CN113553926B - Automatic detection method for drawing of steel wire rope core conveyer belt joint based on image matching - Google Patents
Automatic detection method for drawing of steel wire rope core conveyer belt joint based on image matching Download PDFInfo
- Publication number
- CN113553926B CN113553926B CN202110770909.5A CN202110770909A CN113553926B CN 113553926 B CN113553926 B CN 113553926B CN 202110770909 A CN202110770909 A CN 202110770909A CN 113553926 B CN113553926 B CN 113553926B
- Authority
- CN
- China
- Prior art keywords
- wire rope
- steel wire
- joint
- image
- points
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Analysing Materials By The Use Of Radiation (AREA)
- Image Analysis (AREA)
Abstract
An automatic detection method for the drawing of a joint of a steel wire rope core conveyer belt based on image matching belongs to the field of nondestructive detection. The invention aims to accurately and efficiently calculate the joint extraction quantity of the steel wire rope core conveyer belt and realize the judgment of the joint condition. The invention extracts the end points of the steel wire rope in the X-ray image by using the iterative self-adaptive threshold value, and realizes one-to-one correspondence to the end points of the steel wire rope by matching and optimizing based on BRIEF descriptors; and then calculating the single wire rope end pumping quantity, the wire rope upper (lower) end average pumping quantity and the joint overall pumping quantity by a joint pumping measurement method based on a reference point set, detecting and identifying the position and the variation of the abnormal wire rope end, and realizing automatic detection and comprehensive judgment of joint pumping. The algorithm provided by the invention is efficient and high in accuracy, realizes automatic comprehensive detection of the pumping of the joint of the X-ray steel wire rope core conveyer belt, is beneficial to preventing major belt breakage safety accidents, and has a great application value.
Description
Technical Field
The invention relates to an automatic detection method for joint pumping of a steel wire rope core conveyer belt based on an X-ray image, in particular to a method for calculating joint pumping quantity through matching of the X-ray image of the joint of the steel wire rope core conveyer belt, and belongs to the field of nondestructive detection.
Background
The steel wire rope core conveyer belt is a rubber conveyer belt taking a steel wire rope as a framework material, has the advantages of high tensile strength, impact resistance, friction resistance, long service life and the like, and is widely applied to the fields of coal, mines, ports, electric power, chemical industry and the like. In practical application, a complete steel wire rope core conveyer belt is formed by overlapping a plurality of sections of steel wire rope core conveyer belts through a vulcanization process, and the overlapping part is called a joint (called a joint for short) of the steel wire rope core conveyer belt. The joint area is the part with the lowest tensile strength and weakest tensile strength in the whole steel wire rope core conveyer belt, and the joint pumping fault is easy to occur, so that the belt breakage accident is caused. Therefore, it is necessary to detect the state of the joint of the wire rope core conveyor belt accurately in time.
The X-ray detection technology is a mature nondestructive detection technology and has the advantages of intuitiveness, high speed, high detection precision and the like, so that the detection of the pumping fault of the joint of the steel wire rope core conveyer belt in recent years is mainly carried out on an X-ray image based on the joint. However, the existing joint pumping capacity detection method has the problems of low detection efficiency, poor accuracy and the like, and small changes are not easy to judge by means of manual monitoring, so that human resources are wasted. In view of the foregoing, there is a need for an automatic detection method that can efficiently and reliably determine joint twitches.
