CN114972337B - Anomaly identification method based on waterproof cloth Hough space data processing - Google Patents
Anomaly identification method based on waterproof cloth Hough space data processing Download PDFInfo
- Publication number
- CN114972337B CN114972337B CN202210881129.2A CN202210881129A CN114972337B CN 114972337 B CN114972337 B CN 114972337B CN 202210881129 A CN202210881129 A CN 202210881129A CN 114972337 B CN114972337 B CN 114972337B
- Authority
- CN
- China
- Prior art keywords
- straight lines
- obtaining
- difference
- suspected
- degree
- 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
- 239000004744 fabric Substances 0.000 title claims abstract description 33
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000012545 processing Methods 0.000 title claims abstract description 18
- 230000007547 defect Effects 0.000 claims abstract description 118
- 230000001186 cumulative effect Effects 0.000 claims description 22
- 238000011156 evaluation Methods 0.000 claims description 16
- 230000009466 transformation Effects 0.000 claims description 13
- 230000005856 abnormality Effects 0.000 claims description 4
- 230000007797 corrosion Effects 0.000 claims description 3
- 238000005260 corrosion Methods 0.000 claims description 3
- 230000002950 deficient Effects 0.000 claims description 2
- 239000006185 dispersion Substances 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 23
- 238000007405 data analysis Methods 0.000 abstract description 2
- 230000002159 abnormal effect Effects 0.000 abstract 9
- 238000004458 analytical method Methods 0.000 description 6
- 230000009191 jumping Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000005530 etching Methods 0.000 description 1
- 238000007429 general method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Images
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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/48—Extraction of image or video features by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
- G06T2207/20061—Hough transform
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to an abnormity identification method based on waterproof cloth Hough space data processing. Processing data acquired by electronic equipment through probabilistic Hough transform to obtain Hough space data, marking straight lines corresponding to highlight points in all spaces, dividing waterproof cloth with all marked straight lines into a plurality of sub-blocks, and acquiring suspected abnormal straight lines in each sub-block; positioning the suspected abnormal straight line on the waterproof cloth, and constructing a suspected abnormal area by taking the suspected defect straight line as a center; and acquiring the average gray value and the arrangement entropy of the suspected abnormal area, acquiring the abnormal degree of the suspected abnormal area according to the average gray value and the arrangement entropy, and when the abnormal degree is greater than a preset threshold value, determining the suspected abnormal area as the abnormal area. The detection efficiency is guaranteed, the detection accuracy is improved, and the reliability of data analysis is improved.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an abnormity identification method based on waterproof cloth Hough space data processing.
Background
Tarpaulins are important materials, and if the quality of the tarpaulins is maintained at a certain level during the production process, the tarpaulins with unqualified quality may have serious influence in practical application.
In the process of producing the waterproof cloth, the quality of the waterproof cloth is generally required to be preliminarily detected; the general method for detecting the waterproof cloth is manual detection or machine detection, the efficiency of the manual detection is low, and the error of the detection result is large; detecting all straight lines on the surface of the waterproof cloth by using a machine, wherein the detection is usually carried out according to Hough transform or accumulated probability Hough transform, and then carrying out detection analysis on each straight line; however, the calculation amount of hough transform analysis is large, and accumulated probability hough transform often causes incomplete detection straight lines due to inappropriate threshold setting, so that the accuracy of defect detection is reduced.
Disclosure of Invention
In order to solve the technical problem, the invention aims to provide an anomaly identification method based on tarpaulin hough space data processing, which comprises the following steps:
acquiring a multiframe initial image of a surface area of the waterproof cloth; performing cumulative probability Hough transform on any initial image based on an optimal threshold value to obtain all straight lines in the initial image, and marking all the straight lines in the initial image;
dividing the initial image marked with all the straight lines into a plurality of sub-blocks, and acquiring suspected defect straight lines in each sub-block; positioning the suspected defect straight line into the initial image, and constructing a suspected defect area by taking the suspected defect straight line as a center;
acquiring an average gray value and an arrangement entropy of the suspected defect area, acquiring a defect degree of the suspected defect area according to the average gray value and the arrangement entropy, and when the defect degree is greater than a preset threshold value, determining that the suspected defect area is a defect area;
the method for obtaining the optimal threshold value comprises the following steps:
performing standard Hough transform on the initial image to obtain a plurality of standard straight lines;
sequentially carrying out cumulative probability Hough transformation on pixel points in the initial image, and obtaining a plurality of straight lines in the initial image based on a self-set threshold; obtaining the difference degree according to the difference between all the straight lines and all the standard straight lines;
obtaining evaluation indexes according to the difference degree and the self-set threshold values, wherein each self-set threshold value corresponds to one of the self-set threshold values
And the self-set threshold corresponding to the minimum evaluation index is the optimal threshold.
