannot accurately represent the real gray distribution of the plastic film.
Specifically, in order to implement the plastic film extrusion quality evaluation method based on image processing provided in this embodiment, firstly, a gray image of a plastic film needs to be collected, and the specific process is as follows: shooting plastic films produced by a working plastic film extruder by using an industrial camera, shooting every 5 seconds for 1 hour, and obtaining a plurality of plastic film images; and carrying out graying treatment on each plastic film image to obtain a gray image, and recording the gray image as a plastic gray image of each plastic film image. The graying process is a known technique, and this embodiment will not be described.
So far, the plastic gray level images of a plurality of plastic films are obtained by the method.
Step S002: dividing the plastic gray image into a plurality of areas to obtain a plurality of plastic areas; obtaining a plurality of local neighborhood difference values of each pixel point in each plastic region; obtaining a plurality of local difference probabilities of each pixel point in each plastic region according to the local neighborhood difference values; and obtaining the brightness difference parameter of each pixel point in each plastic region according to the local difference probability.
It should be noted that, in general, the more uniform the plastic film after extrusion, the better the quality of the plastic film, and the better the quality of the finished product manufactured later; in the plastic gray level image, the gray level distribution corresponding to the plastic film areas with different thicknesses is different, so that the uniformity degree of the gray level distribution can well represent the uniformity degree of the plastic film; however, since the gradation information in the plastic gradation image is disturbed by the influence of light, the reflected gradation distribution is not actually the gradation distribution of the plastic film. Therefore, the embodiment restores the gray value of the pixel point in the plastic gray image through the relation among the pixel points and the gray characteristic information, so that the plastic gray image reflects the real gray information of the plastic film.
Specifically, a plastic area number T1 is preset, where the embodiment is described by taking t1=9 as an example, and the embodiment is not specifically limited, where T1 may be determined according to specific implementation conditions; taking any plastic gray level image as an example, dividing the plastic gray level image into T1 plastic areas uniformly; taking any pixel point in any plastic region on the plastic gray level image as an example, taking the pixel point as a center, acquiring a gray level value of each pixel point in eight adjacent regions of the pixel point, and recording the gray level value as a local adjacent region gray level value of the pixel point; taking any local neighborhood gray value of the pixel point as an example, and recording the absolute value of the difference value between the gray value of the pixel point and the local neighborhood gray value as the local neighborhood difference value of the pixel point; obtaining all local neighborhood differences of the pixel points in the plastic area, and obtaining all local neighborhood differences of all pixel points in the plastic area; each pixel point corresponds to a plurality of local neighborhood gray values, and each pixel point corresponds to a plurality of local neighborhood difference values; if the pixel points are located at the boundary of the plastic gray image, the eight adjacent areas exceed the image boundary, and then the pixel points which can be actually obtained in the eight adjacent areas of the pixel points are analyzed and processed.
Further, in all local neighborhood differences of all pixel points in the plastic region, obtaining the occurrence frequency of each local neighborhood difference of each pixel point in the plastic region, and marking the occurrence frequency as the local difference probability of each pixel point in the plastic region; and obtaining all local difference probabilities of all pixel points in the plastic region. Each pixel point corresponds to a plurality of local difference probabilities, and one local neighborhood difference value corresponds to one local difference probability.
Further, taking any pixel point in the plastic area as an example, obtaining a brightness difference parameter of the pixel point according to a plurality of local difference probabilities of the pixel point; the calculation method of the brightness difference parameter of the pixel point comprises the following steps:
in the method, in the process of the invention,a lighting difference parameter indicating the pixel point; />Representing the number of pixel points contained in the plastic region to which the pixel points belong; />A gray value representing the pixel; />Indicating the +.>Gray values of the individual pixels; />Representing hyper-parameters, preset in this example->For preventing denominator from being 0; />A mean value representing all local difference probabilities for the pixel point; />Representing the number of pixels except the pixel in the plastic region to which the pixel belongs; / >The degree of variation in the gradation value of the pixel is represented. The smaller the gray level deviation degree of the pixel point is, the larger the local difference frequency of the pixel point in the adjacent area is, the gray level value of the pixel point is more close to the gray level value uniformly distributed in the plastic area, which indicates that the more uniform the gray level value of the pixel point is distributed in the plastic area, the larger the brightness difference parameter is, and the smaller the degree of the influence of external illumination on the pixel point is reflected. Acquiring brightness difference parameters of all pixel points in the plastic area; and acquiring the brightness difference parameters of all pixel points in all plastic areas.