Disclosure of Invention
The invention provides an automatic detection method for joint drawing of a steel wire rope core conveyer belt based on image matching, and aims to improve the operability and accuracy of joint drawing detection by accurately calculating joint drawing quantity and judging joint conditions by utilizing an X-ray image of the joint of the steel wire rope core conveyer belt based on the image matching method. The method provided by the invention comprises the following steps:
step 1, taking a joint image acquired when a nondestructive testing system of a steel wire rope core conveyer belt based on X-ray is operated for the first time as a reference image, and taking the joint image acquired again as an image to be tested;
step 2, respectively preprocessing a reference image and an image to be detected, carrying out image enhancement through histogram equalization, and then carrying out noise reduction processing by using Gaussian filtering to obtain a preprocessed image I;
step 3, detecting the number M of steel wires in the non-joint area below the reference image and the image to be detected;
step 4, extracting wire rope end points in the reference image and the image to be detected by using a wire rope end point extraction algorithm;
step 5, carrying out feature description through BRIEF description algorithm, matching by using Hamming distance, and then restraining by using distance between matching points and slope of connecting line of the matching points, and removing mismatching points to obtain a plurality of reliable matching point pairs;
step 6, optimizing the image matching result in the step 5 by using a layering method;
step 7, taking the reliable matching point pairs as a reference point set, and calculating the single wire rope end extraction quantity, the average extraction quantity of the wire rope upper end, the average extraction quantity of the wire rope lower end and the joint overall extraction quantity;
and 8, judging the joint whipping, wherein if the joint whipping amount is larger than the average whipping amount of the corresponding steel wire rope end by 5, judging the steel wire rope end as an abnormal steel wire rope end, and if the joint overall whipping amount is larger than 10, judging that the joint has serious joint whipping.
The method for detecting the number M of the steel wire ropes in the lower non-joint area in the reference image and the image to be detected in the step 3 comprises the following steps:
inputting a joint image with resolution of m×n, wherein m represents m pixels in each row, n represents n pixels in each column, taking m×20 region without joint part at the lowest part of the joint image as detection part, dividing into 10The block (2) is divided into binary images by using an OTSU algorithm threshold value, and the binary images are processed in an opposite phase; and then searching points which are not zero along the middle row from left to right, and eliminating points which are less than 3 from the previous point which is not zero, wherein the number of the obtained points is the number M of the steel wire ropes in the non-joint area, and the number M is approximately equal to the number of the upper ends of the steel wire ropes in the joint image.
The step 4 of the invention extracts the end points in the reference image and the image to be detected by using a steel wire rope end point extraction algorithm, and the specific method comprises the following steps:
step a, carrying out convolution operation on the preprocessed image I by using a Scharr operator to obtain R, wherein the numerical value at the upper end of the steel wire rope in the R is a negative value, and the numerical value at the lower end of the steel wire rope is a positive value;
step b, calculating a pixel value mean E (I) and a pixel value variance Var (I) of the image I, setting a step length l=Var (I)/E (I), and setting an initial value of a threshold value T for binarization of the image to be-101;
c, setting the pixel value of the point smaller than T in R as 255 and setting other points as 0 to obtain a binary image of the upper end of the steel wire rope;
step d, carrying out contour detection on the binary images of the upper end of the steel wire rope, calculating a minimum circumcircle for each contour, taking the circle center of each contour as coordinates of the end points of the steel wire rope, calculating the number N of the end points on the steel wire rope, comparing the number N with the number M of the steel wire ropes in the non-joint area obtained in the step 4 of the claim 1, if N is smaller than M, the threshold value T=T+0.51, and if N is larger than M+5, the threshold value T=T-0.51, and then returning to the step d, otherwise, carrying out the next step;
and e, setting a threshold value T= -T, setting a pixel value larger than a point T in R as 255, setting other points as 0, and extracting a lower end point of the steel wire rope.
The BRIEF is an algorithm for rapidly calculating the feature descriptors, the feature points form a binary coded descriptor through the BRIEF description, and the feature points can be matched by utilizing the Hamming distance between the descriptors. Therefore, in the step 5, feature description is carried out through a BRIEF description algorithm, hamming distances are used for matching, the distance between the matching points and the slope of the connecting line of the matching points are used for constraint, and mismatching points are removed, so that a plurality of reliable matching point pairs are obtained;
the step 6 of the invention optimizes the image matching result of the step 5 by using a layering method, and the specific method comprises the following steps:
step 1, layering the end points of the steel wire ropes in a reference image and an image to be detected respectively, and dividing the S-level lap joint into S+1 layers according to a joint lap joint mode, wherein a front S layer is a common layer, S steel wire rope end points at the leftmost end are divided into S layers through distance constraint, then the rest steel wire rope end points are divided into an ith layer according to the distance in sequence, i is less than or equal to S, and the rest steel wire rope end points are put into a garbage layer;
step 2, dividing reliable matching points of end points of the steel wire rope into S+1 layers;
step 3, according to the reliable matching point pair of the steel wire rope end points, the layer is corresponding to the layer, the point A is the steel wire rope end point in the reference image, the point B is the point in the image to be detected which is reliably matched with the point A, and if A is the ith layer of the steel wire rope end point of the reference image and B is the jth layer of the steel wire rope end point of the image to be detected, i and j are two corresponding layers;
and 4, utilizing a transverse position relation to correspond the end points of the steel wire rope between two adjacent reliable matching points of each layer, and completing one-to-one correspondence and matching optimization of the end points of the steel wire rope.
