CN111681217A - Machine vision image intelligent analysis method, device and system for fatigue fracture - Google Patents

Machine vision image intelligent analysis method, device and system for fatigue fracture Download PDF

Info

Publication number
CN111681217A
CN111681217A CN202010478691.1A CN202010478691A CN111681217A CN 111681217 A CN111681217 A CN 111681217A CN 202010478691 A CN202010478691 A CN 202010478691A CN 111681217 A CN111681217 A CN 111681217A
Authority
CN
China
Prior art keywords
image
fracture
metal
crack
fatigue
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.)
Granted
Application number
CN202010478691.1A
Other languages
Chinese (zh)
Other versions
CN111681217B (en
Inventor
黎伟华
杨浩坤
梁皓
陈呼龙
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hong Kong Productivity Council
Original Assignee
Hong Kong Productivity Council
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hong Kong Productivity Council filed Critical Hong Kong Productivity Council
Priority to CN202010478691.1A priority Critical patent/CN111681217B/en
Publication of CN111681217A publication Critical patent/CN111681217A/en
Application granted granted Critical
Publication of CN111681217B publication Critical patent/CN111681217B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
    • 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
    • G06T2207/10004—Still image; Photographic image
    • 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
    • G06T2207/10024—Color image
    • 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
    • G06T2207/20036—Morphological image processing
    • 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
    • G06T2207/20081—Training; Learning
    • 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
    • G06T2207/20084—Artificial neural networks [ANN]
    • 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
    • G06T2207/30164—Workpiece; Machine component
    • 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)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Image Analysis (AREA)

Abstract

The invention provides a method, a device and a system for intelligently analyzing machine vision images of fatigue fractures, wherein the method comprises the following steps: obtaining a metal fracture image and a metal source image; analyzing the metal fracture image based on the deep learning convolution neural network to determine the metal fracture type; when the metal fracture type is a fatigue fracture type, distinguishing the metal fracture image to obtain a crack expansion area and a crack instantaneous fracture area; carrying out statistical analysis on the crack expansion area and the crack instantaneous fracture area; and simulating the forming process of the metal fracture based on an image deformation algorithm according to the metal fracture image and the metal source image, and simulating to generate a metal fracture forming process image. According to the scheme, the metal fracture image is analyzed through a complete artificial intelligence image analysis technology, so that the risk of accuracy reduction caused by long-term work of personnel can be avoided, and the cost of analysis working time is reduced; the cost investment of metal fatigue fracture failure analysis is reduced; and a high-efficiency and reliable fatigue fracture analysis result can be obtained.

