CN115330802B - Method for extracting debonding defect of X-ray image of carbon fiber composite gas cylinder - Google Patents
Method for extracting debonding defect of X-ray image of carbon fiber composite gas cylinder Download PDFInfo
- Publication number
- CN115330802B CN115330802B CN202211263930.7A CN202211263930A CN115330802B CN 115330802 B CN115330802 B CN 115330802B CN 202211263930 A CN202211263930 A CN 202211263930A CN 115330802 B CN115330802 B CN 115330802B
- Authority
- CN
- China
- Prior art keywords
- image
- defect
- gas cylinder
- gradient
- edge
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 121
- 238000000034 method Methods 0.000 title claims abstract description 42
- 229920000049 Carbon (fiber) Polymers 0.000 title claims abstract description 22
- 239000004917 carbon fiber Substances 0.000 title claims abstract description 22
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 title claims abstract description 22
- 239000002131 composite material Substances 0.000 title claims abstract description 17
- 238000012545 processing Methods 0.000 claims abstract description 25
- 238000000605 extraction Methods 0.000 claims abstract description 17
- 238000003708 edge detection Methods 0.000 claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims abstract description 5
- 239000011159 matrix material Substances 0.000 claims description 27
- 238000004364 calculation method Methods 0.000 claims description 14
- 238000002372 labelling Methods 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 6
- 230000007797 corrosion Effects 0.000 claims description 3
- 238000005260 corrosion Methods 0.000 claims description 3
- 230000002950 deficient Effects 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 19
- 239000007789 gas Substances 0.000 description 50
- 230000008569 process Effects 0.000 description 6
- 238000013135 deep learning Methods 0.000 description 4
- 230000000877 morphologic effect Effects 0.000 description 4
- 230000005855 radiation Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000011179 visual inspection Methods 0.000 description 3
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 2
- 238000010521 absorption reaction Methods 0.000 description 2
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 2
- 229910052782 aluminium Inorganic materials 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000009659 non-destructive testing Methods 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000009661 fatigue test Methods 0.000 description 1
- 239000001257 hydrogen Substances 0.000 description 1
- 229910052739 hydrogen Inorganic materials 0.000 description 1
- 125000004435 hydrogen atom Chemical class [H]* 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000005389 magnetism Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 229910052757 nitrogen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 238000004064 recycling Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000012360 testing method 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- 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
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10116—X-ray image
Abstract
The invention relates to the technical field of engineering detection, in particular to a method for extracting debonding defects of an X-ray image of a carbon fiber composite gas cylinder, which comprises the following steps: acquiring an original gas cylinder X-ray image, and preprocessing the original gas cylinder X-ray image to obtain a first processed image; obtaining a gas cylinder debonding defect image characteristic based on the first processing image, and performing debonding edge detection based on the gas cylinder debonding defect image characteristic to obtain a second processing image; and performing defect extraction on the second processed image, performing contour extraction on the extracted defect features to obtain defect contour features, and marking the defect contour features on the original gas cylinder X-ray image to obtain a final defect marking image. The invention can extract the debonding defect in the X-ray image of the composite carbon fiber gas cylinder, and conveniently distinguish the number, shape and other characteristics of the defect; meanwhile, the outline of the defect can be marked on the X-ray image of the composite material carbon fiber gas cylinder, and the position of the defect can be observed more intuitively.
Description
Technical Field
The invention relates to the technical field of engineering detection, in particular to a method for extracting debonding defects of an X-ray image of a carbon fiber composite gas cylinder.
Background
Gas cylinders are pressure vessels that store gases and are commonly used in the industry for the storage and transportation of gases such as hydrogen, oxygen, or nitrogen. It is characterized by convenient transportation and recycling. Because of the huge pressure generated inside the gas cylinder after filling with gas, once accidents occur, severe explosion is often generated, and immeasurable loss is caused. Therefore, our country has extremely high demands on the integrity, tightness and quality of the gas cylinder. There are many methods for detecting the quality of the gas cylinder, such as analysis of chemical components of materials, detection of mechanical properties of the inner container, fatigue test, etc., and the detection procedures are complicated and require a lot of time and money. In addition, the detection means have certain destructiveness, are only suitable for sampling detection of the gas cylinders, and cannot guarantee the quality safety of each gas cylinder. In recent years, with the progress of nondestructive testing technology, nondestructive testing is becoming the main method for detecting the quality of gas cylinders.