So far, the brightness difference parameters of all pixel points in all plastic areas are obtained through the method.
Step S003: obtaining a path termination point and a plurality of path starting points of each plastic area according to the brightness difference parameters; obtaining a path direction of each path starting point in each plastic area according to the path ending point and the path starting point; and obtaining a plurality of external interference paths of the starting point of each path in each plastic area according to the path direction and the brightness difference parameters.
It should be noted that the conventional random path Retinex algorithm is random for the selection of two reference points and the selection of a path between two points. Under ideal conditions, the color characteristics of the surface of the plastic film are single, the color distribution is uniform, and the influence among pixel points is small; under the influence of external illumination, the color characteristics of the plastic surface are more various, the color distribution is in irregular distribution, and the influence among pixel points is larger; meanwhile, the brightness difference parameter can reflect the influence of external illumination on the pixel point to a certain extent, so that two reference points and a path between the two points can be determined according to the brightness difference parameter.
Specifically, taking any plastic area as an example, and marking a pixel point with the largest brightness difference parameter in the plastic area as a path termination point of the plastic area; taking any pixel point except a path ending point in the plastic area as an example, and marking the pixel point as a path starting point of the plastic area; judging the direction relation between the path starting point and the path ending point, and acquiring the path direction of the path starting point, wherein the specific method is as follows:
if the path ending point is positioned in the upper right area of the path starting point, the path direction of the path starting point is 45 degrees upwards right and is marked as the upper right direction;
if the path ending point is positioned in the right lower area of the path starting point, the path direction of the path starting point is the direction of 45 degrees below right and is marked as the direction below right;
if the path ending point is positioned in the upper left area of the path starting point, the path direction of the path starting point is the direction of 45 degrees below left and is marked as the direction below left;
if the path ending point is positioned in the left lower area of the path starting point, the path direction of the path starting point is 45 degrees left-biased upwards and is marked as the left upper direction;
if the path ending point is positioned in the horizontal right direction of the path starting point, the path direction of the path starting point is horizontal right;
If the path ending point is positioned in the horizontal left direction of the path starting point, the path direction of the path starting point is horizontal left;
if the path ending point is positioned in the vertical upward direction of the path starting point, the path direction of the path starting point is vertical upward;
if the path ending point is positioned in the vertically downward direction of the path starting point, the path direction of the path starting point is vertically downward; and acquiring the path direction of the starting point of each path in each plastic area.
Further, taking any one path direction as a path starting point in a horizontal right direction as an example, the path ending point and the path starting point are on the same horizontal line, and the path ending point is on the right side of the path starting point, taking a pixel point in an eight-neighborhood of the path starting point in the horizontal right direction as a first path point, stopping traversing until the newly obtained path point is the path ending point, and marking the image area occupied by all the path points as an external interference path of the path starting point; similarly, if the path direction of the path starting point is horizontal left, vertical upward or vertical downward, the pixel points corresponding to the path directions in the eight neighborhoods are taken as first path points, and subsequent path points are acquired according to the path directions until the path ending point is traversed, so that external interference paths of which all the path directions are horizontal right, horizontal left, vertical upward and vertical downward are obtained.