Description

Machine vision image intelligent analysis method, device and system for fatigue fracture
Technical Field
The invention relates to the technical field of metal fatigue analysis, in particular to a method, a device and a system for intelligently analyzing machine vision images of fatigue fractures.
Background
Mechanical equipment is an important tool in the fields of life, work, construction and the like of people, and metal materials are widely used on the mechanical equipment due to the advantages of high strength and processability. However, in the operation process of the equipment, a component machined from a metal material is subjected to the action of alternate force load, a fatigue crack source is gradually formed in the material or on the surface of the material, a crack tip gradually expands to form a crack expansion area, and the effective section of a sample is reduced, so that the instantaneous fracture area cannot bear the external force load. Fatigue failure of metal materials can lead to damage to metal components and the shutting down of machinery, causing property and personal injury. By analyzing the fatigue fracture of the metal material, the distribution of a crack initiation area, an expansion area and a transient fracture area of the metal component on the fatigue fracture can be known, the failure mechanism of the component in the service process is deduced according to the three fracture characteristic areas, the failure reason is found out to avoid the secondary damage of other similar equipment, or the failure reason is used as an analysis evidence to discuss property loss. Reliable fatigue fracture failure analysis requires experienced professionals and the analysis process takes a significant amount of time and material. Considering the high labor cost of professionals and the fatigue caused by long-time work, the accuracy of the failure analysis is seriously hampered.
In order to improve the failure analysis work efficiency of the fatigue fracture and reduce the cost, an artificial intelligence analysis scheme is provided, the scheme can improve the failure analysis work efficiency, avoid error analysis caused by personnel fatigue, and finally achieve the purposes of high efficiency and low cost of the failure analysis. At present, artificial intelligence image analysis is widely applied to computer image recognition, the application of the image processing digitization technologies covers various technologies, such as common technologies of image enhancement, image weakening and the like, and by carrying out discussion and research on a computer image analysis algorithm, favorable guarantee can be provided for an image processing effect, and the definition and the resolution of an image are practically improved. Typical analysis methods include OTSU thresholding, which is widely used in medical imaging; the k-means clustering method is commonly used for clustering analysis in data mining at present. Hough transform is also a popular technique in image processing. If a shape can be represented mathematically, it can be used to detect any shape. Even if the image shape is slightly distorted or destroyed, it can detect the shape therefrom. However, no complete artificial intelligence image analysis technology is available for analyzing metal fracture images.
Disclosure of Invention
The embodiment of the invention provides a method, a device and a system for intelligently analyzing machine vision images of fatigue fractures, and solves the technical problem that a complete artificial intelligence image analysis technology is lacked in the prior art to analyze metal fracture images.
The embodiment of the invention provides a machine vision image intelligent analysis method of a fatigue fracture, which comprises the following steps:
obtaining a metal fracture image and a metal source image;
analyzing the metal fracture image based on a deep learning convolution neural network to determine the type of the metal fracture;
when the metal fracture type is a fatigue fracture type, distinguishing the metal fracture image to obtain a crack expansion area and a crack instantaneous fracture area;
carrying out statistical analysis on the crack expansion area and the crack instantaneous fracture area;
and simulating the forming process of the metal fracture based on an image deformation algorithm according to the metal fracture image and the metal source image, and simulating to generate a metal fracture forming process image.
The embodiment of the invention also provides a machine vision image intelligent analysis device of the fatigue fracture, which comprises the following components:
the image obtaining module is used for obtaining a metal fracture image and a metal source image;
the metal fracture type determining module is used for analyzing the metal fracture image based on a deep learning convolution neural network and determining the metal fracture type;
the crack expansion area and crack snap-off area determining module is used for distinguishing the metal fracture image when the metal fracture type is a fatigue fracture type to obtain a crack expansion area and a crack snap-off area;