Nondestructive detection is to detect physical and chemical parameters of a detected object by utilizing technologies such as rays, ultrasonic waves, infrared rays, electromagnetic waves and the like under the premise of not affecting the service performance of the detected object and not damaging the internal structure of the detected object through changes such as light, sound, heat, electricity, magnetism and the like caused by abnormal tissue structure or defects of the detected object. Compared with the detection mode, the nondestructive detection has the advantages that physical damage to the gas cylinder is avoided to the greatest extent in the detection process, the integrity of the gas cylinder tissue structure can be guaranteed, the performance of the gas cylinder cannot be influenced, and the detected gas cylinder can be used continuously. The nondestructive test has the advantages of high detection speed, low requirement on detection environment, capability of on-site detection on production sites and the like.
Nondestructive X-ray detection is a technique of transilluminating an object to be inspected with X-rays having a uniform intensity and imaging to display defects inside the object. Because of the defect areas with different radiation absorption degrees in the object, the intensity of the radiation transmitted through the object presents a state of uneven distribution, and the sensor on the back of the object receives the radiation and generates images with different gray values. In general, the thinner the region, the lower the extent of radiation absorption, and the higher the gray value of the image.
In the process of positioning defects through nondestructive detection images of carbon fiber gas cylinders, a manual visual inspection method and a deep learning-based method are often adopted for defect positioning. The manual visual inspection method refers to that a worker judges the type and the position of the defect through naked eyes by virtue of own experience, and marks the defect on an image manually. The deep learning-based method uses a large number of marked images and uses algorithms such as fast-RCNN and YOLO to extract the characteristics of the defect images and automatically marks the images.
Currently, the disadvantages of manual visual inspection are:
1. the identification of defective areas in the cylinder image is dependent on the experience of workers, and is severely insufficient in accuracy and speed of defect identification.
2. The standards of each person for defects may vary, making it difficult to formulate uniform discriminant standards.
The defect positioning method based on deep learning has the following defects:
1. in the process of training a model, the required data volume is large, and the probability of defects generated by a normal gas cylinder is low, so that enough data is difficult to provide for training.
2. The deep learning method takes longer time in the process of training the model, and requires professional software engineers to carry out parameter adjustment and repeated experiments, so that the later maintenance cost is higher.
Disclosure of Invention
The invention aims to provide a method for positioning debonding defects of a carbon fiber composite gas cylinder image, which is used for different carbon fiber gas cylinder images, and simultaneously carries out morphological treatment on the images to better display the positions of the defects.
In order to achieve the above object, the present invention provides the following solutions:
a method for extracting debonding defects of an X-ray image of a carbon fiber composite gas cylinder comprises the following steps:
acquiring an original gas cylinder X-ray image, and preprocessing the original gas cylinder X-ray image to obtain a first processed image;
acquiring a gas cylinder debonding defect image characteristic based on the first processing image, and performing debonding edge detection based on the gas cylinder debonding defect image characteristic to acquire a second processing image;
and performing defect extraction on the second processed image, performing contour extraction on the extracted defect features to obtain defect contour features, and labeling the defect contour features on the original gas cylinder X-ray image to obtain a final defect labeling image.
Preferably, preprocessing the raw cylinder X-ray image includes: and carrying out graying treatment on the original gas cylinder X-ray image to obtain a gray image, wherein the gray image is the first treatment image.
Preferably, obtaining the gas cylinder debonding defect image feature based on the first processed image includes:
denoising the first processed image to obtain a gray matrix of the original gas cylinder X-ray image; performing gradient calculation processing on the denoised image to obtain a gradient matrix of the original gas cylinder X-ray image;
and performing debonding edge detection on the gray matrix and the gradient matrix to obtain the image characteristics of the gas cylinder debonding defect.
Preferably, denoising the first processed image includes:
initializing a Gaussian convolution template through a Gaussian function, converting the first processed image into a gray value matrix, and expanding the gray image by using a zero filling method;
and based on the Gaussian convolution template, carrying out weighted average calculation on the gray values subjected to zero filling processing in a sliding window mode to obtain a denoising image.
