CN114037705B - Metal fracture fatigue source detection method and system based on moire lines - Google Patents

Metal fracture fatigue source detection method and system based on moire lines Download PDF

Info

Publication number
CN114037705B
CN114037705B CN202210024107.4A CN202210024107A CN114037705B CN 114037705 B CN114037705 B CN 114037705B CN 202210024107 A CN202210024107 A CN 202210024107A CN 114037705 B CN114037705 B CN 114037705B
Authority
CN
China
Prior art keywords
fatigue
area
region
fracture
pixel
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202210024107.4A
Other languages
Chinese (zh)
Other versions
CN114037705A (en
Inventor
保柳柳
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.)
Nantong Gaoya Steel Structure Co ltd
Original Assignee
Nantong Gaoya Steel Structure Co ltd
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 Nantong Gaoya Steel Structure Co ltd filed Critical Nantong Gaoya Steel Structure Co ltd
Priority to CN202210024107.4A priority Critical patent/CN114037705B/en
Publication of CN114037705A publication Critical patent/CN114037705A/en
Application granted granted Critical
Publication of CN114037705B publication Critical patent/CN114037705B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/0002Inspection of images, e.g. flaw detection
    • G06T7/0004Industrial image inspection
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • G06N3/045Combinations of networks
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/08Learning methods
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/73Deblurring; Sharpening
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/10Segmentation; Edge detection
    • G06T7/13Edge detection
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/70Determining position or orientation of objects or cameras
    • G06T7/73Determining position or orientation of objects or cameras using feature-based methods
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20081Training; Learning
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/20Special algorithmic details
    • G06T2207/20084Artificial neural networks [ANN]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30108Industrial image inspection
    • G06T2207/30136Metal
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/30Subject of image; Context of image processing
    • G06T2207/30241Trajectory

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Quality & Reliability (AREA)
  • Analysing Materials By The Use Of Radiation (AREA)

Abstract

The invention relates to the field of image processing, in particular to a method and a system for detecting the position of a metal fatigue source according to an image. The method comprises the following steps: extracting a fracture area and a central point of the part; processing the extracted fracture area image of the part, and establishing a central coordinate system by taking the central point as an origin; calculating the change condition of the pixel values in all directions of the central point of the area to obtain a pulse graph of the pixel value change, calculating the mean square difference value of a peripheral density sequence according to a pulse curve oscillogram, dividing the area type of the fracture area of the part, determining the boundary point of a fatigue expansion area and a transient fracture area, fitting the boundary point to obtain a scallop line, determining the moving track of the scallop line, and obtaining the final intersection position of the scallop line and the edge of the part as the position of the fatigue source. According to the method, the shell line in the fracture is searched for according to the gray value fluctuation condition in the fracture area, the shell line can be found more accurately, and the condition that the fatigue source detection effect is poor due to the fact that the shell line in the fracture is not clear can be effectively avoided.