the statistical analysis module is used for performing statistical analysis on the crack expansion area and the crack instantaneous fracture area;
and the metal fracture formation process simulation module is used for simulating the formation process of the metal fracture based on an image deformation algorithm according to the metal fracture image and the metal source image, and simulating to generate a metal fracture formation process image.
The embodiment of the invention also provides a machine vision image intelligent analysis system of the fatigue fracture, which comprises the following steps: a body type microscope and a machine vision image intelligent analysis device of the fatigue fracture;
the body microscope is used for: shooting a metal fracture image and a metal source image, and sending the metal fracture image to a machine vision image intelligent analysis device of the fatigue fracture.
The embodiment of the present invention further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the method when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium, and the computer readable storage medium stores a computer program for executing the method.
In the embodiment of the invention, the metal fracture image is analyzed based on the deep learning convolution neural network, and the metal fracture type is determined; when the metal fracture type is a fatigue fracture type, distinguishing the metal fracture image to obtain a crack expansion area and a crack instantaneous fracture area; and carrying out statistical analysis on the crack expansion area and the crack instantaneous fracture area, simulating the forming process of the metal fracture based on an image deformation algorithm according to the metal fracture image and the metal source image, and simulating to generate a metal fracture forming process image. Therefore, the metal fracture image is analyzed through a complete artificial intelligence image analysis technology, the risk of accuracy reduction caused by long-term work of personnel can be avoided, and the cost of analysis working time is reduced; the cost investment of metal fatigue fracture failure analysis is reduced; and a high-efficiency and reliable fatigue fracture analysis result can be obtained.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or 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 for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flowchart of a method for intelligently analyzing a machine vision image of a fatigue fracture according to an embodiment of the present invention;
fig. 2 is a flowchart of classifying metal fracture categories based on a deep learning Convolutional Neural Network (CNN) according to an embodiment of the present invention;
FIG. 3 is a flowchart for distinguishing an extended area from a transient fracture area of a metal fatigue fracture image according to an embodiment of the present invention;
fig. 4 is a specific schematic diagram of a forming process for simulating generation of a fracture metal based on an image deformation algorithm (Morphing), according to an embodiment of the present invention;
FIG. 5 is a source image I provided by an embodiment of the present inventionSAnd a target image IDA schematic diagram;
FIG. 6 is a sample artwork (under an industrial magnifier) provided by an embodiment of the present invention;
FIG. 7 illustrates a fracture image contour detection and background segmentation (adaptive binary threshold detection) according to an embodiment of the present invention;
FIG. 8 is a schematic diagram of crack propagation regions (standard samples) obtained using the original method;
FIG. 9 is a schematic view of a crack propagation region obtained by the method of the present invention according to an embodiment of the present invention;
FIG. 10 is a schematic diagram of a crack snap zone (standard sample) obtained using the original method;
FIG. 11 is a schematic diagram of a crack transient region obtained by the method of the present invention according to an embodiment of the present invention;
fig. 12 is a broken metal diagram obtained by simulating a forming process of generating a broken metal based on an image deformation algorithm (Morphing), according to an embodiment of the present invention;
fig. 13 is a structural block diagram of a machine vision image intelligent analysis device for fatigue fracture according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In an embodiment of the present invention, there is provided a method for intelligently analyzing machine vision images of fatigue fractures, as shown in fig. 1, the method includes:
step 102: obtaining a metal fracture image and a metal source image;
step 104: analyzing the metal fracture image based on a deep learning convolutional neural network to determine the type of metal fracture (fatigue fracture or tensile fracture);
step 106: when the metal fracture type is a fatigue fracture type, distinguishing the metal fracture image to obtain a crack expansion area and a crack instantaneous fracture area;
step 108: carrying out statistical analysis on the crack expansion area and the crack instantaneous fracture area;