Preferably, the step of obtaining a straight line of suspected defects in each of the sub-blocks includes:
calculating the average length of all straight lines in each sub-block, and obtaining the spread degree in the sub-block according to the average length;
and when the spread is larger than a preset threshold value, calculating the difference between the length of each straight line in the sub-block and the average length, and obtaining the straight line suspected of being defective according to the difference.
Preferably, the step of obtaining the dispersion degree in the sub-blocks according to the average length includes:
and obtaining the difference between the length of each straight line in any subblock and the average length, and obtaining the spread degree of the subblock according to the difference.
Preferably, the step of constructing a suspected defect area with the suspected defect straight line as a center includes:
and corroding and expanding a certain number of pixel points towards the periphery by taking the suspected defect straight line as a center, wherein a corroded and expanded area is the suspected defect area.
Preferably, the step of obtaining the defect degree of the suspected defect area according to the average gray value and the arrangement entropy includes:
selecting a normal area with any size as a target area, calculating the average gray value and the arrangement entropy corresponding to the target area, and obtaining the defect degree according to the average gray value and the arrangement entropy respectively corresponding to the suspected defect area and the target area as follows:
wherein,
indicating the defect degree corresponding to the suspected defect area;
representing the average gray value corresponding to the suspected defect area;
representing the average gray value corresponding to the target area;
representing the arrangement entropy corresponding to the suspected defect area;
indicating the corresponding permutation entropy of the target area.
Preferably, the step of obtaining the difference degree according to the difference between all the straight lines and all the standard straight lines comprises:
respectively acquiring the number and the sum of the lengths of all the straight lines and the number and the sum of the lengths of all the standard straight lines;
and obtaining the difference degree according to the difference between the number of all the straight lines and the number of all the standard straight lines, and the difference between the sum of the lengths of all the straight lines and the sum of the lengths of all the standard straight lines.
Preferably, the step of obtaining an evaluation index according to the difference degree and the self-set threshold includes:
and obtaining all different straight lines based on each self-set threshold, obtaining corresponding difference degrees according to all different straight lines, and carrying out weighted summation on the difference degrees and the self-set thresholds to obtain the evaluation index.
The invention has the following beneficial effects: in the embodiment of the invention, the quantity of all straight lines in the initial image on the surface of the waterproof cloth is detected through cumulative probability Hough transformation, obtaining the best threshold value of detection through the linear information detected by the accumulated probability Hough transform of a plurality of different self-set threshold values, preliminarily judging the suspected defect straight line in the waterproof cloth based on the gray value and the length information of all the detected straight lines, constructing suspected defect areas according to the suspected defect straight lines, obtaining the defect degree of each suspected defect area by combining the arrangement entropy and the average gray value, thereby determining whether each suspected defect straight line is a determined defect straight line, obtaining complete straight line detection according to the Hough transform of the accumulated probability of partial pixel points, the detection efficiency is guaranteed, meanwhile, the detection accuracy is improved, the subsequent analysis is carried out by combining the gray value and the permutation entropy, and the reliability of data analysis is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flowchart of a method for detecting defects of a tarpaulin based on hough transform according to an embodiment of the present invention.
Detailed Description
To further illustrate the technical means and effects of the present invention adopted to achieve the predetermined invention purpose, the following detailed description, the structure, the features and the effects of the abnormality identification method based on the waterproof cloth hough space data processing according to the present invention are provided with the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
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 method is suitable for a scene of detecting straight lines of defects on the surface of the waterproof cloth, and aims to solve the problems that the existing cumulative probability Hough transform threshold is not well set and the relative defect detection efficiency is low.