Description

Metal fracture fatigue source detection method and system based on moire lines
Technical Field
The invention relates to the field of image processing, in particular to a method and a system for detecting a metal fracture fatigue source based on a moire line.
Background
The fracture of the part records the whole process from the initiation and the propagation of the crack to the fracture, and through morphological analysis of the fracture, some basic problems of the part fracture can be obtained, such as fracture cause, fracture mode, crack propagation trend and the like, and the finding of the fatigue source position of the fracture area is the basis of the whole analysis. Most of existing methods for searching for the position of the fatigue source are manual detection, and detection is carried out by experienced workers, or a texture detection method is used for analyzing a shell line in a fracture, so that the position of the fatigue source is searched for according to the method, but the method is limited in detection of the condition that the shell line in the fracture is clear, and poor in detection effect of the position of the fatigue source under the condition that the shell line is not clear.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides the metal fracture fatigue source detection method and system based on the shell lines, which can accurately find the shell lines in the fracture area and accurately position the fatigue source position through the determined shell lines.
In order to achieve the purpose, the invention adopts the following technical scheme, namely a method and a system for detecting a metal fracture fatigue source based on a moire line.
The implementation of the method comprises the following steps:
s1: extracting the central point of the fracture region image of the part, performing binarization image processing on the extracted region image, and establishing a coordinate system by taking the central point of the region image as an origin;
s2: taking the central point of the region image as an origin, extending and generating rays in all directions in a radial manner, obtaining a pixel pulse curve graph for ray pairs in all directions according to pixel values on the generated rays, calculating the peripheral density of each pulse in the pixel pulse curve graph, obtaining a peripheral density sequence according to the peripheral density of each pulse in the pixel pulse curve graph, and obtaining the mean square error of the peripheral density sequence by using the peripheral density sequence;
judging the areas through which the rays pass in each direction according to the mean square error of the peripheral density sequence; the passing area is a fatigue expansion area or a transient interruption area or a common area passing through the fatigue expansion area and the transient interruption area simultaneously;
dividing pixel points on rays passing through the common region into fatigue expansion regions or instantaneous interruption regions, and determining demarcation points on rays in all directions according to the division result;
fitting the obtained demarcation points on the rays in each direction to obtain a shell line;
s3: extracting two edge rays passing through the common area, and finding out a median line of an included angle formed by the two rays, wherein the obtained median line is the motion track of the Bayes line;
and (4) moving the shell line in parallel along the motion track to the direction of the fatigue expansion area, and determining the position of the fatigue source of the fracture area of the part.
The binarization processing of the area image is to sharpen and enhance the area of the area image and to binarize the sharpened and enhanced image.
The method for obtaining the peripheral intensity sequence of the pulse curve of the pixel values on the directional ray is as follows:
extracting the abscissa of the starting point and the ending point of the waveform in a rectangular pulse curve on rays in each direction in a pixel pulse curve graph, calculating the concentration of each pulse and two adjacent pulses to obtain the peripheral concentration of each pulse, wherein the calculation formula is as follows:
Figure 196854DEST_PATH_IMAGE002
in the formula:
Figure DEST_PATH_IMAGE003
representing the peripheral concentration of the pulse;
Figure 538843DEST_PATH_IMAGE004
is the waveform width of the pulse;
Figure DEST_PATH_IMAGE005
the pixel distance between the waveform and the end point of the adjacent waveform on the left side;
Figure 360213DEST_PATH_IMAGE006
the pixel distance between the waveform and the starting point of the adjacent waveform on the right side;
and according to the calculated peripheral density of all the pulses on the pulse curve, combining the peripheral density of each pulse according to the pulse sequence on the pulse curve to obtain the peripheral density sequence of the pulse curve.
The step of dividing the region types passed by the rays in each direction is as follows:
the mean square error value of the peripheral dense sequence of the pulse curve is calculated by the following formula:
Figure 208084DEST_PATH_IMAGE008
in the formula:
Figure DEST_PATH_IMAGE009
is the mean square error of the surrounding intensity sequences,
Figure 195631DEST_PATH_IMAGE010
for the data in the peripheral intensity sequence,
Figure DEST_PATH_IMAGE011
representing the position of the data in the perimeter intensity sequence,
Figure 810152DEST_PATH_IMAGE012
is the average of the data in the surrounding intensity sequence,
Figure DEST_PATH_IMAGE013
is a pulse curve and
Figure 904272DEST_PATH_IMAGE014
the number of the intersection points of the axes;