step 110: and simulating the forming process of the metal fracture based on an image deformation algorithm according to the metal fracture image and the metal source image, and simulating to generate a metal fracture forming process image.
In the embodiment of the invention, the fatigue fracture is obtained firstly, and then the sectional surface of the fatigue fracture is shot by adopting the body type microscope to obtain a complete and clear image, so that the definition of the fracture picture is greatly improved, and the success rate of fatigue fracture analysis is further improved. The metal source image can also be shot by a body type microscope.
In the embodiment of the present invention, as shown in fig. 2, step 104 specifically includes:
step 1041: denoising and graying the images including the metal fatigue fracture and the tensile fracture to obtain grayed metal fatigue fracture and tensile fracture images;
step 1042: training the gray metal fatigue fracture and tensile fracture images based on a deep learning convolutional neural network to obtain a trained deep network detection model;
step 1043: denoising and graying the metal fracture image to obtain a grayed metal fracture image;
step 1044: and detecting the gray-scale metal fracture image by using the trained deep network detection model, and determining the metal fracture type (fatigue fracture or tensile fracture).
Wherein the images including metal fatigue fracture and tensile fracture exist in a training set, the training set comprises a plurality of images including metal fatigue fracture and tensile fracture, and the images are used for training the deep learning convolutional neural network
The deep learning Convolutional Neural Network (CNN) has at least:
1)2 convolutional layers, 1 full-link layer and 1 output layer;
2) the activation function in the convolution layer adopts a linear rectification function;
3) the full connection layer adopts a linear rectification activation function (ReLu) f (x) max (0, x) and a logistic regression function (Sigmoid)
Figure BDA0002516630040000051
In the embodiment of the present invention, as shown in fig. 3, step 106 specifically includes:
crack propagation zone:
step 1061: denoising and graying the metal fracture image to obtain a grayed metal fracture image;
step 1062: detecting the Contour of the grayed metal fracture image by using a Gaussian Active Contour Detection method (Gaussian Active Contour Detection) to obtain a metal image;
step 1063: detecting the metal image by using an Adaptive Binary Threshold detection method (Adaptive Binary Threshold) to obtain an initial image of a metal image expansion area;
step 1064: carrying out corrosion expansion processing on the initial image of the metal image expansion area by using an openCV image corrosion algorithm and an expansion algorithm (Erosis and scaling) to obtain a metal image expansion area outline template;
step 1065: determining a metal crack expansion area image according to the metal fracture image and the metal image expansion area outline template;
crack snap zone:
step 1066: based on the metal crack expansion area image, obtaining an initial instantaneous fracture area contour map according to a preset threshold starting point value and a preset threshold end point value;
step 1067: carrying out corrosion expansion processing on the initial transient interruption region outline by using an openCV image corrosion algorithm and an expansion algorithm (Erosis and scaling) to obtain a transient interruption region outline template;
step 1068: and determining the metal crack instantaneous fracture area image according to the metal fracture image and the instantaneous fracture area outline template.
In the embodiment of the present invention, as shown in fig. 4, step 110 specifically includes:
performing bending deformation processing on the metal source image to obtain a source image with the same shape as the metal fracture image;
performing bending deformation processing on the metal fracture image to obtain a target image with the same shape as the metal source image;
and averaging the corresponding colors of the source image and the target image to obtain a metal fracture forming process image.
The line-pair-based image deformation algorithm (Morphing) is described below.
The image deformation is the effect of progressing from one image to another image, the method is to calculate the intermediate images of all frames in the image change time, and finally, the series of intermediate images are played continuously to achieve the deformation effect, and the following steps are the generation method of the intermediate images of a certain frame:
(1) according to the method using the feature points, two pictures (source image I as shown in FIG. 5) are takenSAnd a target image ID) And performing multi-linear interpolation on the corresponding features to obtain feature information of the intermediate image. Specifically, the multiline line-pair (Multi line-pair) is transformed as follows:
let X ∈ target image IDFor each pair i of straight lines, there is a point X' corresponding to it, as shown in fig. 5. Let Di＝X′i-X, evidently DiIs the amount of shift in pixel position caused by the i-th to linear transformation. And carrying out weighted average on all the offsets to obtain:
Figure BDA0002516630040000061
wherein n is the total number of the straight line pairs; x 'may be taken as X' ═ X + D.
With only one pair of straight lines, then: x ═ X + D1＝X+(X′1-X)＝X′1；
The weight factor is taken as:
Figure BDA0002516630040000062
wherein, lengthiRepresenting a line segment PiQiLength of (d); distiRepresents a straight line PiQiThe distance of (d); a. b and p are parameters for adjusting the relative influence effect of the line pair.
(2) Both images are warped to the intermediate image.
Warp distortion of a single image is represented by warping. To realize two-dimensional images ISAnd IDThe morphing process of (2) first establishes the corresponding relationship between the image features through simple geometric elements.
The geometric elements can be grid nodes, line segments, curves, points and the like, and then the geometric transformation required by morphing is calculated according to the corresponding relation of the characteristics, wherein the geometric transformation defines two images ISAnd IDGeometric correspondence between upper points.
Full shot C0: IS◇IDThe first image ISIs mapped to a second image IDThe geometry of (1), full shot C1: ID◇ISThe second image IDIs mapped to a first map ISThe geometry of the image.
Let P0Being points on the source image, P1As points on the target image, a warping function (warping) W of the source and target images0And W1Are respectively defined as:
Figure BDA0002516630040000071
the two curved and converted images are cross-blended, i.e. the corresponding colors of the two images are averaged, so as to obtain the intermediate image in the frame.
The method of the invention is described below by means of practical examples and images.
1) Obtaining the sample artwork (under an industrial magnifying glass) shown in fig. 6;
2) analyzing the sample original image shown in fig. 6 based on the deep learning convolutional neural network, and determining the type of metal fracture (fatigue fracture or tensile fracture);
3) determining the metal fatigue fracture, and performing fracture image contour detection and background segmentation on the metal fatigue fracture image by using a Gaussian active contour detection method, as shown in FIG. 7;
4) the crack extension zone was obtained using the original method (standard sample, as shown in fig. 8) and using the method of the invention (as shown in fig. 9):
the algorithm accuracy of the invention is calculated according to the area (sum of pixel points) as shown in table 1:
TABLE 1
Figure BDA0002516630040000072
The accuracy of the algorithm is (403422+256140)/(403422+256140+103841+ 38299). times.100% -82.3%.
5) The crack snap zone was obtained using the original method (standard sample, as shown in fig. 10) and using the method of the invention (as shown in fig. 11):
the algorithm accuracy of the invention is calculated according to the area (sum of pixel points) as shown in table 2:
TABLE 2
Figure BDA0002516630040000081
The accuracy of the algorithm is (241787+477199)/(241787+30064+52652+ 477199). times.100% -89.7%.
6) The formation process for simulating the generation of fractured metal based on the image deformation algorithm (Morphing) is shown in fig. 12.
Based on the same inventive concept, the embodiment of the invention also provides an intelligent analysis device for machine vision images of fatigue fracture, which is described in the following embodiment. The principle of solving the problems of the machine vision image intelligent analysis device for the fatigue fracture is similar to that of the machine vision image intelligent analysis method for the fatigue fracture, so the implementation of the machine vision image intelligent analysis device for the fatigue fracture can refer to the implementation of the machine vision image intelligent analysis method for the fatigue fracture, and repeated parts are not repeated. As used hereinafter, the term "unit" or "module" may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.
Fig. 13 is a structural block diagram of an intelligent analysis apparatus for machine vision images of fatigue fracture according to an embodiment of the present invention, as shown in fig. 13, including:
the image obtaining module 02 is used for obtaining a metal fracture image and a metal source image;
the metal fracture type determining module 04 is configured to analyze the metal fracture image based on a deep learning convolutional neural network to determine a metal fracture type;
a crack expansion area and crack snap-off area determining module 06, configured to distinguish the metal fracture image to obtain a crack expansion area and a crack snap-off area when the metal fracture type is a fatigue fracture type;
the statistical analysis module 08 is used for performing statistical analysis on the crack expansion area and the crack instantaneous fracture area;
the metal fracture formation process simulation module 10 is configured to simulate a metal fracture formation process based on an image deformation algorithm according to a metal fracture image and a metal source image, and simulate to generate a metal fracture formation process image. In the embodiment of the present invention, the metal fracture type determining module 04 is specifically configured to:
denoising and graying the images including the metal fatigue fracture and the tensile fracture to obtain grayed metal fatigue fracture and tensile fracture images;
training the gray metal fatigue fracture and tensile fracture images based on a deep learning convolutional neural network to obtain a trained deep network detection model;
denoising and graying the metal fracture image to obtain a grayed metal fracture image;
and detecting the gray metal fracture image by using the trained deep network detection model to determine the metal fracture type.
In the embodiment of the present invention, the crack propagation region and crack transient region determining module 06 is specifically configured to:
denoising and graying the metal fracture image to obtain a grayed metal fracture image;
detecting the contour of the grayed metal fracture image by using a Gaussian active contour detection method to obtain a metal image;
detecting the metal image by using a self-adaptive binary threshold detection method to obtain an initial image of a metal image expansion area;
carrying out corrosion expansion processing on the initial image of the metal image expansion area by using an openCV image corrosion expansion algorithm to obtain a metal image expansion area outline template;
determining a metal crack expansion area image according to the metal fracture image and the metal image expansion area outline template;
based on the metal crack expansion area image, obtaining an initial instantaneous fracture area contour map according to a preset threshold starting point value and a preset threshold end point value;
carrying out corrosion expansion treatment on the initial instantaneous fault area profile graph by using an openCV image corrosion expansion algorithm to obtain an instantaneous fault area profile template;
and determining the metal crack instantaneous fracture area image according to the metal fracture image and the instantaneous fracture area outline template.
In the embodiment of the present invention, the image obtaining module 02 is further configured to: obtaining a metal source image;
in the embodiment of the present invention, the metal fracture formation process simulation module 10 is specifically configured to:
performing bending deformation processing on the metal source image to obtain a source image with the same shape as the metal fracture image;
performing bending deformation processing on the metal fracture image to obtain a target image with the same shape as the metal source image;
and averaging the corresponding colors of the source image and the target image to obtain a metal fracture forming process image.
Based on the same inventive concept, the embodiment of the invention also provides a machine vision image intelligent analysis system of the fatigue fracture, which is described in the following embodiment.
This machine vision image intelligent analysis system of fatigue fracture includes: the machine vision image intelligent analysis device of the fatigue fracture is connected with the body type microscope;
the body microscope is used for: shooting a metal fracture image and a metal source image, and sending the metal fracture image to a machine vision image intelligent analysis device of the fatigue fracture.
The embodiment of the present invention further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the method when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium, and the computer readable storage medium stores a computer program for executing the method.
In conclusion, the intelligent analysis method, device and system for machine vision images of fatigue fracture, provided by the invention, have the following great commercial advantages and values for analyzing the fatigue fracture of metal materials:
1. the risk of accuracy reduction caused by long-term work of personnel can be avoided, the analysis system can work uninterruptedly all the year around, and the cost of analysis working time is greatly reduced on the premise of avoiding wrong analysis results;
2. the analysis system only needs initial cost investment, and does not need to pay analysis work according to hours like experts, so that the cost investment of metal fatigue fracture failure analysis is greatly reduced;
3. the analysis system has extremely low popularization cost, and can obtain a high-efficiency and reliable fatigue fracture analysis result without spending expensive failure expert culture cost and equipment maintenance cost.
4. The invention further improves the intelligent fracture analysis efficiency and reduces the investment cost of equipment and a system, and the integrated body type microscope and the analysis system achieve the aim of simplifying the process design and the equipment investment cost through integrated design, thereby having market development potential.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes may be made to the embodiment of the present invention by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (10)