The step 7 of the invention takes the reliable matching point pair as a reference point set to calculate the extraction quantity delta L of the end of the single steel wire rope elongation (p) average extraction amount DeltaL of the upper end head of the steel wire rope all The method comprises the following steps of:
and step 1, when the extraction quantity of the upper end point of the steel wire rope is calculated, reliable matching point pairs of the lower end point of the steel wire rope are used as a reference point set. Respectively calculating the difference value L between the upper end point of the p steel wire rope to be detected in the reference image and the image to be detected and the q datum point reference (p，q)、L detection (p，q)；
Step 2, calculating the difference DeltaL (p, q) =L of the corresponding point pairs of the reference image and the image to be detected reference (p，q)-L detection (p，q)；
Step 3, dnum is the number of reliably matched point pairs, and the pumping quantity of the p-th point is calculated
Step 4, calculating the average extraction quantity of the upper ends of the steel wire rope, wherein the number of the upper ends of the steel wire rope is tipnum
Step 5, calculating the average extraction quantity of the lower end head of the steel wire rope in a similar way;
and 6, calculating an average value of the average extraction quantity of the upper end head of the steel wire rope and the average extraction quantity of the lower end head of the steel wire rope as the integral extraction quantity delta L of the joint.
The invention has the positive effects that: (1) The joint pumping quantity calculating method is efficient and high in accuracy, and can be effectively applied to a nondestructive testing system of the steel wire rope core conveyer belt based on X rays; (2) And detecting and marking the position and the variation of the end head of the abnormal steel wire rope, and realizing automatic detection and comprehensive judgment of the joint drawing. Is beneficial to preventing major belt breakage safety accidents, can effectively improve the production efficiency, and has great application value.
Drawings
Fig. 1 is a flow chart of the overall scheme of the invention.
Fig. 2 is an X-ray image of a steel cord conveyor belt joint.
Fig. 3 is an X-ray image processing result of a steel wire rope core conveyer belt joint of a certain mine, wherein (a), (b), (c), (d), (e), (f), (g) and (h) are respectively a reference image, an image to be detected, a reference image steel wire rope end point extraction diagram, an image to be detected steel wire rope end point extraction diagram, a characteristic point-based matching result diagram, a steel wire rope end point matching result optimization diagram, a steel wire rope end point-to-reference point vertical direction longitudinal coordinate difference diagram, a drawing amount and an abnormal point labeling diagram.
Detailed Description
The following describes specific embodiments of the invention with reference to the drawings.
The whole scheme flow of the invention is shown in figure 1, and the image pretreatment, the extraction of the end point of the steel wire rope, the matching of the end point of the steel wire rope and the drawing calculation of the joint are sequentially carried out on the joint of the steel wire rope core conveyer belt.
Fig. 2 shows an X-ray image of a steel cord conveyor belt joint.