setting a mean square error threshold value of a peripheral dense sequence of a pulse curve, and dividing the region type of a part fracture region according to the set threshold value:
Figure 782098DEST_PATH_IMAGE016
in the formula:
Figure DEST_PATH_IMAGE017
a region-indicating mark is shown which indicates,
Figure 522521DEST_PATH_IMAGE018
the degree of an included angle formed by each direction ray and the horizontal axis of the coordinate system is expressed;
Figure DEST_PATH_IMAGE019
representing the situation that the ray made by the central point to each direction passes through the transient interruption region,
Figure 81679DEST_PATH_IMAGE020
representing the case where the ray from the center point in the direction to each direction passes through the fatigue extension region,
Figure DEST_PATH_IMAGE021
which represents the case where rays made by the center point in the respective directions pass through the common area.
Further, a method for determining a demarcation point on a ray in each direction comprises the following steps:
firstly, extracting the peripheral density of all pixel points on the directional ray, respectively comparing with the average value of the peripheral density sequence, and determining the type of the region to which the pixel points belong;
then marking all pixel points on the directional ray, and marking the pixel points belonging to the fatigue expansion area
Figure 154939DEST_PATH_IMAGE022
Marking of pixel points belonging to a transient interruption zone
Figure DEST_PATH_IMAGE023
And then comparing the marking values of the pixel points at two adjacent sides of all the pixel points in the direction, and when the marking values of the pixel points at two sides of the pixel point are different, the pixel point is a demarcation point in the direction.
The method for determining the position of the fatigue source comprises the following steps:
extracting two edge rays passing through the common region, finding out a median line of an included angle formed by the two edge rays, wherein the median line is a motion track of the shell line, horizontally moving the shell line along the motion track to the fatigue expansion region, and when the shell line is moved to be intersected with only one pixel point of the region image, the pixel point is the position of the fatigue source.
The system comprises the following modules: the device comprises an image processing module, a shell line fitting module and a fatigue source determining module;
the image acquisition module is used for acquiring RGB image information of part fracture and extracting a part fracture area.
And the image processing module is used for carrying out sharpening processing and binarization processing on the fracture area image and providing image detail information for determining the position of the fatigue source.
And the scallop line fitting module is used for determining regional boundary points of the instantaneous fracture region and the fatigue expansion region from the detail information, and then fitting all the boundary points to obtain the scallop lines.
And the fatigue source determining module is used for determining the final position of the fatigue source according to the fitted beta line and the region type.
The invention has the beneficial effects that: the shell line can be more accurately found out through the fluctuation condition of the gray value in the fracture area, and the condition that the fatigue source is inaccurately detected due to the unclear shell line in the fracture can be effectively avoided.
1. The method and the device have the advantages that the image of the fracture area of the part is sharpened, enhanced and binarized, the detail information of the image is highlighted, and the fatigue expansion area and the instantaneous fracture area in the fracture area of the part can be determined more clearly.
2. Through the fatigue expansion area and the instantaneous interruption area which are judged, the scallop lines obtained by the least square method fitting of the boundary points are more accurate, so that the movement locus of the scallop lines is more accurate, and the accuracy of the position of the fatigue source is improved.
Drawings
FIG. 1 is a schematic flow diagram of the present invention.
Fig. 2 is a schematic view of a metal fracture in the present embodiment.
Fig. 3 is a schematic diagram of the embodiment after the sharpening of the metal fracture is enhanced.
Fig. 4 is a schematic diagram of the center coordinate system in the present embodiment.
Fig. 5 is a schematic view of rays in each direction in the present embodiment.
Fig. 6 is a graph of the pixel pulse in this embodiment.
Fig. 7 is a schematic view of the peripheral density in the present embodiment.
Fig. 8 is a block diagram of a system in the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and examples.
In the description of the present invention, it is to be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience of description and simplicity of description, and do not indicate or imply that the referenced devices or elements must have a particular orientation, be constructed and operated in a particular orientation, and thus, are not to be construed as limiting the present invention.
The terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature; in the description of the present invention, "a plurality" means two or more unless otherwise specified.
Example 1
The method mainly aims to utilize an image processing technology to search the shell line according to the difference of the roughness of a transient fracture area and the roughness of a fatigue expansion area, and judge the unclear fracture fatigue source position;