1. A machine vision image intelligent analysis method for fatigue fracture is characterized by comprising the following steps:
obtaining a metal fracture image and a metal source image;
analyzing the metal fracture image based on a deep learning convolution neural network to determine the type of the metal fracture;
when the metal fracture type is a fatigue fracture type, distinguishing the metal fracture image to obtain a crack expansion area and a crack instantaneous fracture area;
carrying out statistical analysis on the crack expansion area and the crack instantaneous fracture area;
and simulating the forming process of the metal fracture based on an image deformation algorithm according to the metal fracture image and the metal source image, and simulating to generate a metal fracture forming process image.
2. The machine vision image intelligent analysis method for the fatigue fracture as claimed in claim 1, wherein the metal fracture image is analyzed based on a deep learning convolutional neural network to determine the type of the metal fracture, and the method comprises the following steps:
denoising and graying the images including the metal fatigue fracture and the tensile fracture to obtain grayed metal fatigue fracture and tensile fracture images;
training the gray metal fatigue fracture and tensile fracture images based on a deep learning convolutional neural network to obtain a trained deep network detection model;
denoising and graying the metal fracture image to obtain a grayed metal fracture image;
and detecting the gray metal fracture image by using the trained deep network detection model to determine the metal fracture type.
3. The method for intelligently analyzing machine vision images of fatigue fractures as claimed in claim 1, wherein the step of distinguishing the metal fracture images to obtain crack propagation zones comprises:
denoising and graying the metal fracture image to obtain a grayed metal fracture image;
detecting the contour of the grayed metal fracture image by using a Gaussian active contour detection method to obtain a metal image;
detecting the metal image by using a self-adaptive binary threshold detection method to obtain an initial image of a metal image expansion area;
carrying out corrosion expansion processing on the initial image of the metal image expansion area by using an openCV image corrosion expansion algorithm to obtain a metal image expansion area outline template;
and determining the metal crack expansion area image according to the metal fracture image and the metal image expansion area outline template.
4. The machine vision image intelligent analysis method for the fatigue fracture as claimed in claim 3, wherein the step of distinguishing the metal fracture image to obtain a crack transient fracture zone comprises:
based on the metal crack expansion area image, obtaining an initial instantaneous fracture area contour map according to a preset threshold starting point value and a preset threshold end point value;
carrying out corrosion expansion treatment on the initial instantaneous fault area profile graph by using an openCV image corrosion expansion algorithm to obtain an instantaneous fault area profile template;
and determining the metal crack instantaneous fracture area image according to the metal fracture image and the instantaneous fracture area outline template.
5. The machine vision image intelligent analysis method for the fatigue fracture as claimed in claim 1, wherein the metal fracture forming process is generated based on an image deformation algorithm simulation according to the metal fracture image and the metal source image, and the metal fracture forming process image is generated by simulation, and the method comprises the following steps:
performing bending deformation processing on the metal source image to obtain a source image with the same shape as the metal fracture image;
performing bending deformation processing on the metal fracture image to obtain a target image with the same shape as the metal source image;
and averaging the corresponding colors of the source image and the target image to obtain a metal fracture forming process image.
6. The utility model provides a machine vision image intelligent analysis device of fatigue fracture which characterized in that includes:
the image obtaining module is used for obtaining a metal fracture image and a metal source image;
the metal fracture type determining module is used for analyzing the metal fracture image based on a deep learning convolution neural network and determining the metal fracture type;
the crack expansion area and crack snap-off area determining module is used for distinguishing the metal fracture image when the metal fracture type is a fatigue fracture type to obtain a crack expansion area and a crack snap-off area;
the statistical analysis module is used for performing statistical analysis on the crack expansion area and the crack instantaneous fracture area;
and the metal fracture formation process simulation module is used for simulating the formation process of the metal fracture based on an image deformation algorithm according to the metal fracture image and the metal source image, and simulating to generate a metal fracture formation process image.
7. The machine-vision graphical intelligent analysis device of fatigue fractures of claim 6, wherein the metal fracture formation process simulation module is configured to:
performing bending deformation processing on the metal source image to obtain a source image with the same shape as the metal fracture image;
performing bending deformation processing on the metal fracture image to obtain a target image with the same shape as the metal source image;
and averaging the corresponding colors of the source image and the target image to obtain a metal fracture forming process image.
8. A machine vision image intelligent analysis system of fatigue fracture is characterized by comprising: machine vision image intelligent analysis device of fatigue fracture of style microscope, any one of claims 6 to 7;
the body microscope is used for: shooting a metal fracture image and a metal source image, and sending the metal fracture image to a machine vision image intelligent analysis device of the fatigue fracture.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of claims 1 to 5 when executing the computer program.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program for executing the method of any one of claims 1 to 5.
CN202010478691.1A 2020-05-29 2020-05-29 Machine vision image intelligent analysis method, device and system for fatigue fracture Active CN111681217B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010478691.1A CN111681217B (en) 2020-05-29 2020-05-29 Machine vision image intelligent analysis method, device and system for fatigue fracture