Fig. 3 shows the implementation of the method according to the invention, the main flow comprises:
step 1, a certain coal mine is selected, a certain joint of a 3-level lap joint steel wire rope core conveyer belt is adopted as an experimental object for testing, an image acquired by 20 th month in 2020 is used as a reference image, an image acquired by 25 th month in 2020 is used as an image to be detected, and in order to better verify the effect of the method, the upper and lower steel wire rope ends in the image to be detected are respectively subjected to pumping treatment;
step 2, respectively preprocessing a reference image and an image to be detected, estimating the number of upper ends of the steel wire rope in the image, then, based on an edge detection result of a Scharr operator, obtaining a binary image of the steel wire rope ends by using iterative self-adaptive threshold segmentation, and respectively obtaining steel wire rope end points by detecting and extracting the minimum circumscribed circle center through the outline, as shown in fig. 3 (c) and (d);
step 3, using the result obtained by the steel wire rope end point extraction algorithm to a BRIEF algorithm, and matching by using a Hamming distance to obtain a reliable matching point pair between a reference image and an image to be detected, as shown in fig. 3 (e);
step 4, due to the condition of miss-matching, the matching result is further optimized by using a layering method, so that the one-to-one correspondence of the end points of the steel wire rope is realized, as shown in fig. 3 (f);
step 5, calculating a vertical coordinate difference value from the end point to the reference point, as shown in fig. 3 (g);
and 6, calculating the pumping quantity of the single end, the average pumping quantity of the upper (lower) end of the steel wire rope and the overall pumping quantity of the joint, marking abnormal end points in the drawing, marking abnormal point positions in the reference image, pumping rectangular frames of the end points of the steel wire rope in the image to be detected, and indicating the pumping length by the length in the vertical direction, wherein the drawing length is shown in fig. 3 (h).
The method of the invention is used for calculating the pumping quantity of one steel wire rope core conveyer belt joint of a certain mine by collecting an X-ray joint image through an X-ray nondestructive testing system, and a final testing result is shown in fig. 3 (h), so that the whole steel wire rope core conveyer belt joint has slight joint pumping change. The method automatically detects the end points of the two abnormal steel wire ropes, and the detected pumping quantity is consistent with the result obtained by manual detection through image processing software. The invention is illustrated in terms of validity and reliability of the joint twitch detection.
Tests show that the invention can accurately and automatically detect the pumping state of the joint of the steel wire rope core conveyer belt, and has good use value in a nondestructive testing system of the steel wire rope core conveyer belt based on X rays.
Finally, it should be pointed out that the above embodiments are only intended to illustrate the technical solution of the invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art will understand that modifications may be made to the technical solutions described in the foregoing embodiments, or equivalents may be substituted for some of the technical features thereof, without departing from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (3)
1. The automatic detection method for the drawing of the joint of the steel wire rope core conveyer belt based on image matching is characterized by comprising the following steps:
step 1, taking a joint image acquired when a nondestructive testing system of a steel wire rope core conveyer belt based on X-rays operates for the first time as a reference image, and taking the joint image acquired again as an image to be tested;
step 2, respectively preprocessing a reference image and an image to be detected, carrying out image enhancement through histogram equalization, and then carrying out noise reduction processing by using Gaussian filtering to obtain a preprocessed reference image I and a preprocessed image I1 to be detected;
step 3, detecting the number M of the steel wires in the lower non-joint area in the I, and detecting the number M1 of the steel wires in the lower non-joint area in the I1;
and 4, respectively extracting the wire rope end points in I and I1 according to the same steps by using a wire rope end point extraction algorithm, wherein the specific steps of extracting the wire rope end points in I by using the wire rope end point extraction algorithm are as follows:
step a, carrying out convolution operation on I by using a Scharr operator to obtain R, wherein the numerical value at the upper end of the steel wire rope in the R is a negative value, the numerical value at the lower end of the steel wire rope is a positive value,
step b, calculating the pixel value mean E (I) and the pixel value variance Var (I) of I, setting the step length l=Var (I)/E (I), setting the initial value of the threshold value T used for image binarization to be-10 l,
c, setting the pixel value of the point smaller than T in R as 255 and setting other points as 0 to obtain a binary image of the upper end of the steel wire rope,
step d, carrying out contour detection on the binary images of the upper end heads of the steel wire ropes, calculating the minimum circumcircle for each contour, taking the circle center of each contour as the coordinates of the end heads of the steel wire ropes, calculating the number N of the end heads on the steel wire ropes, comparing the number N with the number M of the steel wire ropes, if N is smaller than M, the threshold value T=T+0.5l, if N is larger than M+5, the threshold value T=T-0.5l, then returning to the step d, otherwise, carrying out the next step,
step e, making a threshold T= -T, setting a pixel value larger than a T point in R as 255, setting other points as 0, and extracting a lower end point of the steel wire rope;
step 5, performing feature description on the extracted end points of the steel wire rope through a BRIEF description algorithm, performing matching by using Hamming distances, and then performing constraint by using the distances between the matching points and the slope of the connecting line of the matching points to remove mismatching points so as to obtain a plurality of reliable matching point pairs;
step 6, optimizing the image matching result in the step 5 by using a layering method;
step 7, taking the reliable matching point pairs as a reference point set, and calculating the single wire rope end extraction quantity, the average extraction quantity of the wire rope upper end, the average extraction quantity of the wire rope lower end and the joint overall extraction quantity;
and 8, judging the joint whipping, wherein if the joint whipping amount is larger than the average whipping amount of the corresponding steel wire rope end by 5, judging the steel wire rope end as an abnormal steel wire rope end, and if the joint overall whipping amount is larger than 10, judging that the joint has serious joint whipping.