Further, taking a path starting point with any one path direction as a lower right direction as an example, in the eight adjacent areas of the path starting point, marking a pixel point of the path starting point in a horizontal right direction as a horizontal right side starting point of the path starting point, marking a pixel point of the path starting point in a horizontal left direction as a horizontal left side starting point of the path starting point, marking a pixel point of the path starting point in a vertical upward direction as a vertical upper side starting point of the path starting point, and marking a pixel point of the path starting point in a vertical downward direction as a vertical lower side starting point of the path starting point; judging the path direction of the path starting point, and acquiring a plurality of external interference paths of the path starting point, wherein the specific method comprises the following steps:
taking a horizontal left starting point of a path starting point as an example, taking the horizontal left starting point as a first path point, and taking a pixel point vertically downward in eight adjacent areas of the first path point as a second path point; if the brightness difference parameter of the pixel points in the eight neighborhood of the second path point to the right is more than or equal to the brightness difference parameter of the pixel points in the vertical direction, marking the pixel points in the eight neighborhood of the second path point to the right as a third path point; if the brightness difference parameter of the pixel points in the eight neighborhood of the second path point, which is horizontally right, is smaller than the brightness difference parameter of the pixel points in the vertical direction, the pixel points in the eight neighborhood of the second path point, which are vertically downward, are marked as a third path point; if the brightness difference parameter of the pixel points in the eight neighborhood of the third path point to the right is more than or equal to the brightness difference parameter of the pixel points in the vertical direction, marking the pixel points in the eight neighborhood of the third path point to the right as a fourth path point; if the brightness difference parameter of the pixel points in the eight neighborhood of the third path point, which is horizontally right, is smaller than the brightness difference parameter of the pixel points in the vertical direction, the pixel points in the eight neighborhood of the third path point, which are vertically downward, are marked as fourth path points; similarly, if the newly traversed acquired path point and the path ending point are in the same horizontal line, the subsequently acquired path point is a horizontal rightward pixel point in the eight neighborhood of the previous path point; if the latest traversing acquired path point and the path ending point are in the same vertical line, the subsequently acquired path point is a vertically downward pixel point in the eight neighborhood of the previous path point; and similarly, if the latest traversed and acquired path point and the path ending point are not in the same horizontal line or vertical line, continuing to acquire the next path point according to the brightness difference parameter until the latest traversed and acquired path point and the path ending point are in the same horizontal line or vertical line, acquiring the subsequent path point according to the method, stopping traversing until the latest acquired path point is the path ending point, and recording the image area occupied by all the path points as an interference path of a horizontal left starting point;
At this time, the horizontal left starting point of the path starting point is taken as the first path point, the second path point is acquired according to the path direction, and the path starting point is avoided at the same time, namely, the path direction of the path starting point is the right lower direction, the next path point needs to be selected from two pixel points horizontally right and vertically downward in eight adjacent domains, and if one pixel point is the path starting point, the other pixel point is selected as the next path point; if the two pixel points are not the path starting points, taking the largest pixel point in the brightness difference parameters of the two pixel points as the next path point, and selecting the pixel points to the right horizontally under the condition of equality; according to the method, the next path point can be obtained according to the brightness difference parameters of two pixel points in eight adjacent areas corresponding to the path direction and whether the brightness difference parameters are the path starting points or not until the interference paths corresponding to the first starting points respectively under three conditions are finally obtained; and respectively marking the interference path of the horizontal left starting point, the interference path of the horizontal right starting point, the interference path of the vertical upper starting point and the interference path of the vertical lower starting point as one external interference path of which the path direction is the path starting point in the lower right direction, and acquiring a plurality of external interference paths of which each path direction is the path starting point in the lower right direction according to the method.
Further, for the path starting point with the path direction being the upper right direction, the upper left direction or the lower left direction, after the path starting point is obtained according to the method and is used for obtaining four first path points of four interference paths, for the upper right direction, two pixel points in eight adjacent domains, namely the two pixel points in the horizontal direction and the vertical direction, need to be selected for the next path point; for the upper left direction, two pixel points in the eight neighborhood, which are horizontally left and vertically upwards, are required to be selected for the next path point; for the lower left direction, two pixel points horizontally leftwards and vertically downwards in the eight neighborhood are required to be selected for the next path point; if the two pixel points to be selected have the path starting point, taking the pixel point which is not the path starting point as the next path point according to the method; if the two pixel points to be selected do not have the path starting point, taking the largest pixel point in the brightness difference parameters of the two pixel points as the next path point, and selecting the pixel point in the horizontal direction (horizontal right or horizontal left according to the path direction) as the next path point if the brightness difference parameters are equal to each other; according to the method, for the path starting points of which the path directions are the right upper direction, the left upper direction or the left lower direction, the corresponding interference paths under the condition of the four first path points of each path starting point are obtained, and the four interference paths of each path starting point are used as the external interference paths of each path starting point, so that a plurality of external interference paths of each path starting point in each plastic area are obtained.