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010478691.1A CN111681217B (en) 2020-05-29 2020-05-29 Machine vision image intelligent analysis method, device and system for fatigue fracture

Publications (2)

Publication Number Publication Date
CN111681217A true CN111681217A (en) 2020-09-18
CN111681217B CN111681217B (en) 2023-06-20

Family

ID=72453278

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010478691.1A Active CN111681217B (en) 2020-05-29 2020-05-29 Machine vision image intelligent analysis method, device and system for fatigue fracture

Country Status (1)

Country Link
CN (1) CN111681217B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114037705A (en) * 2022-01-11 2022-02-11 南通皋亚钢结构有限公司 Metal fracture fatigue source detection method and system based on moire lines

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102692188A (en) * 2012-05-08 2012-09-26 浙江工业大学 Dynamic crack length measurement method for machine vision fatigue crack propagation test
US20130129187A1 (en) * 2010-08-09 2013-05-23 Bt Imaging Pty Ltd Persistent feature detection
CN104657548A (en) * 2015-02-02 2015-05-27 西安电子科技大学 Modeling method for radiation array plane error of planar slot array antenna
CN110009606A (en) * 2019-03-22 2019-07-12 北京航空航天大学 A kind of crack propagation dynamic monitoring method and device based on image recognition

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130129187A1 (en) * 2010-08-09 2013-05-23 Bt Imaging Pty Ltd Persistent feature detection
CN102692188A (en) * 2012-05-08 2012-09-26 浙江工业大学 Dynamic crack length measurement method for machine vision fatigue crack propagation test
CN104657548A (en) * 2015-02-02 2015-05-27 西安电子科技大学 Modeling method for radiation array plane error of planar slot array antenna
CN110009606A (en) * 2019-03-22 2019-07-12 北京航空航天大学 A kind of crack propagation dynamic monitoring method and device based on image recognition

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
JINGWEI DONG 等: "The Review of New NDT Methods of Metal Material Fatigue Monitoring" *
周志宇: "基于Grouplet变换的金属断口图像处理方法研究" *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114037705A (en) * 2022-01-11 2022-02-11 南通皋亚钢结构有限公司 Metal fracture fatigue source detection method and system based on moire lines

Also Published As

Publication number Publication date
CN111681217B (en) 2023-06-20

Similar Documents

Publication Publication Date Title
CN106960202B (en) Smiling face identification method based on visible light and infrared image fusion
KR100903348B1 (en) Emotion recognition mothod and system based on feature fusion
Youssif et al. Automatic facial expression recognition system based on geometric and appearance features
Sarkar et al. Deep learning for structural health monitoring: A damage characterization application
JP5505409B2 (en) Feature point generation system, feature point generation method, and feature point generation program
CN106845384B (en) gesture recognition method based on recursive model
CN110889332A (en) Lie detection method based on micro expression in interview
CN111291701B (en) Sight tracking method based on image gradient and ellipse fitting algorithm
CN104268515A (en) Sperm morphology anomaly detection method
CN112270658A (en) Elevator steel wire rope detection method based on machine vision
CN111798404B (en) Iris image quality evaluation method and system based on deep neural network
CN112949560B (en) Method for identifying continuous expression change of long video expression interval under two-channel feature fusion
CN111667425B (en) Facial expression image shielding and repairing method based on priori algorithm
CN111681217A (en) Machine vision image intelligent analysis method, device and system for fatigue fracture
CN113822157A (en) Mask wearing face recognition method based on multi-branch network and image restoration
Konovalenko et al. Fuzzy logic analysis of parameters of dimples of ductile tearing on the digital image of fracture surface
CN106919898A (en) Feature modeling method in recognition of face
CN114170686A (en) Elbow bending behavior detection method based on human body key points
CN111209886B (en) Rapid pedestrian re-identification method based on deep neural network
CN112686872A (en) Wood counting method based on deep learning
Rehman An ensemble of neural networks for nonlinear segmentation of overlapped cursive script
Zhang Snake Model Based on Improved Genetic Algorithm in Fingerprint Image Segmentation.
Ren et al. A linear hybrid classifier for fingerprint segmentation
Kshirsagar Segmentation Of One And Two Hand Gesture Recognition Using Key Frame Selection
CN112418085A (en) Facial expression recognition method under partial shielding working condition

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