Preferably, performing gradient computation processing on the denoised image includes:
selecting a gradient operator template, calculating gradient values of all pixel points in the gradient operator template, and calculating gradient directions through the gradient values to obtain a gradient matrix of the original gas cylinder X-ray image; wherein the gradient operator template is a Sobel one-step gradient operator template.
Preferably, the method for calculating the gradient direction is as follows:
wherein,is the direction of the gradient,to use Sobel one step degreeOperator calculatedyThe gradient value in the direction is set,calculated for using Sobel one-step degree operatorxGradient values in the direction.
Preferably, obtaining the second processed image includes:
setting a gray level threshold value and a gradient threshold value according to the gray level matrix and the gradient matrix and combining gray level information of the original gas cylinder X-ray image;
traversing the original gas cylinder X-ray image based on the gradient threshold value and the gray threshold value, and determining the edge of a liner layer of the gas cylinder, wherein the edge of the liner layer comprises a first edge and a second edge;
calculating the gray average value of the middle area of the first edge and the second edge, and marking the gray average value as a strong edge if the gray average value is larger than the gray value of the first edge and the second edge in the horizontal direction, otherwise marking the gray average value as a weak edge;
and performing neighborhood traversal on the strong edge, setting a gradient threshold value, and communicating the strong edge with a weak edge with a distance of one pixel point to obtain a complete edge image, namely the second processed image.
Preferably, the defect extraction of the second processed image includes:
performing binarization processing on the second processed image to obtain a binary image, and performing corrosion operation based on the binary image to obtain a refined edge image;
performing a first closing operation based on the thinned edge image to obtain a first defect area image, and performing a second closing operation on the first defect area image to obtain a second defect area image;
performing difference operation on the first defect area image and the second defect area image to obtain a defect area image;
wherein the structural elements of the first closed operation are smaller than the structural elements of the second closed operation.
Preferably, the contour extraction of the defect area image includes:
setting the minimum side length, the minimum perimeter and the minimum area of the defect area image, and removing the defect area outline which does not meet the conditions after screening to obtain a determined defect area outline;
and drawing the determined outline of the defect area on the original gas cylinder X-ray image in a rectangular frame form and displaying the outline of the determined defect area to obtain the final defect labeling image.
The beneficial effects of the invention are as follows:
1. the invention can extract the debonding defect in the X-ray image of the composite carbon fiber gas cylinder, and conveniently distinguish the number, shape and other characteristics of the defect; meanwhile, the outline of the defect can be marked on the X-ray image of the composite material carbon fiber gas cylinder, so that the position of the defect can be observed more intuitively;
2. the invention uses the self-defined Gaussian filter to carry out denoising treatment on the gray level image, thereby realizing better denoising effect, eliminating the influence of noise generated in the process of collecting the X-ray image on the identification and extraction of image defects, and solving the problem of inaccurate image edge positioning and extraction caused by noise interference when the image is processed;
3. according to the invention, the self-defined gradient operator method is utilized to perform gradient calculation processing on the denoising image, so that the gradient characteristics of the image can be accurately extracted, and the edge area is easier to position;
4. according to the invention, the edge detection and binarization processing are carried out on the obtained de-noised image by using the edge detection method based on the self-adaptive gradient-gray threshold value, and the two side edges of the de-sticking area are determined according to the gradient-gray information, so that the problems that in the prior art, for example, in the intelligent image defect identification and analysis method using Canny edge detection, the false edge area is too many, the edge of the defect area cannot be accurately positioned, and the false detection and the difficult positioning of the defect area are easy to occur are solved;
5. according to the invention, the extracted contour is screened by using a contour screening method based on the side length and the area, so that the problem that the area and the side length of a contour area cannot be screened in the prior art, such as an OpenCV contour searching method, is solved, the false detection condition is reduced, and more accurate defect contour extraction is realized.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings that are needed in the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present 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 flow chart of a method for positioning debonding defects of an image of a carbon fiber composite gas cylinder according to an embodiment of the present invention;
FIG. 2 is a denoised image of an embodiment of the present invention;
FIG. 3 is a first edge image obtained in an embodiment of the present invention;
FIG. 4 is a second edge image obtained in an embodiment of the present invention;
FIG. 5 is a morphologically processed image according to an embodiment of the invention;
FIG. 6 is a final image of a defect label in an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
The invention combines manual priori knowledge and a self-adaptive algorithm to identify the defects of the gas cylinder.