2. The automatic detection method for the joint drawing of the steel wire rope core conveyer belt based on image matching as recited in claim 1, wherein the step 3 is used for detecting the number M of the steel wire ropes in the lower non-joint area in the I, and the method is as follows:
inputting a joint image with resolution of m×n, wherein m represents m pixels in each row, n represents n pixels in each column, taking m×20 region without joint part at the lowest part of the joint image as detection part, dividing into 10The block (2) is divided into binary images by using an OTSU algorithm threshold value, and the binary images are processed in an opposite phase; however, the method is thatSearching non-zero points from left to right in the middle of the rear edge, removing points less than 3 from the previous non-zero points, and obtaining the number M of the steel wire ropes with the number of non-joint areas, wherein the number M of the steel wire ropes is approximately equal to the number of the upper ends of the steel wire ropes in the joint image.
3. The automatic detection method for the pumping of the steel wire rope core conveyer belt joint based on the image matching as recited in claim 1, wherein the step 6 uses a layering method to optimize the image matching result of the step 5, and the specific method comprises the following steps:
step 1, layering the end points of the steel wire ropes in the I and the I1 respectively, and dividing the S-level lap joint into S+1 layers according to a joint lap joint mode, wherein the front S layer is a common layer, S steel wire rope end points at the leftmost end are divided into S layers through distance constraint, then the rest steel wire rope end points are divided into the ith layer according to the distance in sequence, I is less than or equal to S, and the rest steel wire rope end points are put into a garbage layer;
step 2, dividing reliable matching points of end points of the steel wire rope into S+1 layers;
step 3, according to the reliable matching point pair of the steel wire rope end points, the layer is corresponding to the layer, the point A is the steel wire rope end point in I, the point B is the point in I1 which is reliably matched with the point A, if the layer A is the ith layer of the steel wire rope end point in I, and the layer B is the jth layer of the steel wire rope end point in I1, I and j are two corresponding layers;
and 4, utilizing a transverse position relation to correspond the end points of the steel wire rope between two adjacent reliable matching points of each layer, and completing one-to-one correspondence and matching optimization of the end points of the steel wire rope.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110770909.5A CN113553926B (en) | 2021-07-08 | 2021-07-08 | Automatic detection method for drawing of steel wire rope core conveyer belt joint based on image matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110770909.5A CN113553926B (en) | 2021-07-08 | 2021-07-08 | Automatic detection method for drawing of steel wire rope core conveyer belt joint based on image matching |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113553926A CN113553926A (en) | 2021-10-26 |
| CN113553926B true CN113553926B (en) | 2023-08-11 |
Family
ID=78131418
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110770909.5A Active CN113553926B (en) | 2021-07-08 | 2021-07-08 | Automatic detection method for drawing of steel wire rope core conveyer belt joint based on image matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113553926B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117830416B (en) * | 2024-03-05 | 2024-05-17 | 山西戴德测控技术股份有限公司 | Method, device, equipment and medium for positioning abnormal position of conveying belt |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105911074A (en) * | 2016-04-07 | 2016-08-31 | 山西大学 | Calibration method for self-adaptive threshold in X-ray on-line detection of joint of wire-cored belt |
| CN106296700A (en) * | 2016-08-15 | 2017-01-04 | 南京工程学院 | Detection method twitched by a kind of steel cord conveyor belt joint |