The anomaly identification method based on waterproof cloth Hough space data processing provided by the invention is specifically described below by combining the attached drawings
The specific scheme of (1).
Referring to fig. 1, a flowchart of a method for detecting defects of a tarpaulin based on hough transform according to an embodiment of the present invention is shown, where the method includes the following steps:
s100, acquiring a multi-frame initial image of the surface area of the waterproof cloth; and performing cumulative probability Hough transformation on any initial image based on the optimal threshold to obtain all straight lines in the initial image, and marking all the straight lines in the initial image.
In the transportation process after the production of the waterproof cloth is finished, whether the defects of wire jumping, wire breakage and the like exist on the surface of the waterproof cloth needs to be detected, in order to improve the analysis and detection efficiency, the industrial camera is arranged to collect images right above a conveyor belt for conveying the waterproof cloth, and the shooting frequency of the industrial camera is adaptively set according to the conveying speed and the length of the conveyor belt, so that the condition that the waterproof cloth is missed in shooting is avoided. Therefore, multiple frames of surface images of the waterproof cloth are obtained, and in order to reduce the influence of image noise, the shot surface images of each frame are subjected to denoising treatment; and (4) performing graying after denoising the multi-frame surface images respectively to obtain corresponding multi-frame initial images.
Furthermore, as the warp and weft of the waterproof cloth are uniformly distributed, and the interference of factors such as pattern texture and the like is basically avoided except the warp and weft, all straight lines in the initial image of the surface of the waterproof cloth can be detected by Hough transform, and the straight lines are used for analyzing whether the defects such as wire jumping, wire breakage and the like exist in the initial image of the surface of the waterproof cloth.
The traditional Hough transform algorithm projects all pixel points in an image to be analyzed into Hough space, and all straight lines in the image are obtained by combining a set threshold value, the speed of detecting the straight lines by the traditional Hough transform algorithm is very low, in order to accelerate the speed of identifying the straight lines, the embodiment of the invention adopts the improved algorithm of the Hough transform algorithm to accumulate probability Hough transform, and the accumulated probability Hough transform is to select partial points to complete the detection of the straight lines, so the setting of the threshold value plays a crucial role in the integrity of the detected straight lines, and when a multi-frame initial image of a waterproof cloth is processed, the optimal threshold value is required to be screened.
Carrying out standard Hough transform on the initial image to obtain a plurality of standard straight lines; sequentially carrying out Hough transform on the accumulated probability of pixel points in the initial image, and obtaining a plurality of straight lines in the initial image based on a self-set threshold; obtaining the difference degree according to the difference between all the straight lines and all the standard straight lines; and obtaining evaluation indexes according to the difference and the self-set threshold values, wherein each self-set threshold value corresponds to one evaluation index, and the corresponding self-set threshold value is the optimal threshold value when the evaluation index is minimum.
Specifically, the method for obtaining the optimal threshold in the embodiment of the present invention is as follows:
taking the first frame initial image corresponding to the tarpaulin as an example, performing standard hough transform on the initial image to obtain a plurality of standard straight lines, and analyzing by taking the plurality of standard straight lines detected by the standard hough transform as a reference.
In order to improve the detection efficiency, in the embodiment of the invention, one pixel point is randomly selected as a target point in the initial image, the target point is subjected to cumulative probability Hough transformation, and by analogy, different target points are continuously selected from the initial image to be cumulatively added
And probability Hough transformation, wherein when the accumulated time value in the Hough space exceeds a self-set threshold value, a plurality of target points corresponding to the accumulated point are a straight line in the initial image, and the straight line is marked.
Further, removing the determined straight lines from the initial image; and then, sequentially carrying out cumulative probability Hough transformation on all the remaining target points, and similarly, screening and marking the corresponding target points when the cumulative number exceeds a set threshold value. And by analogy, obtaining a plurality of straight lines in the initial image based on the cumulative probability Hough transform of the self-set threshold. Since the standard hough transform detects straight lines in the same direction, there may be a straight line formed by two or more line segments, and since the cumulative probability hough transform is a sampling detection, it detects the presence of a short line segment, and merges the line segments in the same direction detected by the cumulative probability hough transform into a straight line.