So far, all external interference paths of the starting points of all paths in all plastic areas are obtained through the method.
Step S004: obtaining a brightness threshold value of each external interference path according to the external interference path and the brightness difference parameter; reducing the starting point of each path in each plastic region according to the brightness threshold value to obtain each reduced plastic region; quality assessment was performed based on the reduced plastic area.
Specifically, taking any external interference path of any one path starting point in any one plastic area as an example, and obtaining a brightness threshold of the external interference path according to the brightness difference parameter of each pixel point in the plastic area and the path point contained in the external interference path; the method for calculating the brightness threshold of the external interference path comprises the following steps:
in the method, in the process of the invention,a bright-dark threshold value representing the external interference path; />Indicating hyper-parameters, preset +.>；Representing the number of pixels contained within the plastic region; />Indicating the>A lighting difference parameter of each pixel point; />Representing the maximum value of the brightness difference parameters of all pixel points in the plastic area; />Representing the minimum value of the brightness difference parameters of all pixel points in the plastic area; / >Indicating hyper-parameters, preset +.>For preventing denominator from being 0; />Representing the number of path points contained in the ambient interference path; />Representing the average value of the brightness difference parameters of all pixel points in the plastic area; />An exponential function that is based on a natural constant; example use->The functions represent inverse proportion relation and normalization processing, and an implementer can select the inverse proportion functions and the normalization functions according to actual conditions; />Indicating the degree of external influence to which the plastic region is subjected; />For taking->Normalization. If the brightness threshold value of the external interference path is larger, the degree that the external interference path is influenced by external interference is larger, and the influence of the external interference path is required to be eliminated, so that the elimination strength is larger.
Further, a Retinex algorithm is utilized to obtain the relative brightness relationship of the external interference path according to the brightness threshold of the external interference path; acquiring the relative brightness relationship of all external interference paths of the path starting point in the plastic region; the bright-dark threshold is used as the threshold of the Retinex algorithm, and the acquisition of the adjacent bright-dark relationship is a well-known content of the Retinex algorithm, which is not described in this embodiment.
Further, according to the relative brightness relationship of all external interference paths of the path starting point in the plastic region, the path starting point is restored, and the specific method is as follows:
In the method, in the process of the invention,representing the gray value of the path after the initial point is restored; />A gray value representing a path termination point within the plastic region; />And the average value of the relative brightness relations of all external interference paths at the starting point of the path is represented. If the gray value of the path starting point after the restoration is larger, the degree that the path starting point is influenced by external interference is larger, and the influence needs to be eliminated when the path starting point is reflected.
Further, reducing all path starting points in the plastic area, marking the reduced plastic area as a reduced plastic area, marking the accumulated sum of gray values of all pixel points in each reduced plastic area as the initial uniformity degree of each reduced plastic area, carrying out linear normalization on the initial uniformity degree of all the reduced plastic areas, and marking the normalized initial uniformity degree as the uniformity degree of each reduced plastic area.
Further, a uniformity threshold T2 is preset, where the embodiment is described by taking t2=0.5 as an example, and the embodiment is not specifically limited, where T2 may be determined according to the specific implementation situation; if the uniformity degree of any one of the reduced plastic areas is larger than the uniformity degree threshold T2, marking the plastic gray-scale image as a plastic gray-scale image with unqualified quality; if the uniformity degree of any one of the non-existing reduced plastic areas is larger than the uniformity degree threshold T2, no processing is carried out on the plastic gray image. And obtaining all plastic gray-scale images with unqualified quality.
So far, all plastic gray-scale images with unqualified quality are obtained through the method.