1. Based on a priori knowledge, the position of the gas cylinder debonding defect is determined, wherein the defect only exists in the area between the aluminum liner layer and the carbon fiber composite material layer. Aiming at the characteristics of the debonding region, the image gray scale characteristics of the debonding defect region are extracted, for example, the aluminum liner layer is thicker, and the image gray scale value is lower due to the fact that X rays are difficult to penetrate. The debonded area is not tightly combined, and a gap which can be penetrated by X rays is reserved, so that the gray value of an image is higher, and the whole image has the characteristics of low gray values at two sides and high gray value in the middle part. The characteristic is summarized by using priori knowledge, and the recognition accuracy of the debonded area is greatly improved by the algorithm.
2. By using the edge detection method of the self-adaptive gray-gradient threshold, the threshold is automatically set for different characteristics of each picture, and a good detection effect can still be obtained under the condition of insufficient defect sample size.
Referring to fig. 1, the embodiment provides a method for extracting a debonding defect of an X-ray image of a carbon fiber composite gas cylinder, which includes the following steps:
1. acquiring an X-ray image to be processed;
2. graying treatment is carried out on the obtained X-ray image, so that a treated gray image, namely a first treatment image, is obtained;
3. obtaining a gas cylinder debonding defect image characteristic based on the first processing image, and performing debonding edge detection based on the gas cylinder debonding defect image characteristic to obtain a second processing image;
the obtaining the gas cylinder debonding defect image feature based on the first processed image includes:
denoising the first processed image to obtain a gray matrix of the original gas cylinder X-ray image; performing gradient calculation processing on the denoised image to obtain a gradient matrix of the original gas cylinder X-ray image;
and performing debonding edge detection on the gray matrix and the gradient matrix to obtain the image characteristics of the gas cylinder debonding defect.
4. Performing defect extraction on the image subjected to edge detection processing by using a morphological method;
5. and carrying out contour extraction on the extracted defect features, and marking the extracted defect features on the original X-ray image.
Denoising the obtained gray image, wherein the denoising method comprises the following steps of:
(1) A gaussian convolution template of 3*3 is initialized using a gaussian function.
(2) The entire image is converted into a form represented by a gray value matrix and the image is expanded using zero padding.
(3) And carrying out weighted average calculation on the gray values of the image subjected to zero filling processing in a sliding window mode by using the generated Gaussian convolution template to obtain a Gaussian filtered gray value matrix.
The weight calculation formula of the Gaussian template is as follows:
wherein,and (5) the coefficients of the corresponding positions of the calculated Gaussian templates.Is the standard deviation of the two-dimensional image,is a natural constant.In order to use the center point as the origin of coordinates, the X-axis and Y-axis coordinates of the relative positions are included in the coordinate system of the X-axis and Y-axis.
Further, the denoising image gradient calculation process comprises the following steps:
(1) And selecting a gradient operator template, and calculating gradient values of all pixel points.
(2) The gradient direction is determined.
The calculation formula for calculating the gradient value in the x direction by using the Sobel one-step gradient operator is as follows:
the calculation formula for calculating the gradient value in the y direction by using the Sobel one-step gradient operator is as follows:
wherein,is the gray value of the corresponding point of the image.
The total gradient calculation formula is:
wherein,to calculate the gradient value in the x-direction using the Sobel one-step gradient operator,gradient values in the y-direction calculated for using the Sobel one-step gradient operator.
The gradient direction calculation formula is:
the obtained denoised image is shown in figure 2;
further, the image gradient matrix and the gray matrix are used for debonding edge detection, which comprises the following steps:
(1) According to the characteristics of the X-ray image, setting gray level and gradient threshold values according to gray level information of the image, traversing the image, and determining the edge of the inner container layer of the gas cylinder.