It should be noted that, in the embodiment of the present invention, the value range of the self-set threshold is set as
Wherein
is the threshold of the standard hough transform. Preferably, in the embodiment of the present invention, the threshold of the standard hough transform is set to be
I.e. the value range of the self-set threshold is
(ii) a Obtaining straight lines with different numbers and lengths according to Hough transform of cumulative probability through self-set thresholds with different sizes; in other embodiments, the value range of the self-setting threshold value can be set according to the actual situation.
Considering the characteristics of longitude and latitude lines on the surface of the waterproof cloth, the direction of the accumulated probability Hough transform detection is specified in the embodiment of the invention
And
，
in two intervals, only straight lines in the two intervals are considered when probability hough transformation is carried out.
Further, respectively acquiring the number and the sum of the lengths of all the straight lines and the number and the sum of the lengths of the standard straight lines; and obtaining the difference degree according to the difference between the number of all straight lines and the number of all standard straight lines and the difference between the sum of the lengths of all straight lines and the sum of the lengths of all standard straight lines.
Specifically, the sum of the number of all straight lines and the length of the straight lines obtained by the hough transform of the cumulative probability under an arbitrary self-set threshold is obtained, and then the sum of the number of all standard straight lines and the length of all standard straight lines obtained by the hough transform of the standard probability is obtained, so that the difference between the detected straight lines obtained by the hough transform of the cumulative probability and the standard hough transform is:
wherein,
representing the degree of difference;
representing the number of all straight lines obtained by accumulative probability Hough transformation;
representing the number of all standard straight lines obtained by standard Hough transform;
representing the sum of the lengths of all straight lines obtained by accumulative probability Hough transformation;
the standard hough transform is expressed to obtain the sum of the lengths of all standard straight lines.
Further, due to the continuous change of the self-set threshold, part of oblique straight lines may be detected through the cumulative probability hough transform, so that the number of the detected straight lines is increased, different straight lines are obtained based on each self-set threshold, corresponding difference degrees are obtained according to the different straight lines, and the difference degrees and the self-set threshold are subjected to weighted summation to obtain an evaluation index:
wherein,
to representEvaluating the index;
representing a current self-set threshold size;
representing the degree of difference;
representing a weight of a self-set threshold;
and a weight value representing the degree of difference.
Preferably, the device is arranged in the embodiment of the invention
。
By analogy, a plurality of evaluation indexes are obtained according to the difference degree of straight lines obtained by Hough transform of different self-set thresholds and the cumulative probabilities of the different self-set thresholds, when the evaluation index is minimum, the self-set threshold is smaller and the difference degree is smaller, the efficiency of the corresponding algorithm is better, and therefore the self-set threshold corresponding to the minimum evaluation index is used as the optimal threshold, and the cumulative probability Hough transform is carried out on the initial image of the waterproof cloth by adopting the optimal threshold. When the line detection analysis is carried out on the multi-frame initial images of the same piece of waterproof cloth, the line in each frame of initial image can be directly analyzed and detected by adopting the cumulative probability Hough transform of the optimal threshold value.
Step S200, dividing the initial image marked with the straight line into a plurality of sub blocks, and acquiring a suspected defect straight line in each sub block; and positioning the suspected defect straight line into the initial image, and constructing a suspected defect area by taking the suspected defect straight line as the center.
In step S100, all the straight lines in the initial image are obtained and marked, and in order to ensure that only one warp or weft is detected when a defect exists, in the embodiment of the present invention, the marked initial image is divided into a plurality of sub-blocks, and each sub-block is analyzed.
Specifically, the average length of all straight lines in any subblock is obtained, the difference between the length of each straight line in the subblock and the average length is obtained, and the parameter of the subblock is obtained according to the difference; namely, the spread of the straight line in each sub-block is as follows:
wherein,
is shown as
The corresponding spread of each sub-block;
representing the number of all lines in the sub-block;
indicates the first in the sub-block
The length of the bar line;
is shown as
Average length of all lines in a sub-block.
And acquiring the corresponding spread degree of each sub-block in the initial image, and when the spread degree is greater than a preset threshold, indicating that the length difference of straight lines in the sub-blocks is large, and the conditions of wire jumping, wire breaking and the like may exist.