This embodiment is completed.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (7)
1. The plastic film extrusion quality evaluation method based on image processing is characterized by comprising the following steps of:
collecting plastic gray images of a plurality of plastic films;
dividing the plastic gray image into a plurality of areas to obtain a plurality of plastic areas; obtaining a plurality of local neighborhood difference values of each pixel point in each plastic region; obtaining a plurality of local difference probabilities of each pixel point in each plastic region according to the local neighborhood difference values; obtaining a brightness difference parameter of each pixel point in each plastic area according to the local difference probability;
obtaining a path termination point and a plurality of path starting points of each plastic area according to the brightness difference parameters; obtaining a path direction of each path starting point in each plastic area according to the path ending point and the path starting point; obtaining a plurality of external interference paths of the starting point of each path in each plastic area according to the path direction and the brightness difference parameters;
Obtaining a brightness threshold value of each external interference path according to the external interference path and the brightness difference parameter; reducing the starting point of each path in each plastic region according to the brightness threshold value to obtain each reduced plastic region; performing quality assessment according to the reduced plastic region;
the method for obtaining a plurality of external interference paths of the starting point of each path in each plastic area according to the path direction and the brightness difference parameter comprises the following specific steps:
for any path starting point with the path direction being horizontal to the right, taking the pixel points in the eight neighborhood of the path starting point and horizontal to the right as the first path point, stopping traversing until the latest acquired path point is the path ending point, and marking the image area occupied by all the path points as the external interference path of the path starting point; if the path direction of the path starting point is horizontal left, vertical upward or vertical downward, taking pixel points corresponding to the path directions in eight adjacent domains as first path points, and acquiring subsequent path points according to the path directions until traversing to a path ending point, so as to obtain external interference paths of which all the path directions are horizontal right, horizontal left, vertical upward and vertical downward path starting points;
For any one path starting point with the path direction being the right lower direction, in the eight adjacent areas of the path starting point, the pixel point of the path starting point at the horizontal right is marked as the horizontal right side starting point of the path starting point, the pixel point of the path starting point at the horizontal left is marked as the horizontal left side starting point of the path starting point, the pixel point of the path starting point at the vertical upper side of the path starting point is marked as the vertical upper side starting point of the path starting point, and the pixel point of the path starting point at the vertical lower side of the path starting point is marked as the vertical lower side starting point of the path starting point;
for a horizontal left starting point of the path starting point, taking the horizontal left starting point as a first path point, and taking a pixel point vertically downward in eight adjacent areas of the first path point as a second path point; if the brightness difference parameter of the pixel points in the eight neighborhood of the second path point to the right is more than or equal to the brightness difference parameter of the pixel points in the vertical direction, marking the pixel points in the eight neighborhood of the second path point to the right as a third path point; if the brightness difference parameter of the pixel points in the eight neighborhood of the second path point, which is horizontally right, is smaller than the brightness difference parameter of the pixel points in the vertical direction, the pixel points in the eight neighborhood of the second path point, which are vertically downward, are marked as a third path point; if the brightness difference parameter of the pixel points in the eight neighborhood of the third path point to the right is more than or equal to the brightness difference parameter of the pixel points in the vertical direction, marking the pixel points in the eight neighborhood of the third path point to the right as a fourth path point; if the brightness difference parameter of the pixel points in the eight neighborhood of the third path point, which is horizontally right, is smaller than the brightness difference parameter of the pixel points in the vertical direction, the pixel points in the eight neighborhood of the third path point, which are vertically downward, are marked as fourth path points;
In the subsequent path point acquisition process, if the path point acquired by the latest traversal and the path ending point are in the same horizontal line, the subsequently acquired path point is a horizontal rightward pixel point in the eight neighborhood of the previous path point; if the latest traversing acquired path point and the path ending point are in the same vertical line, the subsequently acquired path point is a vertically downward pixel point in the eight neighborhood of the previous path point; if the latest traversed and acquired path point and the path ending point are still not in the same horizontal line or the same vertical line, continuing to acquire the next path point according to the brightness difference parameter until the latest traversed and acquired path point and the path ending point are in the same horizontal line or the same vertical line, acquiring the subsequent path point, stopping traversing until the latest acquired path point is the path ending point, and recording the image area occupied by all the path points as an interference path of a horizontal left starting point;