(2) Near the first edge, a carbon fiber layer edge with a larger gradient value and a gradient direction opposite to the first edge is found.
(3) And calculating a gray average value of the middle area of the two edges, and ensuring that the gray average value of the debonded area is larger than that of the two edges in the horizontal direction, otherwise, marking the edges as weak edges.
(4) And 8, performing 8-neighborhood traversal on the determined strong edge, and setting a gradient threshold value to communicate the strong edge with surrounding weak edges to obtain a complete edge image, namely a second processed image.
The obtained first and second edge images are shown in fig. 3 and 4.
Further, defect feature extraction is performed on the image after edge detection processing by using a morphological method, and the method comprises the following steps:
(1) Performing binarization processing on the image with the determined edge, setting the gray value of the edge to 255 and setting the gray value of other areas to 0;
(2) And (3) performing corrosion operation on the binarized image by using the structural elements with the size of 3*3, and eliminating isolated points and possible noise points in the image to obtain a thinned edge image.
(3) Performing a closing operation on the thinned edge image by using 5*5 structural elements, and filling the tiny defect part under the condition of keeping the edge of the image unchanged to obtain a defect area image M after preliminary processing;
(4) Performing a closing operation on the defect area image M obtained in the previous step by using structural elements with the size of 15 x 15, and filling a larger defect part under the condition that the edge of the image is kept unchanged greatly to obtain a defect area image N after further processing;
(5) Performing a difference operation on the images M and N obtained in the steps (4) and (3) to obtain an accurate defect area image Q;
the morphologically processed image is shown in fig. 5.
Further, the contour extraction is performed on the extracted defect area image Q, and the extracted defect area is marked on the original X-ray image, which comprises the following steps:
(1) And extracting the outline of the defective area image, and storing the outline position in an array form.
(2) The minimum side length, the minimum perimeter and the minimum area of the defect area are set.
(3) Traversing the defect area according to the set threshold value, and screening out the outline of the defect area which does not meet the condition.
The set threshold includes three constraints, a minimum side length, a minimum perimeter, and a minimum area. When a defect area is defined, fitting is carried out by using rectangular frames, for the rectangular frames which are fit by the defect area, firstly, the minimum side length is used as a screening condition, then, the rectangular frames with the minimum side length of the rectangle being larger than a set threshold value are screened by the minimum circumference, and finally, the minimum area screening is carried out. The remaining rectangular box is the determined defect area outline.
(4) And drawing and displaying the outline of the determined defect area on the original image in a rectangular frame form.
The final defect labeling image is shown in fig. 6.
The method can be used for different carbon fiber gas cylinder images, has a good detection effect, and can better display the defect positions by morphological processing of the images.
The above embodiments are merely illustrative of the preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, but various modifications and improvements made by those skilled in the art to which the present invention pertains are made without departing from the spirit of the present invention, and all modifications and improvements fall within the scope of the present invention as defined in the appended claims.
Claims (3)
1. The method for extracting the debonding defect of the X-ray image of the carbon fiber composite gas cylinder is characterized by comprising the following steps of:
acquiring an original gas cylinder X-ray image, and preprocessing the original gas cylinder X-ray image to obtain a first processed image, wherein the original gas cylinder X-ray image is a carbon fiber composite gas cylinder X-ray image;
acquiring a gas cylinder debonding defect image characteristic based on the first processing image, and performing debonding edge detection based on the gas cylinder debonding defect image characteristic to acquire a second processing image;
performing defect extraction on the second processed image, performing contour extraction on the extracted defect features to obtain defect contour features, and labeling the defect contour features on the original gas cylinder X-ray image to obtain a final defect labeling image;