Preferably, in the embodiment of the present invention, the preset threshold is set as
I.e. when the subblock is presentWhen the corresponding spread is larger than 10, it indicates that the subblock region may have the defects of wire jumping, wire breaking and the like.
Further, the difference between the length of each straight line in the subblock region and the average length is obtained as follows:
wherein,
representing the difference between the length of any one straight line and the average length;
represents the length of any straight line;
is shown as
Average length of all lines in a sub-block.
When the difference is larger than the preset threshold value, the difference between the length of the straight line and the average length is larger, and the straight line is divided into a plurality of straight lines
The bar lines are marked as suspected defect lines.
Preferably, in the embodiment of the present invention, the preset threshold is set as
I.e. when the difference between the length of any one straight line and the average length is greater than 5, marking the straight line as a suspected-defect straight line.
And projecting all the obtained suspected defect straight lines in the sub-blocks to an original initial image, and corroding and expanding a certain number of pixel points respectively from top to bottom and from left to right by taking the suspected defect straight lines as the center, wherein a corrosion expansion area is a suspected defect area. Namely, the upper, lower, left and right etching expansion is performed with each suspected defect straight line as the center position, so that one area is obtained and is marked as a suspected defect area.
Preferably, in the embodiment of the present invention, the suspected defect straight line is used as a center, and the pixels are respectively expanded by 20 pixels up, down, left, and right to form the suspected defect area.
And by analogy, the suspected defect area corresponding to each suspected defect straight line in each sub-block is obtained, so that a plurality of suspected defect areas are obtained.
Step S300, obtaining an average gray value and an arrangement entropy of the suspected defect area, obtaining a defect degree of the suspected defect area according to the average gray value and the arrangement entropy, and when the defect degree is larger than a preset threshold value, determining that the suspected defect area is a defect area.
Specifically, a plurality of suspected defect areas are obtained in step S200, and each suspected defect area is further analyzed. Selecting a normal area with any size as a target area; preferably, in the embodiment of the present invention, the target area is set as
The area of (a).
Further, the permutation entropy of each suspected defect area and the permutation entropy of the target area are calculated, the permutation entropy is used for measuring the complexity of time series data, whether the permutation rule of each suspected defect area is different from that of the target area or not is determined according to the permutation entropy, and the embedding dimension of the selected parameter is 3 and the time delay is 1 when the permutation entropy is calculated in the embodiment of the invention.
Thus, the defect degree corresponding to each suspected defect area is calculated as:
wherein,
indicating the defect degree corresponding to the suspected defect area;
representing the average gray value corresponding to the suspected defect area;
representing the average gray value corresponding to the target area;
representing the arrangement entropy corresponding to the suspected defect area;
indicating the corresponding permutation entropy of the target area.
Normalizing the acquired defect degree corresponding to each suspected defect area; when the average gray value difference and the arrangement entropy difference between the suspected defect area and the target area are large, namely the defect degree of the suspected defect area is larger than a preset threshold value, the suspected defect area is indicated to have the defect conditions of wire jumping, wire breaking and the like, the suspected defect area is the defect area, and the suspected defect straight line in the defect area is the determined defect straight line.
Preferably, in the embodiment of the present invention, the preset threshold is an empirical value
That is, when the value of the defect degree is greater than 0.05, the suspected defect area is a defect area, and the suspected defect straight line is a defect on the waterproof cloth surface having a defect
A straight line.
In summary, in the embodiment of the present invention, cumulative probability hough transformation is sequentially performed on pixel points in an initial image of the tarpaulin, and the cumulative probability hough transformation based on an optimal threshold is performed to obtain a plurality of straight lines, where the optimal threshold of the probability hough transformation is obtained by detecting a difference between a plurality of straight lines corresponding to different self-set thresholds each time and a standard straight line; marking a plurality of straight lines obtained based on an optimal threshold value in an initial image, dividing the initial image into a plurality of sub-blocks for analysis, obtaining suspected defect straight lines in each sub-block according to length information corresponding to each straight line in each sub-block, further performing corrosion expansion according to each suspected defect straight line to obtain a plurality of suspected defect areas, obtaining arrangement entropy and average gray value in each suspected defect area, determining whether each suspected defect area is a defect area according to the difference between the arrangement entropy and the average gray value of the suspected defect area and determining whether each suspected defect straight line is a defect straight line, thereby finding out the position of the surface of the waterproof cloth with defects, reducing the calculation amount of detection processing while ensuring the detection accuracy, and greatly improving the efficiency.