the following description is made with reference to the examples and fig. 1:
the method comprises the following steps: extracting a part fracture area from the obtained RGB image by using the enclosure frame; sharpening and enhancing the extracted region, and establishing a central coordinate system;
the DNN can realize target detection automatically, but when the accuracy result is related, the DNN accuracy is limited, and labeling of the lines of the shell and the fatigue source needs to be performed by professional personnel, and the operation is complex. And the section is identified only by using the form of the bounding box, so that the labeling is simple, and the DNN result is reliable.
Only DNN is used to detect the slices and not to identify more other information.
The invention uses DNN neural network to carry out target detection on the fracture in the image by using an Encoder-Decoder-Bbox structure, and the specific process is as follows:
1. the network adopts an Encoder-Decoder form, and firstly encodes and then decodes the image. The input of the network is an image and the output is the center of the bounding box.
2. The network takes the image with the detected target as input, firstly decodes, namely uses convolution and pooling operation to extract spatial domain characteristics in the image in the process of downsampling the image, and the output of the encoder is the extracted characteristic vector.
3. The input of the decoder is the output characteristic vector of the encoder, and the decoder returns the central point of the target corresponding bounding box in the image after up-sampling. The output of the decoder is the output of the network.
4. The data set used for training the network is an image with a part fracture;
5. the label of the image is an enclosure frame corresponding to the fracture of the part, and comprises a central point coordinate of the enclosure frame, wherein the central point is marked in the image in a hot spot mode, namely the mark is corresponding to the mark
Figure 672508DEST_PATH_IMAGE024
After the coordinates are obtained, a corresponding hot spot is obtained by blurring the coordinates using a gaussian kernel.
6, the Loss function is a mean square error Loss function;
at this point, extracting the fracture area of the part shown in fig. 2, and acquiring the central point of the area, which is recorded as
Figure DEST_PATH_IMAGE025
Using the extracted region
Figure 837910DEST_PATH_IMAGE026
The gradient sharpening of the operator is enhanced, the details in the fracture area are highlighted, then the binarization is carried out, and the obtained result is shown in figure 3.
The gray gradient of the projection positions in the instantaneous interruption area is changed greatly, and the positions are converted into white pixel points after the operation; the fatigue expansion area is smooth, the gradient change of the gray scale is small, and most of the area is converted into black background pixel points.
The image is sharpened and enhanced and binarized, so that the detail information of the fracture area image can be obtained more clearly, and the boundary point between the fatigue expansion area and the instantaneous fracture area can be obtained more accurately when the area type in the area image is analyzed.
To be provided with
Figure DEST_PATH_IMAGE027
As the origin, the horizontal direction to the right is the positive X-axis direction, and the vertical direction is the positive Y-axis direction, a central coordinate system is established, as shown in fig. 4.
Step two: calculating the change condition of the pixel values in all directions of the central point of the area, determining the boundary position of the instantaneous interruption area and the expansion area, and fitting the boundary points according to the boundary position to obtain a shell line;
in the embodiment, the boundary point of the transient interruption area and the fatigue expansion area is found through the change condition of the pixel values in each direction of the central point of the section, so that the shell line is fitted. According to the relative position of the instantaneous fracture area and the fatigue expansion area, moving a shell line to the fatigue expansion area, and finally, taking the position intersected with the fracture edge as the position of a fatigue source, wherein the specific process is as follows:
a) obtaining a pixel pulse curve graph of the central point in each direction according to the gray value in each direction;
b) calculating the pulse peripheral density
Figure 233382DEST_PATH_IMAGE003
Searching a region boundary point;
c) and fitting the boundary points according to all the obtained boundary points to obtain a shell line, and determining the relative positions of the fatigue expansion area and the instantaneous fracture area.
The following is an expanded description of the above process:
corresponding to the step a), obtaining a pixel pulse curve graph of the central point in each direction according to the pixel values in each direction;
1. at the center point
Figure 128525DEST_PATH_IMAGE025
Starting at a point around the centreIn the range, every time the included angle formed by the positive direction of the X axis is 1 degree, 2 degrees and … 360 degrees, the central point of the region image is taken as the original point, the rays are generated by extending in all directions in a radial mode, and the rays generated in all directions are shown in FIG. 5;