preprocessing the original gas cylinder X-ray image comprises the following steps: graying treatment is carried out on the original gas cylinder X-ray image to obtain a gray image, wherein the gray image is the first treatment image;
the obtaining the gas cylinder debonding defect image feature based on the first processed image includes:
denoising the first processed image to obtain a gray matrix of the original gas cylinder X-ray image; performing gradient calculation processing on the denoised image to obtain a gradient matrix of the original gas cylinder X-ray image;
performing debonding edge detection on the gray matrix and the gradient matrix to obtain the image characteristics of the gas cylinder debonding defect;
denoising the first processed image, including:
initializing a Gaussian convolution template through a Gaussian function, converting the first processed image into a gray value matrix, and expanding the gray image by using a zero filling method;
based on the Gaussian convolution template, carrying out weighted average calculation on gray values subjected to zero filling treatment in a sliding window mode to obtain a denoising image;
obtaining the second processed image includes:
setting a gray level threshold value and a gradient threshold value according to the gray level matrix and the gradient matrix and combining gray level information of the original gas cylinder X-ray image;
traversing the original gas cylinder X-ray image based on the gradient threshold value and the gray threshold value, and determining the edge of a liner layer of the gas cylinder, wherein the edge of the liner layer comprises a first edge and a second edge, and the second edge is a carbon fiber layer edge with a large gradient value and a gradient direction opposite to the first edge;
calculating the gray average value of the middle area of the first edge and the second edge, and marking the gray average value as a strong edge if the gray average value is larger than the gray value of the first edge and the second edge in the horizontal direction, otherwise marking the gray average value as a weak edge;
performing neighborhood traversal on the strong edge, setting a gradient threshold value to communicate the strong edge with a weak edge with a distance of one pixel point to obtain a complete edge image, namely the second processed image;
performing defect extraction on the second processed image, including:
performing binarization processing on the second processed image to obtain a binary image, and performing corrosion operation based on the binary image to obtain a refined edge image;
performing a first closing operation based on the thinned edge image to obtain a first defect area image, and performing a second closing operation on the first defect area image to obtain a second defect area image;
performing difference operation on the first defect area image and the second defect area image to obtain a defect area image;
wherein the structural elements of the first closed operation are smaller than the structural elements of the second closed operation;
the gradient calculation processing of the denoising image comprises the following steps:
selecting a gradient operator template, calculating gradient values of all pixel points in the gradient operator template, and calculating gradient directions through the gradient values to obtain a gradient matrix of the original gas cylinder X-ray image; wherein the gradient operator template is a Sobel one-step gradient operator template;
the gradient value calculating method comprises the following steps:
G x (x,y)＝g(x+1,y-1)+2*g(x+1,y)+g(x+1,y+1)-g(x-1,y-1)-2*g(x-1,y)-g(x-1,y+1)
G y (x,y)＝g(x-1,y+1)+2*g(x,y+1)+g(x+1,y+1)-g(x-1,y-1)-2*g(x,y-1)-g(x+1,y-1)
wherein g (x, y) is the imageThe value of the gray matrix at coordinates (x, y) obtained after noise, G x (x, y) is the gradient value of the image in the x direction, G y (x, y) is the gradient value of the image in the y direction;
calculating the value of the gradient matrix at coordinates (x, y) according to the change degree of each pixel point of the image in the x and y directions:
wherein G is (x,y) Is the value of the image gradient matrix at coordinates (x, y).
2. The method for extracting the debonding defect of the X-ray image of the carbon fiber composite gas cylinder according to claim 1, wherein the method for calculating the gradient direction is as follows:
wherein θ is the gradient direction, G y (x, y) is the gradient value in the y direction calculated using Sobel one-step gradient operator, G x (x, y) is the gradient value in the x-direction calculated using the Sobel one-step gradient operator.