It should be noted that: the sequence of the above embodiments of the present invention is only for description, and does not represent the advantages or disadvantages of the embodiments. And specific embodiments thereof have been described above. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (6)
1. An abnormity identification method based on waterproof cloth Hough space data processing is characterized by comprising the following steps:
acquiring a multiframe initial image of the surface area of the waterproof cloth; performing cumulative probability Hough transform on any initial image based on an optimal threshold value to obtain all straight lines in the initial image, and marking all the straight lines in the initial image;
dividing the initial image marked with all the straight lines into a plurality of sub-blocks, and acquiring suspected defect straight lines in each sub-block; positioning the suspected defect straight line into the initial image, corroding and expanding a certain number of pixel points to the periphery by taking the suspected defect straight line as a center, and marking a corrosion expansion area as a suspected defect area;
acquiring an average gray value and an arrangement entropy of the suspected defect area, acquiring a defect degree of the suspected defect area according to the average gray value and the arrangement entropy, and when the defect degree is greater than a preset threshold value, determining that the suspected defect area is a defect area;
the method for obtaining the optimal threshold value comprises the following steps:
carrying out standard Hough transform on the initial image to obtain a plurality of standard straight lines;
sequentially carrying out cumulative probability Hough transformation on pixel points in the initial image, and obtaining a plurality of straight lines in the initial image based on a self-set threshold; obtaining the difference degree according to the difference between all the straight lines and all the standard straight lines;
and obtaining evaluation indexes according to the difference and the self-set threshold values, wherein each self-set threshold value corresponds to one evaluation index, and the self-set threshold value corresponding to the minimum evaluation index is the optimal threshold value.
2. The anomaly identification method based on the tarpaulin hough space data processing as claimed in claim 1, wherein the step of obtaining the straight line of suspected defect in each sub-block comprises:
calculating the average length of all straight lines in each sub-block, and obtaining the spread degree in the sub-block according to the average length;
and when the spread is larger than a preset threshold value, calculating the difference between the length of each straight line in the sub-block and the average length, and obtaining the straight line suspected of being defective according to the difference.
3. The method for identifying the abnormality based on the tarpaulin hough space data processing according to claim 2, wherein the step of obtaining the dispersion degree in the sub-blocks according to the average length comprises:
and obtaining the difference between the length of each straight line in any subblock and the average length, and obtaining the spread degree of the subblock according to the difference.
4. The method for identifying the abnormality based on the tarpaulin hough space data processing according to claim 1, wherein the step of obtaining the defect degree of the suspected defect area according to the average gray value and the arrangement entropy comprises:
selecting a normal area with any size as a target area, calculating the average gray value and the arrangement entropy corresponding to the target area, and obtaining the defect degree according to the average gray value and the arrangement entropy respectively corresponding to the suspected defect area and the target area as follows:
wherein,
indicating the defect degree corresponding to the suspected defect area;
representing the average gray value corresponding to the suspected defect area;
representing the average gray value corresponding to the target area;
representing the arrangement entropy corresponding to the suspected defect area;
indicating the corresponding permutation entropy of the target area.
5. The method for identifying the abnormality based on the tarpaulin hough space data processing according to claim 1, wherein the step of obtaining the degree of difference according to the difference between all the straight lines and all the standard straight lines comprises:
respectively acquiring the number and the sum of the straight lines and the number and the sum of the straight line lengths of all the standard straight lines;
and obtaining the difference degree according to the difference between the number of all the straight lines and the number of all the standard straight lines, and the difference between the sum of the lengths of all the straight lines and the sum of the lengths of all the standard straight lines.