2. and fitting each pixel point and pixel value thereof obtained by each ray into a pixel pulse curve graph which starts from an original point, takes the central point of the area image as the original point, extends and generates the ray in each direction in a radial mode, and obtains the ray pair in each direction according to the pixel value on the generated ray.
Since the extracted fracture area is binarized in the processing of the pair of data in the step, the pixel values of the pixel points in the area are only black and white, namely, in a certain direction, starting from a central point, the pixel values jump between 0 and 255 along with the increase of the number of pixels, and are in a rectangular pulse curve shape as shown in fig. 6;
the pixel point corresponding to the pixel value of 255 is a white pixel point in the sharpened image, and the pixel point corresponding to the pixel value of 0 is a background pixel point.
Corresponding to the step b), calculating the proportion of background pixels in the midpoint range of two adjacent waveform intervals, judging the type of the region where the pixel points are located, and searching region boundary points;
and (3) logical level: the instantaneous interruption region has more and dense white pixels, so that the width of the waveform in the instantaneous interruption region is larger and the distance between the waveforms is smaller, and the specific expression is that the peripheral concentration of the pulse is larger, namely the ratio of the width of the pulse to the distance between two adjacent pulses is larger; the fatigue expansion area has fewer and sparse white pixels, so the waveform width of the fatigue expansion area is smaller and the distance between the waveforms is larger, and the specific expression is that the peripheral concentration of the pulse is smaller, namely the value of the ratio of the width of the pulse to the distance between two adjacent pulses is smaller; the waveform pitch of a curve on a direction ray passing through the common region (namely a region passing through the fatigue expansion region and the instantaneous interruption region simultaneously) is shown as sparse first and then dense or dense first and then sparse;
the specific steps of the part are as follows:
1. separately recording a) stepsThe abscissa of the start point and the end point of the pulse curve waveform of the pixel values on the rays in the respective directions obtained in the step is obtained as
Figure 285837DEST_PATH_IMAGE028
Of the pixel pulse profile of (1), wherein
Figure DEST_PATH_IMAGE029
Is an integer and
Figure 735273DEST_PATH_IMAGE029
is taken as
Figure 573916DEST_PATH_IMAGE030
(co)
Figure DEST_PATH_IMAGE031
One waveform) as shown in fig. 7.
2. Calculating the peripheral concentration of the pulse
Figure 293873DEST_PATH_IMAGE003
The abscissa of the starting point and the end point of the waveform to be calculated is set as
Figure 887665DEST_PATH_IMAGE032
Then the horizontal coordinates of the starting point and the ending point of the waveform on the left and right sides are respectively
Figure DEST_PATH_IMAGE033
The calculation formula is as follows:
Figure 558818DEST_PATH_IMAGE002
as shown in fig. 7, wherein
Figure 591365DEST_PATH_IMAGE004
Is wave-shaped in width, i.e.
Figure 70888DEST_PATH_IMAGE034
Figure 133784DEST_PATH_IMAGE005
The pixel distance of the waveform from the end of the left adjacent waveform, i.e.
Figure DEST_PATH_IMAGE035
Figure 495495DEST_PATH_IMAGE006
The pixel distance of the start of the waveform from the right-hand adjacent waveform, i.e.
Figure 66154DEST_PATH_IMAGE036
It should be noted that when the waveform is the first waveform, i.e. the first waveform
Figure DEST_PATH_IMAGE037
When the temperature of the water is higher than the set temperature,
Figure 993658DEST_PATH_IMAGE038
(ii) a When the waveform is the last waveform, i.e.
Figure DEST_PATH_IMAGE039
When the temperature of the water is higher than the set temperature,
Figure 899560DEST_PATH_IMAGE040
arranging all the obtained peripheral densities according to the sequence of the pulses in the pulse curve, and integrating to obtain a peripheral density sequence of the pulse curve:
Figure DEST_PATH_IMAGE041
by calculating the peripheral density of each pulse according to the pulse curve on each direction ray, the change condition of the gray level of the pixel points on the direction ray can be reflected more clearly, and the demarcation point on the direction ray can be determined more accurately.
3. The mean square error of the peripheral density sequence is obtained by calculation
Figure 545305DEST_PATH_IMAGE009
The consistency of the pulse peripheral density in the direction is judged, so that the area of the fracture area of the part is divided;
the mean square error of the surrounding intensity sequence is calculated by the formula:
Figure 388496DEST_PATH_IMAGE042
in the formula:
Figure 108190DEST_PATH_IMAGE009
is the mean square error of the surrounding intensity sequences,
Figure 214687DEST_PATH_IMAGE010
for the data in the peripheral intensity sequence,
Figure 380351DEST_PATH_IMAGE011
representing the position of the data in the perimeter intensity sequence,
Figure 902599DEST_PATH_IMAGE012
is the average of the data in the surrounding intensity sequence,
Figure 335854DEST_PATH_IMAGE013
is a pulse curve and
Figure 550935DEST_PATH_IMAGE014
the number of intersections of the axes.
According to the priori knowledge, when the pixel points in a certain direction are all located in a transient interruption area or a fatigue expansion area, the pixel points are
Figure 436852DEST_PATH_IMAGE003
The degree of change of the mean square error is small, so that the mean square error is small; when the pixel point in the direction passes through the transient interruption region and the fatigue expansion region, the pixel point in the direction passes through the transient interruption region and the fatigue expansion region