for the case that a horizontal right starting point, a vertical upper starting point and a vertical lower starting point of the path starting point are respectively used as a first path point, acquiring a next path point according to the brightness difference parameters of two pixel points in the eight adjacent areas corresponding to the path direction and whether the brightness difference parameters are the path starting points or not until interference paths respectively corresponding to the first starting points under the three conditions are finally obtained; respectively marking an interference path of a horizontal left starting point, an interference path of a horizontal right starting point, an interference path of a vertical upper starting point and an interference path of a vertical lower starting point as an external interference path of which the path direction is a path starting point in the lower right direction, and acquiring a plurality of external interference paths of which each path direction is a path starting point in the lower right direction;
For the path starting points of which the path directions are the upper right direction, the upper left direction or the lower left direction, four interference paths of each path starting point are obtained, and the four interference paths of each path starting point are used as external interference paths of each path starting point;
acquiring a plurality of external interference paths of each path starting point in each plastic region;
the method for obtaining the brightness threshold of each external interference path according to the external interference path and the brightness difference parameter comprises the following specific steps:
for any external disturbance path at the start point of any path in any plastic region, in the formula,a bright-dark threshold value representing an external interference path; />Representing preset super parameters; />Representing the number of pixels contained within the plastic region;indicating the%>A lighting difference parameter of each pixel point; />Representing the maximum value of the brightness difference parameters of all pixel points in the plastic area; />Representing the minimum value of the brightness difference parameters of all pixel points in the plastic area; />Representing preset super parameters; />Representing the number of path points contained in the external interference path; />Representing the average value of the brightness difference parameters of all pixel points in the plastic area;an exponential function that is based on a natural constant;
Acquiring a brightness threshold value of each external interference path;
the method for reducing the starting point of each path in each plastic region according to the brightness threshold value to obtain each reduced plastic region comprises the following specific steps:
for any external interference path of any path starting point in any plastic region, obtaining the relative brightness relationship of the external interference path by using a Retinex algorithm according to the brightness threshold of the external interference path; acquiring the relative brightness relationship of all external interference paths of the path starting points in the plastic area;
the method comprises the following steps of restoring the path starting point according to the relative brightness relation of all external interference paths of the path starting point in the plastic area:
in the method, in the process of the invention,representing the gray value of the path after the initial point is restored; />Gray values representing the path termination points within the plastic region; />The average value of the relative brightness relationship of all external interference paths at the starting point of the path is represented;
reducing all path starting points in the plastic area, and marking the reduced plastic area as a reduced plastic area;
each reduced plastic region is acquired.
2. The method for evaluating extrusion quality of plastic film based on image processing according to claim 1, wherein the obtaining a plurality of local neighborhood differences of each pixel point in each plastic region comprises the following specific steps:
Marking any pixel point in any plastic region on any plastic gray level image as a target pixel point, taking the target pixel point as a center, acquiring a gray level value of each pixel point in eight adjacent regions of the target pixel point, and marking the gray level value as a local neighborhood gray level value of the target pixel point; for any local neighborhood gray value of the target pixel, marking the absolute value of the difference value between the gray value of the target pixel and the local neighborhood gray value as the local neighborhood difference value of the target pixel; obtaining all local neighborhood difference values of target pixel points in a plastic region;
and obtaining a plurality of local neighborhood difference values of each pixel point in each plastic region.
3. The method for evaluating extrusion quality of plastic film based on image processing according to claim 1, wherein the obtaining a plurality of local difference probabilities of each pixel point in each plastic region according to the local neighborhood difference value comprises the following specific steps:
for any one plastic region, acquiring the occurrence frequency of each local neighborhood difference value of each pixel point in the plastic region from all local neighborhood difference values of all pixel points in the plastic region, and marking the occurrence frequency as the local difference probability of each pixel point in the plastic region;
And acquiring a plurality of local difference probabilities of each pixel point in each plastic region.