6. The tarpaulin hough space data processing-based anomaly identification method according to claim 1, wherein the step of obtaining an evaluation index according to the difference degree and the self-set threshold comprises the following steps:
and obtaining all different straight lines based on each self-set threshold, obtaining corresponding difference degrees according to all different straight lines, and performing weighted summation on the difference degrees and the self-set thresholds to obtain the evaluation index.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210881129.2A CN114972337B (en) | 2022-07-26 | 2022-07-26 | Anomaly identification method based on waterproof cloth Hough space data processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210881129.2A CN114972337B (en) | 2022-07-26 | 2022-07-26 | Anomaly identification method based on waterproof cloth Hough space data processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114972337A CN114972337A (en) | 2022-08-30 |
| CN114972337B true CN114972337B (en) | 2022-09-30 |
Family
ID=82968573
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210881129.2A Active CN114972337B (en) | 2022-07-26 | 2022-07-26 | Anomaly identification method based on waterproof cloth Hough space data processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114972337B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115222736B (en) * | 2022-09-20 | 2022-12-13 | 南通佳晟宏机械配件有限公司 | Steel pipe production quality detection method based on Hough space |
| CN117113107B (en) * | 2023-10-24 | 2023-12-29 | 广东顺德爱顺机电科技有限公司 | Rotor die casting abnormal error data processing method |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102393422A (en) * | 2011-08-22 | 2012-03-28 | 江苏省产品质量监督检验研究院 | Ultrasonic time of flight diffraction (TOFD)-based offline defect judgment method |
| KR101677875B1 (en) * | 2016-04-12 | 2016-11-18 | 숭실대학교산학협력단 | Apparatus and method for line segment detection |
| CN111696348B (en) * | 2020-06-05 | 2022-08-09 | 南京云创大数据科技股份有限公司 | Multifunctional intelligent signal control system and method |
-
2022
- 2022-07-26 CN CN202210881129.2A patent/CN114972337B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN114972337A (en) | 2022-08-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114972337B (en) | Anomaly identification method based on waterproof cloth Hough space data processing | |
| CN115601364B (en) | Golden finger circuit board detection method based on image analysis | |
| CN114937055B (en) | Image self-adaptive segmentation method and system based on artificial intelligence | |
| CN113538433A (en) | Mechanical casting defect detection method and system based on artificial intelligence | |
| CN115020267B (en) | Semiconductor surface defect detection method | |
| CN115829883A (en) | Surface image denoising method for dissimilar metal structural member | |
| CN115345885A (en) | Method for detecting appearance quality of metal fitness equipment | |
| US20060238755A1 (en) | Method for analyzing defect data and inspection apparatus and review system | |
| CN117593294B (en) | Centrifugal fan blade crack visual detection method based on image feature analysis | |
| CN117173187B (en) | Intelligent valve fault detection system | |
| US7352901B2 (en) | Contour inspection method and apparatus | |
| CN116645364B (en) | Alloy steel casting air hole defect detection method based on image data | |
| CN115311267A (en) | Method for detecting abnormity of check fabric | |
| CN117274291B (en) | Method for detecting mold demolding residues based on computer vision | |
| CN115311279A (en) | Machine vision identification method for warp and weft defects of fabric | |
| CN117218124A (en) | Copper-aluminum switching surface processing quality detection method | |
| CN111784645A (en) | Crack detection method for filling pipeline | |
| CN117078679B (en) | Automatic assembly line production detection method for cooling fan based on machine vision | |
| CN117078676B (en) | Breaking hammer shell visual detection method based on image processing | |
| CN117237350A (en) | Real-time detection method for quality of steel castings | |
| CN113516608A (en) | Tire defect detection method and device, and tire detection device | |
| CN116958182A (en) | Quick concrete crack detection method based on image data | |
| JP2004515830A (en) | Quadratic dispersion plot technique for defect detection | |
| CN114693652A (en) | Fabric defect detection method based on Gaussian mixture model | |
| CN118657778B (en) | Method and system for monitoring fault data of automatic production of metal panel |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231225 Address after: Room 3414, Building 33, Zhongnan Century City, Chongchuan District, Nantong City, Jiangsu Province, 226000 Patentee after: Nantong Yanlu Enterprise Management Consulting Co.,Ltd. Address before: 226000 No.7 Jinggong Road, Qidong Economic Development Zone, Nantong City, Jiangsu Province Patentee before: Qidong Gude waterproof fabric Co.,Ltd. |
|
| TR01 | Transfer of patent right |