| CN108550135A (en) * | 2018-03-07 | 2018-09-18 | 天津工业大学 | A kind of steel cable core conveying belt joint elongation automatic testing method based on x-ray image |
| CN110288562A (en) * | 2019-05-16 | 2019-09-27 | 枣庄学院 | A kind of steel cable core conveying belt joint twitch detection method based on x-ray image |
-
2021
- 2021-07-08 CN CN202110770909.5A patent/CN113553926B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105911074A (en) * | 2016-04-07 | 2016-08-31 | 山西大学 | Calibration method for self-adaptive threshold in X-ray on-line detection of joint of wire-cored belt |
| CN106296700A (en) * | 2016-08-15 | 2017-01-04 | 南京工程学院 | Detection method twitched by a kind of steel cord conveyor belt joint |
| CN108550135A (en) * | 2018-03-07 | 2018-09-18 | 天津工业大学 | A kind of steel cable core conveying belt joint elongation automatic testing method based on x-ray image |
| CN110288562A (en) * | 2019-05-16 | 2019-09-27 | 枣庄学院 | A kind of steel cable core conveying belt joint twitch detection method based on x-ray image |
| WO2020228111A1 (en) * | 2019-05-16 | 2020-11-19 | 枣庄学院 | X-ray image-based spasm detection method for steel cable core conveyor belt connector |
Non-Patent Citations (1)
| Title |
|---|
| 钢丝绳芯输送带接头抽动检测方法;韩现代等;《钢丝绳芯输送带接头抽动检测方法》;第第43卷卷(第第8期期);第67-71页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113553926A (en) | 2021-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110288562B (en) | Method for detecting joint twitching of steel wire rope core conveying belt based on X-ray image | |
| WO2023082418A1 (en) | Power utility tunnel settlement crack identification method based on artificial intelligence technology | |
| CN108550135B (en) | Automatic detection method for elongation of steel wire rope core conveyor belt joint based on X-ray image | |
| CN114581782B (en) | Fine defect detection method based on coarse-to-fine detection strategy | |
| CN113553926B (en) | Automatic detection method for drawing of steel wire rope core conveyer belt joint based on image matching | |
| CN110060251A (en) | A kind of building surface crack detecting method based on U-Net | |
| CN111079734B (en) | Method for detecting foreign matters in triangular holes of railway wagon | |
| CN116166983A (en) | Train bow net fault detection method based on machine learning | |
| CN114897855A (en) | Method for judging defect type based on X-ray picture gray value distribution | |
| CN112241574A (en) | Belt conveyor carrier roller state measurement and control method and device based on image texture information | |
| CN111738367A (en) | Part classification method based on image recognition | |
| CN114820626B (en) | Intelligent detection method for automobile front face part configuration | |
| CN106599894A (en) | Method for identifying pole number of overhead line system based on image identification | |
| CN115526515A (en) | Safety monitoring system of gate for water conservancy and hydropower | |
| CN107392216B (en) | Method for quickly identifying circumferential seams of shield tunnel segments based on gray data | |
| CN114612403A (en) | Intelligent detection method and system for breakage defect of feeding belt | |
| CN113333306B (en) | Sorting method and system for poor chip appearance | |
| CN112598680B (en) | Image segmentation method and system of sticky ore based on artificial intelligence network | |
| CN108470333B (en) | Method for screening and splicing X-ray steel wire rope core conveyor belt joint images | |
| CN115631146A (en) | Image-based pantograph carbon slide strip defect detection method | |
| CN114487014A (en) | Defect geometric feature extraction method for infrared nondestructive testing | |
| CN110598674B (en) | Point switch fault identification method based on pattern identification | |
| CN111325789B (en) | Curvature discontinuous point detection method based on discrete direction change sequence | |
| CN113793327A (en) | High-speed rail foreign matter detection method based on token | |
| CN113538418A (en) | Tire X-ray image defect extraction model construction method based on morphological analysis |
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 |