Figure 762791DEST_PATH_IMAGE003
There are large variations in the whole areaTherefore, the obtained mean square error is large, and the region types can be distinguished.
Setting a mean square error threshold value of the peripheral density sequence, wherein the empirical value is 0.3; setting region marks in various directions
Figure DEST_PATH_IMAGE043
The judgment criteria are as follows:
Figure DEST_PATH_IMAGE045
in the formula:
Figure 829315DEST_PATH_IMAGE017
a region-indicating mark is shown which indicates,
Figure 74352DEST_PATH_IMAGE018
representing the angle of each direction.
Figure 588510DEST_PATH_IMAGE019
Representing the situation that the ray made by the central point to each direction passes through the transient interruption region,
Figure 577194DEST_PATH_IMAGE020
representing the case where the ray from the center point in the direction to each direction passes through the fatigue extension region,
Figure 860408DEST_PATH_IMAGE021
the case where the rays made from the center point in the direction are common to the respective directions is shown.
4. Determining demarcation points on the rays in all directions:
to is directed at
Figure 541925DEST_PATH_IMAGE021
In case of (1), for
Figure 277800DEST_PATH_IMAGE003
The peripheral intensity of all pulses in the sequence is respectively equal to
Figure 571640DEST_PATH_IMAGE003
The average values of the data in the sequence are compared and
Figure 771678DEST_PATH_IMAGE046
the pixel points of (a) consider the waveform to be located in the fatigue extension region, and mark the starting point and the end point of the waveform
Figure 499462DEST_PATH_IMAGE022
(ii) a Will be provided with
Figure DEST_PATH_IMAGE047
The pixel points consider the waveform to be positioned in the transient interruption area, and marks of the starting point and the end point of the waveform
Figure 316109DEST_PATH_IMAGE023
(ii) a After all the waveforms are judged, traversing the pulse sequence and comparing the pixel points
Figure 646596DEST_PATH_IMAGE048
Marking of two-sided pixel points
Figure DEST_PATH_IMAGE049
I.e. by
Figure 733763DEST_PATH_IMAGE050
And
Figure DEST_PATH_IMAGE051
when the two are different (the exclusive or value is 1), the dividing point is the first point from the central point in the direction
Figure 491503DEST_PATH_IMAGE052
Each pixel point records the coordinates of the demarcation point; and repeating the step, traversing all direction rays in the regional image, and recording to obtain demarcation points on all direction rays.
5. Corresponding to the step c), fitting the boundary points according to the recorded boundary points on the rays in each direction to obtain a shell line;
and performing curve fitting on all the demarcation points obtained in the last step by using least square fitting, so that as many demarcation points as possible fall on the curve.
Step three: determining the motion trail of the shell lines according to the obtained shell lines, and further determining the position of the fatigue source;
and according to the obtained shell line, extracting two edge rays which simultaneously pass through a common area of the fatigue expansion area and the instantaneous interruption area to obtain a median line of an included angle formed by the two rays, wherein the median line is the motion track of the shell line.
The fatigue expansion zone is formed firstly when the part is broken from the fatigue source, and the scallop line has the fatigue characteristic and only exists in the fatigue expansion zone, so that the position of the fatigue source can be obtained by moving the obtained scallop line to the fatigue expansion zone, and therefore, the moving direction of the scallop line is always moved to the fatigue expansion zone.
And (4) moving the shell line in parallel to the fatigue expansion region along the motion track, wherein the position of the shell line which is finally intersected with the edge of the part (when only one intersected pixel point is provided) is the position of the fatigue source.
As shown in FIG. 8, the invention provides a fatigue source detection system for metal fracture based on a moire line, which comprises the following modules: the device comprises an image acquisition module, an image processing module, a shell line fitting module and a fatigue source determining module.
The image acquisition module is used for acquiring RGB image information of part fracture and extracting a part fracture area.
And the image processing module is used for carrying out sharpening processing and binarization processing on the fracture area image and providing image detail information for determining the position of the fatigue source.
And the scallop line fitting module is used for extracting pixel point information from the detail information and determining that the pixel points on the ray in each direction belong to a fatigue expansion area or a transient interruption area, so that the boundary points of the transient interruption area and the fatigue expansion area on the ray in the direction are determined, and all the obtained boundary points are fitted to obtain the scallop line.
And the fatigue source determining module is used for acquiring the moving direction of the shell lines according to the fitted shell lines and the region types, horizontally moving the shell lines to the moving direction of the shell lines and determining the final position of the fatigue source.
The above examples are merely illustrative of the present invention and do not constitute the scope of the present invention.