3. The method for extracting the debonding defect of the X-ray image of the carbon fiber composite gas cylinder according to claim 1, wherein the contour extraction of the defective region image comprises:
setting the minimum side length, the minimum perimeter and the minimum area of the defect area image, and removing the defect area outline which does not meet the conditions after screening to obtain a determined defect area outline;
and drawing the determined outline of the defect area on the original gas cylinder X-ray image in a rectangular frame form and displaying the outline of the determined defect area to obtain the final defect labeling image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211263930.7A CN115330802B (en) | 2022-10-17 | 2022-10-17 | Method for extracting debonding defect of X-ray image of carbon fiber composite gas cylinder |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211263930.7A CN115330802B (en) | 2022-10-17 | 2022-10-17 | Method for extracting debonding defect of X-ray image of carbon fiber composite gas cylinder |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115330802A CN115330802A (en) | 2022-11-11 |
| CN115330802B true CN115330802B (en) | 2024-01-19 |
Family
ID=83915258
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211263930.7A Active CN115330802B (en) | 2022-10-17 | 2022-10-17 | Method for extracting debonding defect of X-ray image of carbon fiber composite gas cylinder |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115330802B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116823924B (en) * | 2023-08-24 | 2023-12-12 | 杭州百子尖科技股份有限公司 | Determination method and device for defect area, electronic equipment and storage medium |
| CN116958714B (en) * | 2023-09-20 | 2023-12-01 | 信熙缘(江苏)智能科技有限公司 | Automatic identification method for wafer back damage defect |
| CN117437224B (en) * | 2023-12-20 | 2024-03-29 | 山东特联信息科技有限公司 | Gas cylinder defect image data processing system and method based on artificial intelligence |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107870181A (en) * | 2017-06-20 | 2018-04-03 | 成都飞机工业（集团）有限责任公司 | A kind of later stage recognition methods of composite debonding defect |
| CN108765436A (en) * | 2018-04-21 | 2018-11-06 | 卞家福 | Method for detecting image edge is piled up based on the irregular beverage bottle for improving Roberts operators |
| CN109142396A (en) * | 2018-08-30 | 2019-01-04 | 湖北三江航天江北机械工程有限公司 | A kind of layering of carbon fiber winding shell, debonding defect detecting appraisal method |
| CN109978869A (en) * | 2019-03-29 | 2019-07-05 | 清华大学 | A kind of sea horizon detection method and system based on gray level co-occurrence matrixes and Hough transform |
| CN112819775A (en) * | 2021-01-28 | 2021-05-18 | 中国空气动力研究与发展中心超高速空气动力研究所 | Segmentation and reinforcement method for damage detection image of aerospace composite material |
| CN113592782A (en) * | 2021-07-07 | 2021-11-02 | 山东大学 | Method and system for extracting X-ray image defects of composite carbon fiber core rod |
| WO2022148192A1 (en) * | 2021-01-07 | 2022-07-14 | 新东方教育科技集团有限公司 | Image processing method, image processing apparatus, and non-transitory storage medium |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9453500B2 (en) * | 2013-03-15 | 2016-09-27 | Digital Wind Systems, Inc. | Method and apparatus for remote feature measurement in distorted images |
| CN106780438B (en) * | 2016-11-11 | 2020-09-25 | 广东电网有限责任公司清远供电局 | Insulator defect detection method and system based on image processing |
| CN108444934B (en) * | 2018-01-30 | 2021-12-10 | 四川沐迪圣科技有限公司 | Automatic segmentation and quantification method for debonding defect of composite material |
| FR3086434B1 (en) * | 2018-09-26 | 2020-09-04 | Safran | METHOD AND SYSTEM FOR NON-DESTRUCTIVE INSPECTION OF AN AERONAUTICAL PART BY CONTOUR REPLACEMENT |
| CN109900742B (en) * | 2019-04-03 | 2019-12-17 | 哈尔滨商业大学 | device and method for detecting debonding defect of carbon fiber composite material in linear and nonlinear frequency modulation hybrid excitation refrigeration mode |
| CN114581364A (en) * | 2021-12-15 | 2022-06-03 | 宁波送变电建设有限公司 | GIS defect detection method based on X-ray imaging and Sobel-SCN |
-
2022
- 2022-10-17 CN CN202211263930.7A patent/CN115330802B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107870181A (en) * | 2017-06-20 | 2018-04-03 | 成都飞机工业（集团）有限责任公司 | A kind of later stage recognition methods of composite debonding defect |