4. The method for evaluating the extrusion quality of the plastic film based on the image processing according to claim 1, wherein the step of obtaining the brightness difference parameter of each pixel point in each plastic area according to the local difference probability comprises the following specific steps:
any pixel point in any one plastic area is marked as a first target pixel point, in the formula,a luminance difference parameter representing a first target pixel point; />Representing the number of pixel points contained in the plastic region to which the first target pixel point belongs; />A gray value representing a first target pixel point; />Representing the +.>Gray values of the individual pixels; />Representing preset super parameters; />A mean value representing all local difference probabilities of the first target pixel point;
and acquiring the brightness difference parameters of each pixel point in each plastic region.
5. The method for evaluating the extrusion quality of the plastic film based on the image processing according to claim 1, wherein the method for obtaining the path end point and the plurality of path start points of each plastic area according to the brightness parameter comprises the following specific steps:
For any one plastic region, marking the pixel point with the largest brightness difference parameter in the plastic region as a path termination point of the plastic region; marking each pixel point except the path ending point in the plastic area as a path starting point of the plastic area;
and acquiring a path ending point and a plurality of path starting points of each plastic area.
6. The method for evaluating extrusion quality of plastic film based on image processing according to claim 1, wherein the obtaining the path direction of each path starting point in each plastic region according to the path ending point and the path starting point comprises the following specific steps:
a path ending point for any one of the plastic regions and any one of the path starting points;
if the path ending point is positioned in the upper right area of the path starting point, the path direction of the path starting point is the upper right direction;
if the path ending point is positioned in the right lower area of the path starting point, the path direction of the path starting point is the right lower direction;
if the path ending point is positioned in the upper left area of the path starting point, the path direction of the path starting point is the upper left direction;
if the path ending point is positioned in the left lower area of the path starting point, the path direction of the path starting point is the left lower direction;
If the path ending point is positioned in the horizontal right direction of the path starting point, the path direction of the path starting point is horizontal right;
if the path ending point is positioned in the horizontal left direction of the path starting point, the path direction of the path starting point is horizontal left;
if the path ending point is positioned in the vertical upward direction of the path starting point, the path direction of the path starting point is vertical upward;
if the path ending point is positioned in the vertically downward direction of the path starting point, the path direction of the path starting point is vertically downward;
and acquiring the path direction of the starting point of each path in each plastic area.
7. The image processing-based plastic film extrusion quality evaluation method according to claim 1, wherein the quality evaluation based on the reduced plastic region comprises the following specific steps:
for any plastic gray level image, the accumulated sum of gray level values of all pixel points in each reduced plastic area is recorded as the initial uniformity degree of each reduced plastic area, the initial uniformity degree of all the reduced plastic areas is subjected to linear normalization, and the normalized initial uniformity degree is recorded as the uniformity degree of each reduced plastic area;
marking a preset uniformity degree threshold as T2, and marking the plastic gray level image as a plastic gray level image with unqualified quality if the uniformity degree of any one of the reduced plastic areas is larger than T2; if any reduced plastic area does not exist, the uniformity degree of the reduced plastic area is larger than T2, and no treatment is carried out on the plastic gray image;
And obtaining all plastic gray-scale images with unqualified quality.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311204130.2A CN116958134B (en) | 2023-09-19 | 2023-09-19 | Plastic film extrusion quality evaluation method based on image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311204130.2A CN116958134B (en) | 2023-09-19 | 2023-09-19 | Plastic film extrusion quality evaluation method based on image processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116958134A CN116958134A (en) | 2023-10-27 |