Claims (7)

1. A method for detecting a metal fracture fatigue source based on a moire line is characterized by comprising the following steps:
s1: extracting the central point of the fracture region image of the part, performing binarization image processing on the extracted region image, and establishing a coordinate system by taking the central point of the region image as an origin;
s2: taking the central point of the region image as an origin, extending and generating rays in all directions in a radial manner, obtaining a pixel pulse curve graph for ray pairs in all directions according to pixel values on the generated rays, calculating the peripheral density of each pulse in the pixel pulse curve graph, obtaining a peripheral density sequence according to the peripheral density of each pulse in the pixel pulse curve graph, and obtaining the mean square error of the peripheral density sequence by using the peripheral density sequence;
judging the areas through which the rays pass in each direction according to the mean square error of the peripheral density sequence; the passing area is a fatigue expansion area or a transient interruption area or a common area passing through the fatigue expansion area and the transient interruption area simultaneously;
dividing pixel points on rays passing through the common region into fatigue expansion regions or instantaneous interruption regions, and determining demarcation points on rays in all directions according to the division result;
fitting the obtained demarcation points on the rays in each direction to obtain a shell line;
s3: extracting two edge rays passing through the common area, and finding out a median line of an included angle formed by the two rays, wherein the obtained median line is the motion track of the Bayes line;
and (4) moving the shell line in parallel along the motion track to the fatigue expansion area to determine the fatigue source position of the fracture area of the part.
2. The method for detecting the fatigue source of the metal fracture based on the moire lines as claimed in claim 1, wherein the method for performing binarization processing on the area image comprises the following steps: firstly, sharpening enhancement is carried out on the region image region, and then binarization processing is carried out on the sharpened and enhanced image.
3. The method for detecting the metal fracture fatigue source based on the moire lines as claimed in claim 1, wherein the method for obtaining the peripheral intensity sequence of the pulse curve is as follows:
extracting the abscissa of the starting point and the ending point of the waveform in a rectangular pulse curve on rays in each direction in a pixel pulse curve graph, calculating the concentration of each pulse and two adjacent pulses to obtain the peripheral concentration of each pulse, wherein the calculation formula is as follows:
Figure DEST_PATH_IMAGE002
in the formula:
Figure DEST_PATH_IMAGE004
representing the peripheral concentration of the pulse;
Figure DEST_PATH_IMAGE006
is the waveform width of the pulse;
Figure DEST_PATH_IMAGE008
the pixel distance between the waveform and the end point of the adjacent waveform on the left side;
Figure DEST_PATH_IMAGE010
the pixel distance between the waveform and the starting point of the adjacent waveform on the right side;
and according to the calculated peripheral density of all the pulses on the pulse curve, combining the peripheral density of each pulse according to the pulse sequence on the pulse curve to obtain the peripheral density sequence of the pulse curve.
4. The method for detecting the metal fracture fatigue source based on the moire lines as claimed in claim 1, wherein the step of dividing the region types passed by the rays in each direction is as follows:
the mean square error value of the peripheral dense sequence of the pulse curve is calculated by the following formula:
Figure DEST_PATH_IMAGE012
in the formula:
Figure DEST_PATH_IMAGE014
is the mean square error of the surrounding intensity sequences,
Figure DEST_PATH_IMAGE016
for the data in the peripheral intensity sequence,
Figure DEST_PATH_IMAGE018
representing the position of the data in the perimeter intensity sequence,
Figure DEST_PATH_IMAGE020
is the average of the data in the surrounding intensity sequence,
Figure DEST_PATH_IMAGE022
is a pulse curve and
Figure DEST_PATH_IMAGE024
the number of the intersection points of the axes;
setting a mean square error threshold value of a peripheral dense sequence of a pulse curve, and dividing the region type of a part fracture region according to the set threshold value:
Figure DEST_PATH_IMAGE026
in the formula:
Figure DEST_PATH_IMAGE028
a region-indicating mark is shown which indicates,
Figure DEST_PATH_IMAGE030
the degree of an included angle formed by each direction ray and the horizontal axis of the coordinate system is expressed;
Figure DEST_PATH_IMAGE032
representing the situation that the ray made by the central point to each direction passes through the transient interruption region,
Figure DEST_PATH_IMAGE034
representing the case where the ray from the center point in the direction to each direction passes through the fatigue extension region,
Figure DEST_PATH_IMAGE036
which represents the case where rays made by the center point in the respective directions pass through the common area.
5. The method for detecting the metal fracture fatigue source based on the moire lines as claimed in claim 1, is characterized in that the method for determining the demarcation point on the ray in each direction comprises the following steps:
extracting the peripheral density of all pixel points on the directional ray, and respectively comparing the peripheral density with the average value of the peripheral density sequence to determine the type of the region to which the pixel points belong;
marking all pixel points on the directional ray, and marking the pixel points belonging to the fatigue expansion area
Figure DEST_PATH_IMAGE038
Marking of pixel points belonging to a transient interruption zone
Figure DEST_PATH_IMAGE040
And comparing the marking values of the pixel points at the two adjacent sides of all the pixel points in the direction, and when the marking values of the pixel points at the two sides of the pixel point are different, the pixel point is a demarcation point in the direction.
6. The method for detecting the fatigue source of the metal fracture based on the shell line as claimed in claim 1, wherein the method for determining the position of the fatigue source comprises the following steps:
and horizontally moving a motion track formed by the shell line along a median line of an included angle formed by the two edge rays to the direction of the fatigue expansion area, and when the motion track is moved to the state that only one pixel point of the shell line and the area image is intersected, determining the pixel point as the position of the fatigue source.
7. A metal fracture fatigue source detection system based on a moire line is characterized by comprising the following modules: the device comprises an image processing module, a shell line fitting module and a fatigue source determining module;
the image acquisition module is used for acquiring RGB image information of part fracture and extracting a part fracture area;
the image processing module is used for carrying out sharpening enhancement processing and binarization processing on the fracture area image and providing image detail information for determining the position of the fatigue source;
the shell line fitting module is used for determining a region boundary point of a transient interruption region and a fatigue expansion region from detail information;
and the fatigue source determining module is used for determining the final position of the fatigue source according to the fitted beta line and the region type.
CN202210024107.4A 2022-01-11 2022-01-11 Metal fracture fatigue source detection method and system based on moire lines Active CN114037705B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210024107.4A CN114037705B (en) 2022-01-11 2022-01-11 Metal fracture fatigue source detection method and system based on moire lines

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210024107.4A CN114037705B (en) 2022-01-11 2022-01-11 Metal fracture fatigue source detection method and system based on moire lines

Publications (2)

Publication Number Publication Date
CN114037705A CN114037705A (en) 2022-02-11
CN114037705B true CN114037705B (en) 2022-03-25

Family

ID=80141541

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210024107.4A Active CN114037705B (en) 2022-01-11 2022-01-11 Metal fracture fatigue source detection method and system based on moire lines

Country Status (1)

Country Link
CN (1) CN114037705B (en)

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104867130B (en) * 2015-04-21 2018-09-18 浙江工业大学 A kind of self-adapting division method based on crack image sub-district gray average
KR102015706B1 (en) * 2017-11-23 2019-08-28 울산과학기술원 Apparatus and method for measureing a crack in a concrete surface
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
CN111681217B (en) * 2020-05-29 2023-06-20 香港生产力促进局 Machine vision image intelligent analysis method, device and system for fatigue fracture
CN112881642B (en) * 2021-01-15 2022-05-13 宝武杰富意特殊钢有限公司 Analysis method of material failure crack source
CN113706423B (en) * 2021-10-28 2022-01-07 南通皋亚钢结构有限公司 Artificial intelligence-based mechanical part corrosion crack detection method

Also Published As

Publication number Publication date
CN114037705A (en) 2022-02-11

Similar Documents

Publication Publication Date Title
CN108520274B (en) High-reflectivity surface defect detection method based on image processing and neural network classification
CN112598001B (en) Automatic ship water gauge reading identification method based on multi-model fusion
CN109816644B (en) Bearing defect automatic detection system based on multi-angle light source image
CN104751187B (en) Meter reading automatic distinguishing method for image
CN109900711A (en) Workpiece, defect detection method based on machine vision
CN106600600A (en) Wafer defect detection method based on characteristic matching
CN113610814B (en) Weld quality detection method and device, electronic equipment and storage medium
CN109376740A (en) A kind of water gauge reading detection method based on video
CN105095822B (en) A kind of Chinese letter co pattern image detection method and system
CN109540925B (en) Complex ceramic tile surface defect detection method based on difference method and local variance measurement operator
CN109685766A (en) A kind of Fabric Defect detection method based on region fusion feature
CN102032875A (en) Image-processing-based cable sheath thickness measuring method
CN110189375A (en) A kind of images steganalysis method based on monocular vision measurement
Garz et al. A binarization-free clustering approach to segment curved text lines in historical manuscripts
CN113340909A (en) Glue line defect detection method based on machine vision
CN110096980A (en) Character machining identifying system
CN112085700A (en) Automatic extraction method, system and medium for weld joint region in X-ray image
CN108961262A (en) A kind of Bar code positioning method under complex scene
CN113705564B (en) Pointer type instrument identification reading method
TW201601119A (en) Method for recognizing and locating object
CN114037705B (en) Metal fracture fatigue source detection method and system based on moire lines
CN116596921A (en) Method and system for sorting incinerator slag
CN116452826A (en) Coal gangue contour estimation method based on machine vision under shielding condition
CN118506338A (en) Electronic device printed character recognition and detection method based on deep learning
CN111191534B (en) Road extraction method in fuzzy aviation image

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