| CN108765436A (en) * | 2018-04-21 | 2018-11-06 | 卞家福 | Method for detecting image edge is piled up based on the irregular beverage bottle for improving Roberts operators |
| CN109142396A (en) * | 2018-08-30 | 2019-01-04 | 湖北三江航天江北机械工程有限公司 | A kind of layering of carbon fiber winding shell, debonding defect detecting appraisal method |
| CN109978869A (en) * | 2019-03-29 | 2019-07-05 | 清华大学 | A kind of sea horizon detection method and system based on gray level co-occurrence matrixes and Hough transform |
| WO2022148192A1 (en) * | 2021-01-07 | 2022-07-14 | 新东方教育科技集团有限公司 | Image processing method, image processing apparatus, and non-transitory storage medium |
| CN112819775A (en) * | 2021-01-28 | 2021-05-18 | 中国空气动力研究与发展中心超高速空气动力研究所 | Segmentation and reinforcement method for damage detection image of aerospace composite material |
| CN113592782A (en) * | 2021-07-07 | 2021-11-02 | 山东大学 | Method and system for extracting X-ray image defects of composite carbon fiber core rod |
Non-Patent Citations (8)
| Title |
|---|
| 基于数学形态学的焊接熔池图像边缘提取技术研究;武晓朦;吴凯;王欢;封园;;西安石油大学学报(自然科学版)(03);第115页第3.1-3.2节，图1-3 * |
| 基于灰度-梯度二维最大熵阈值法的赤足迹轮廓提取;李孟歆;贾燕雯;姜佳楠;;电子技术与软件工程(第16期);全文 * |
| 基于灰度-梯度共生矩阵的木材表面缺陷分割方法;白雪冰;邹丽晖;;森林工程(第02期);全文 * |
| 基于灰度-梯度共生矩阵的钢轨表面缺陷检测方法;陈后金;许文达;郝晓莉;;北京交通大学学报(02);第2.1节 * |
| 梯度图像的阈值选取;韩冰;高俊钗;雷鸣;;科学技术与工程(第16期);全文 * |
| 直升机旋翼粘弹阻尼器射线DR/CT检测技术研究;张宝双;《万方学位论文》;全文 * |
| 碳纤维复合材料气瓶的X射线实时成像技术;史建军;李得天;;宇航材料工艺(第06期);全文 * |
| 自适应Canny算法在钢板缺陷边缘检测中的应用;于海川;穆平安;;软件导刊(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115330802A (en) | 2022-11-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115330802B (en) | Method for extracting debonding defect of X-ray image of carbon fiber composite gas cylinder | |
| US11221107B2 (en) | Method for leakage detection of underground pipeline corridor based on dynamic infrared thermal image processing | |
| CN113469177B (en) | Deep learning-based drainage pipeline defect detection method and system | |
| CN111209876B (en) | Oil leakage defect detection method and system | |
| Lin et al. | Detection of a casting defect tracked by deep convolution neural network | |
| CN110148130B (en) | Method and device for detecting part defects | |
| CN111047568B (en) | Method and system for detecting and identifying steam leakage defect | |
| CN111325721A (en) | Gas leakage detection method and system based on infrared thermal imaging | |
| CN108846397B (en) | Automatic detection method for cable semi-conducting layer based on image processing | |
| CN111369516B (en) | Transformer bushing heating defect detection method based on infrared image recognition | |
| CN105510364B (en) | The nondestructive detection system and its detection method of industrial part defect based on X-ray | |
| CN110097547A (en) | A kind of automatic testing method that the weld seam egative film based on deep learning is faked | |
| CN112862744B (en) | Intelligent detection method for internal defects of capacitor based on ultrasonic image | |
| Mery et al. | Image processing for fault detection in aluminum castings | |
| CN111257329A (en) | Smartphone camera defect detection method and detection system | |
| CN112730454A (en) | Intelligent damage detection method for composite material based on fusion of optics, infrared thermal waves and ultrasonic waves | |
| CN112258444A (en) | Elevator steel wire rope detection method | |
| CN114719749B (en) | Metal surface crack detection and real size measurement method and system based on machine vision | |
| CN116823822A (en) | Ship sheet metal part welding defect detection method based on visual characteristics | |
| CN115601359A (en) | Welding seam detection method and device | |
| Peng et al. | Non-uniform illumination image enhancement for surface damage detection of wind turbine blades | |
| CN110084807A (en) | A kind of detection method that seam inspection egative film image is faked | |
| Ghamisi et al. | Anomaly detection in automated fibre placement: learning with data limitations | |
| CN114742823A (en) | Intelligent detection method for scratches on surface of object | |
| KR101782366B1 (en) | Vision inspection method based on learning data using normalization of sample scale |
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 |