| CN116958134B true CN116958134B (en) | 2023-12-19 |
Family
ID=88456850
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311204130.2A Active CN116958134B (en) | 2023-09-19 | 2023-09-19 | Plastic film extrusion quality evaluation method based on image processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116958134B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107622502A (en) * | 2017-07-28 | 2018-01-23 | 南京航空航天大学 | The path extraction of robot vision leading system and recognition methods under the conditions of complex illumination |
| CN113962993A (en) * | 2021-12-21 | 2022-01-21 | 武汉霖杉工贸有限公司 | Paper cup raw material quality detection method based on computer vision |
| WO2022021912A1 (en) * | 2020-07-31 | 2022-02-03 | 南方科技大学 | Low power consumption stereo matching system and method for obtaining depth information |
| CN114862849A (en) * | 2022-07-06 | 2022-08-05 | 山东智领新材料有限公司 | Aluminum alloy plate film coating effect evaluation method based on image processing |
| CN116309579A (en) * | 2023-05-19 | 2023-06-23 | 惠州市宝惠电子科技有限公司 | Transformer welding seam quality detection method using image processing |
| CN116416246A (en) * | 2023-06-08 | 2023-07-11 | 临沂中科芯华新材料科技有限公司 | Machine vision-based fully-degradable plastic product film coating effect evaluation method |
-
2023
- 2023-09-19 CN CN202311204130.2A patent/CN116958134B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107622502A (en) * | 2017-07-28 | 2018-01-23 | 南京航空航天大学 | The path extraction of robot vision leading system and recognition methods under the conditions of complex illumination |
| WO2022021912A1 (en) * | 2020-07-31 | 2022-02-03 | 南方科技大学 | Low power consumption stereo matching system and method for obtaining depth information |
| CN113962993A (en) * | 2021-12-21 | 2022-01-21 | 武汉霖杉工贸有限公司 | Paper cup raw material quality detection method based on computer vision |
| CN114862849A (en) * | 2022-07-06 | 2022-08-05 | 山东智领新材料有限公司 | Aluminum alloy plate film coating effect evaluation method based on image processing |
| CN116309579A (en) * | 2023-05-19 | 2023-06-23 | 惠州市宝惠电子科技有限公司 | Transformer welding seam quality detection method using image processing |
| CN116416246A (en) * | 2023-06-08 | 2023-07-11 | 临沂中科芯华新材料科技有限公司 | Machine vision-based fully-degradable plastic product film coating effect evaluation method |
Non-Patent Citations (1)
| Title |
|---|
| 一种改进的全局Retinex图像增强算法及其仿真研究;乔小燕;姬光荣;陈雾;;系统仿真学报(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116958134A (en) | 2023-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108876743B (en) | Image rapid defogging method, system, terminal and storage medium | |
| CN116740070B (en) | Plastic pipeline appearance defect detection method based on machine vision | |
| CN110070531B (en) | Model training method for detecting fundus picture, and fundus picture detection method and device | |
| CN116310360A (en) | Reactor surface defect detection method | |
| CN116229335B (en) | Livestock and poultry farm environment recognition method based on image data | |
| CN109740721B (en) | Wheat ear counting method and device | |
| CN114926436A (en) | Defect detection method for periodic pattern fabric | |
| CN115511907B (en) | Scratch detection method for LED screen | |
| CN117422712B (en) | Plastic master batch visual detection method and system based on image filtering processing | |
| CN116883408B (en) | Integrating instrument shell defect detection method based on artificial intelligence | |
| CN111814570A (en) | Face recognition method, system and storage medium based on dynamic threshold | |
| CN117058147A (en) | Environment-friendly plastic product defect detection method based on computer vision | |
| CN115937186A (en) | Textile defect identification method and system | |
| CN116228804A (en) | Mineral resource identification method based on image segmentation | |
| CN116843581B (en) | Image enhancement method, system, device and storage medium for multi-scene graph | |
| CN117173180B (en) | Intelligent steel pipe defect detection method based on image enhancement | |
| CN117218029B (en) | Night dim light image intelligent processing method based on neural network | |
| CN116958134B (en) | Plastic film extrusion quality evaluation method based on image processing | |
| CN113989127A (en) | Image contrast adjusting method, system, equipment and computer storage medium | |
| CN115423723B (en) | Self-adaptive railway wagon anomaly detection training set image enhancement method | |
| CN117541582B (en) | IGBT insulation quality detection method for high-frequency converter | |
| CN117495829B (en) | Intelligent watch hardware quality detection method | |
| CN116703930B (en) | Automobile rearview mirror mold forming detection method | |
| CN117314899B (en) | Carbon fiber plate quality detection method based on image characteristics | |
| CN116740056B (en) | Defect detection method for coating layer of whole-core high-pattern conveyer belt |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |