CN112581456B - Image crack real-time extraction method suitable for ultrasonic imaging logging - Google Patents

Image crack real-time extraction method suitable for ultrasonic imaging logging Download PDF

Info

Publication number
CN112581456B
CN112581456B CN202011537010.0A CN202011537010A CN112581456B CN 112581456 B CN112581456 B CN 112581456B CN 202011537010 A CN202011537010 A CN 202011537010A CN 112581456 B CN112581456 B CN 112581456B
Authority
CN
China
Prior art keywords
crack
sub
region
image
area
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
CN202011537010.0A
Other languages
Chinese (zh)
Other versions
CN112581456A (en
Inventor
张伟
李焱骏
师奕兵
吴桐
罗欣怡
李志鹏
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.)
University of Electronic Science and Technology of China
Original Assignee
University of Electronic Science and Technology of China
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 University of Electronic Science and Technology of China filed Critical University of Electronic Science and Technology of China
Priority to CN202011537010.0A priority Critical patent/CN112581456B/en
Publication of CN112581456A publication Critical patent/CN112581456A/en
Application granted granted Critical
Publication of CN112581456B publication Critical patent/CN112581456B/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
    • E—FIXED CONSTRUCTIONS
    • E21—EARTH DRILLING; MINING
    • E21B—EARTH DRILLING, e.g. DEEP DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS
    • E21B47/00—Survey of boreholes or wells
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • 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
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
    • 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
    • G06T2207/10132—Ultrasound 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30132—Masonry; Concrete
    • 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
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/30—Assessment of water resources

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Geology (AREA)
  • Mining & Mineral Resources (AREA)
  • Health & Medical Sciences (AREA)
  • Biophysics (AREA)
  • Environmental & Geological Engineering (AREA)
  • Geophysics (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • Geochemistry & Mineralogy (AREA)
  • Quality & Reliability (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Fluid Mechanics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses an image crack real-time extraction method suitable for ultrasonic imaging logging, which comprises the steps of firstly, carrying out primary segmentation on an original image according to an Otus segmentation algorithm, and extracting effective crack information to establish a search space; then, referring to the crack development form, determining the crack reference position through a voting accumulation mechanism, and properly cutting the area where the crack is located according to the reference position; then, cutting the image obtained by cutting along the longitudinal direction to obtain each subarea with partial crack information, and performing crack search and extraction on each subarea in parallel by using an ant colony algorithm; and finally splicing the sub-regions searched to obtain the crack path to obtain the complete crack.

Description

Image crack real-time extraction method suitable for ultrasonic imaging logging
Technical Field
The invention belongs to the technical field of well logging, and particularly relates to a real-time image crack extraction method suitable for ultrasonic imaging well logging.
Background
In the field of oil well logging, formation fractures are discontinuous sections which are widely distributed in different lithologies and are gradually formed through different diagenesis or tectonic deformation. The presence or absence of fractures is often a key indicator of the producibility of hydrocarbon reservoirs in tight formations. The well periphery imaging well logging technology is one of the mainstream stratum information acquisition methods at present. The imaging logging around the well can reflect the condition of the oil well by visual well wall images, can clearly see the development conditions of cracks and holes on the well wall, and is an important means for evaluating the oil well.
The logging method mainly used in the field of imaging logging around the well comprises an ultrasonic imaging method and a microresistivity scanning method, wherein the ultrasonic imaging method utilizes the echo reflection principle of ultrasonic waves in different media to digitize the arrival time and the amplitude of received echoes, and then the echoes are arrayed in the depth direction to form a logging image. Compared with the microresistivity scanning method, the ultrasonic imaging method has lower resolution, but is widely applied due to strong penetrating power, high well coverage rate and simple instrument structure.
As shown in fig. 1, a probe of the ultrasonic imaging logging tool continuously rotates to scan the borehole wall to form a scan line, the scan line is arranged along the depth to form a logging image, and each pixel in the image corresponds to the arrival time of the ultrasonic signal reflected by the borehole wall. The shaded area indicates that the echo arrival time measured in this region is small, whereas in turn the echo arrival time is large or even no echo reflections.
The crack identification method based on ultrasonic imaging logging is numerous, some mathematical models are used for fitting scattered points meeting the sine shape, but the well wall is damaged in actual drilling, and the crack shape is not necessarily standard sine, so that poor identification is caused; some maximum entropy threshold segmentation algorithms separate pixels corresponding to cracks from images, but part of non-crack information is often left, and detection results are affected. According to the method, an Otus-based threshold segmentation algorithm and an improved ant colony algorithm are combined to filter unnecessary interference information, and cracks are identified and extracted quickly and accurately.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides an image crack real-time extraction method suitable for ultrasonic imaging logging.
In order to achieve the purpose, the invention discloses an image crack real-time extraction method suitable for ultrasonic imaging logging, which is characterized by comprising the following steps of:
(1) Performing primary threshold segmentation on the original ultrasonic well logging image G by using an Otsu segmentation algorithm;
(1.1) calculating the between-class variance of the image G
Figure BDA0002853370930000021
/>
Figure BDA0002853370930000022
μ T Pixels with a gray level of 0 to d, ω (d), which is the gray level mean of the image G, are divided into the crack regions G 1 μ (d) is the mean value of the gray levels of the pixels with gray levels from 0 to d;
(1.2) finding an optimal segmentation threshold T * So that the crack region G 1 And a background region G 2 The sum of the information entropies of (a);
Figure BDA0002853370930000023
(1.3) all pixel points of the image G are subjected to threshold value T * Dividing the image into a crack region G 1 Setting the corresponding gray value as 0, and classifying the rest pixel points as background areas G 2 The corresponding gray value is set as 1;
(1.4) forming a crack region G by using a number method 1 To form a feasible grid, the background region G 2 Converting into an infeasible grid, thereby generating a grid set corresponding to the binary image as a search space;
(2) Through the base line y 0 Determines a target search area G in the search space s ；
(2.1) determining a baseline position; tong (Chinese character of 'tong')Over fracture curve y = A sin (ω x + β) + y 0 Up searching point pair s 1 ＝(x 1 ,y 1 ) And s 2 ＝(x 2 ,y 2 ) To satisfy x 1 -x 2 = T/2, where A is the amplitude of the crack curve, β is the initial phase, ω is the angular velocity, y 0 Is the baseline position, and T is the period of the fracture curve;
then the midpoint s of such a point pair is satisfied 0 ＝(x 0 ,y 0 ) Must fall at base line y 0 Then, the voting accumulation mechanism is used to count the ordinate information of all the midpoints so as to determine the baseline y 0 The position of (a);
(2.2) taking the base line y 0 Cutting off the region to obtain a target search region G s ；
(3) Dividing the target search region G s ；
Searching for a target area G s Dividing the substrate into L different regions along the longitudinal direction, and ensuring each sub-region G in the dividing process sl The inner crack area is still left-right through, wherein L =1,2, \8230, L;
(4) Simultaneously aiming each subarea G by utilizing ant colony algorithm sl Performing path search to complete real-time extraction of image cracks;
(4.1) setting the number of ant colony ants as K; the number of pixel points in the l sub-area is n; any two pixel points in the first sub-region
Figure BDA0002853370930000031
And/or>
Figure BDA0002853370930000032
Is at a distance of->
Figure BDA0002853370930000033
And i is not equal to j; any two pixel points in the first sub-region
Figure BDA0002853370930000034
And/or>
Figure BDA0002853370930000035
In between as initial pheromone concentration>
Figure BDA0002853370930000036
Any two pixel points in the ith sub-area at the moment t>
Figure BDA0002853370930000037
And/or>
Figure BDA0002853370930000038
Has an pheromone concentration of->
Figure BDA0002853370930000039
Randomly selecting one pixel point on the left side and one pixel point on the right side in the first sub-area>
Figure BDA00028533709300000310
And/or>
Figure BDA00028533709300000311
As a starting point and an end point of a path search;
(4.2) calculating slave pixel points of the ant k in the ith sub-area at the moment t
Figure BDA00028533709300000312
Transferred to the pixel point->
Figure BDA00028533709300000313
Is greater than or equal to>
Figure BDA00028533709300000314
Figure BDA00028533709300000315
Wherein, the first and the second end of the pipe are connected with each other,
Figure BDA00028533709300000316
indicating the current time t in the ith sub-regionPrime point->
Figure BDA00028533709300000317
Transferred to a pixel point>
Figure BDA00028533709300000318
The desired degree of; all k For ant k, the set of pixel points to be searched, r represents allow k Any one pixel point; a is a pheromone importance factor; b is a heuristic function importance factor;
(4.3) through roulette selection method, ant k from the beginning in the first subregion
Figure BDA00028533709300000319
Starting to go to the next pixel point according to the transfer probability until the terminal point is reached>
Figure BDA00028533709300000320
And then updating the pheromone concentration between the traversed pixel points>
Figure BDA00028533709300000321
Figure BDA00028533709300000322
Figure BDA00028533709300000323
Wherein rho is the volatilization rate of the pheromone,
Figure BDA00028533709300000324
for ant k in the l sub-area from pixel point->
Figure BDA00028533709300000325
To the pixel point
Figure BDA00028533709300000326
Based on the pheromone concentration of->
Figure BDA00028533709300000327
For all ants in the l sub-area from pixel point->
Figure BDA00028533709300000328
To a pixel point>
Figure BDA00028533709300000329
Q is a constant;
(4.4) when the number of iterations reaches the maximum value, the path search is ended, and each sub-region G sl And obtaining a specific crack search path, and then splicing all sub-areas of the obtained specific crack search path to finish the crack extraction of the logging image.
The invention aims to realize the following steps:
the invention relates to an image crack real-time extraction method suitable for ultrasonic imaging logging, which comprises the steps of firstly, carrying out primary segmentation on an original image according to an Otus segmentation algorithm, and extracting effective crack information to establish a search space; then, referring to the crack development form, determining the crack reference position through a voting accumulation mechanism, and properly cutting the area where the crack is located according to the reference position; then, cutting the cut image along the longitudinal direction to obtain each subarea with partial crack information, and performing crack search and extraction on each subarea in parallel by using an ant colony algorithm; and finally splicing the sub-regions searched to obtain the crack path to obtain the complete crack.
Meanwhile, the image crack real-time extraction method suitable for the ultrasonic imaging logging further has the following beneficial effects:
(1) The invention divides the search space into a plurality of sub-areas for parallel path search, so that the algorithm is quickly converged and the search speed is increased.
(2) And because each area is independently ant colony for path search, the condition that the ant colony falls into local optimal search can be effectively improved.
(3) The method has high identification accuracy, and can accurately keep crack information no matter whether the crack is a sine crack, a horizontal crack or a crack with obvious background noise, so that the identification and extraction of the crack area in the logging image are realized.
Drawings
FIG. 1 is a schematic diagram of an ultrasonic imaging tool configuration;
FIG. 2 is a general block diagram of the image fracture real-time extraction method of the present invention suitable for ultrasonic imaging logging;
FIG. 3 is a diagram of an overall architecture suitable for use with an ultrasonic imaging tool;
FIG. 4 is a flow chart of data acquisition for an ultrasonic imaging tool;
FIG. 5 is a flow chart for establishing a search space;
FIG. 6 is a graph of threshold-partitioned binary matrix data;
FIG. 7 is a determination of a baseline position;
FIG. 8 is a target search area diagram;
FIG. 9 is a diagram of vertically divided sub-regions;
FIG. 10 is a diagram of a small area path parallel search strategy;
FIG. 11 is a crack searched by the ant colony algorithm;
FIG. 12 is a final identified and extracted fracture map.
Detailed Description
The following description of the embodiments of the present invention is provided in order to better understand the present invention for those skilled in the art with reference to the accompanying drawings. It is to be expressly noted that in the following description, a detailed description of known functions and designs will be omitted when it may obscure the subject matter of the present invention.
Examples
FIG. 2 is a general block diagram of the image fracture real-time extraction method suitable for ultrasonic imaging logging according to the invention.
In this embodiment, as shown in fig. 2, the method for extracting an image fracture in real time, which is suitable for ultrasonic imaging logging, of the present invention includes the following steps:
s1, image segmentation
The FPGA internal framework of the ultrasonic imaging logging instrument is shown in figure 3, the receiving command is mainly used for processing commands sent by an upper computer and a PIC, the transmitting module is used for controlling an ultrasonic transmitting channel, the collecting and calculating module is used for storing and processing data of the ADC, the EDIB transmitting module is used for sending the processed data to the upper computer through M2, M5 and M7, the upper computer is enabled to image, and the modules in the FPGA coordinate through various control signals.
As shown in fig. 4, the PIC transmits a mechanical synchronization signal generated by a sensor of a mechanical structure part of the ultrasonic imaging logging tool to the FPGA, transmits a sound wave when the FPGA receives a tooth signal and a bodymar signal, generates a transmission flag bit, and then judges whether the transmission reaches 250 times, namely, one week. When the emission reaches 250 times, the flag bit of the full period of emission is output, and the stage of emission of the mud wave is entered.
When the FPGA receives the transmitting zone bit and transmits the zone bit of one full period, the ADC starts to collect according to the set waiting transition time, the collected data is stored in the RAM, and a once-collecting zone bit is generated, when the zone bit of one week of transmission is received, a zone bit for completing the collection is generated, and then the slurry echo starts to be collected.
And when the FPGA receives the once-collected zone bit, the FPGA acquires data from the RAM and starts to calculate to obtain the arrival time and the amplitude of the first wave, and when the once-collected zone bit is received, the FPGA starts to calculate the sound velocity of the slurry wave.
And then constructing the calculated data into matrix data G corresponding to the image and storing the matrix data G into the RAM.
When the data in the RAM is stored to a certain amount, the FPGA on the main control board reads the data in the RAM to construct an 88 x 88 matrix required by the image G, and an inter-class variance evaluation function which is convenient for simple calculation is adopted
Figure BDA0002853370930000061
As a criterion of threshold segmentation, the specific segmentation steps are shown in fig. 5 as follows:
s1.1, calculating the between-class variance of the image G
Figure BDA0002853370930000062
/>
Figure BDA0002853370930000063
μ T Pixels with omega (d) of 0 to d are classified into the crack region G as the gray-scale mean of the image G 1 μ (d) is the gray mean of pixels with gray levels 0 to d;
in the present embodiment, the inter-class variance is obtained
Figure BDA0002853370930000064
Is 1852.9;
s1.2, obtaining an optimal segmentation threshold value T * So that the crack region G 1 And a background region G 2 The sum of the information entropies of (a);
Figure BDA0002853370930000065
in the present embodiment, the one obtained according to the previous step
Figure BDA0002853370930000066
Finding the optimal threshold T * Is 0.524;
s1.3, all pixel points of the image G are subjected to threshold T * Dividing the image into a crack region G 1 Setting the corresponding gray value as 0, and classifying the rest pixel points as background areas G 2 Setting the corresponding gray value as 1, so that the ultrasonic well logging image after threshold segmentation is converted into a binary image, as shown in fig. 6;
s1.4, utilizing a sequence number method to divide a crack area G 1 To form a feasible grid, the background region G 2 The grid is converted into an infeasible grid, so that a grid set corresponding to the binary image is generated as a search space, as shown in fig. 5;
s2, passing through base line y 0 Determines a target search area G in the search space s ；
S2.1, determining a baseline position; by curve y = A sin (ω x + β) + y at the crack 0 Up looking for point pair s 1 ＝(x 1 ,y 1 ) And s 2 ＝(x 2 ,y 2 ) To satisfy x 1 -x 2 = T/2, where A is the amplitude of the crack curve, β is the initial phase, ω is the angular velocity, y 0 Is the baseline position, and T is the period of the fracture curve;
then the midpoint s of such a point pair is satisfied 0 ＝(x 0 ,y 0 ) Must fall at base line y 0 Then, the address information of all the midpoints is counted by using a voting accumulation mechanism, so as to determine the baseline y 0 Is the row in which y =29 is located, i.e. the position of yellow marked in fig. 7;
s2.2, comparing the base line y 0 Cutting off the region to obtain a target search region G s A 22 x 88 matrix is shown in fig. 8;
s3, dividing a target search area G s ；
Searching for a target area G s Dividing the wafer into L different regions along the longitudinal direction, and ensuring each sub-region G in the dividing process sl The inner crack area is still left-right through, wherein L =1,2, ... in this embodiment, the total division into 4 sub-regions of 22 × 22 size is shown in fig. 9;
s4, simultaneously carrying out comparison on the four sub-regions G by utilizing the ant colony algorithm sl Performing path search to finish real-time extraction of image cracks, as shown in fig. 10;
s4.1, setting the number of ant colony ants as K =20; the number of pixel points in the ith sub-area is n =22 × 22; any two pixel points in the first sub-region
Figure BDA0002853370930000071
And/or>
Figure BDA0002853370930000072
Is at a distance of->
Figure BDA0002853370930000073
And i ≠j; any two pixel points in the l-th sub-area>
Figure BDA0002853370930000074
And/or>
Figure BDA0002853370930000075
Has an initial pheromone concentration of ^ 5>
Figure BDA0002853370930000076
Any two pixel points in the ith sub-area at the moment t>
Figure BDA0002853370930000077
And/or>
Figure BDA0002853370930000078
Has a pheromone concentration of->
Figure BDA0002853370930000079
Randomly selecting one pixel point on the left side and one pixel point on the right side in the first sub-area>
Figure BDA00028533709300000710
And/or>
Figure BDA00028533709300000711
As a starting point and an end point of a path search;
s4.2, calculating slave pixel points of the ant k in the first sub-area at the moment t
Figure BDA00028533709300000712
Transferred to the pixel point->
Figure BDA00028533709300000713
Is greater than or equal to>
Figure BDA00028533709300000714
Figure BDA00028533709300000715
Wherein the content of the first and second substances,
Figure BDA00028533709300000716
indicates that the current time t in the ith sub-area is greater than or equal to>
Figure BDA00028533709300000717
Transferred to the pixel point->
Figure BDA00028533709300000718
The desired degree of; all k For ant k, the set of pixel points to be searched, r represents allow k Any one pixel point; a is a pheromone importance factor; b is a heuristic function importance factor;
s4.3, through roulette selection method, ant k starts from the beginning in the first sub-area
Figure BDA00028533709300000719
Starting to go to the next pixel point according to the transfer probability until the destination is reached>
Figure BDA00028533709300000720
And then updating the pheromone concentration between the traversed pixel points>
Figure BDA00028533709300000721
Figure BDA00028533709300000722
Figure BDA00028533709300000723
Wherein rho is the volatilization rate of the pheromone,
Figure BDA00028533709300000724
is ant k in the first sub-areaSlave pixel point->
Figure BDA00028533709300000725
To the pixel point
Figure BDA00028533709300000726
Based on the pheromone concentration of->
Figure BDA00028533709300000727
Slave pixel point->
Figure BDA00028533709300000728
To the pixel point->
Figure BDA00028533709300000729
Q is a constant;
s4.4, when the iteration number reaches the maximum value, the path search is finished, and the four sub-regions G sl Specific fracture search paths are obtained as shown in fig. 11, then all sub-regions of the specific fracture search paths are spliced, and fracture extraction of the logging image is completed as shown in fig. 12.
Although illustrative embodiments of the present invention have been described above to facilitate the understanding of the present invention by those skilled in the art, it should be understood that the present invention is not limited to the scope of the embodiments, and various changes may be made apparent to those skilled in the art as long as they are within the spirit and scope of the present invention as defined and defined by the appended claims, and all matters of the invention which utilize the inventive concepts are protected.

Claims (1)

1. An image crack real-time extraction method suitable for ultrasonic imaging logging is characterized by comprising the following steps:
(1) Performing primary threshold segmentation on the original ultrasonic well logging image G by using an Otsu segmentation algorithm;
(1.1) calculating the between-class variance of the image G
Figure QLYQS_1
Figure QLYQS_2
μ T The pixels with the gray scale of 0 to d are classified into the crack region G as the gray scale mean of the image G and ω (d) 1 μ (d) is the gray mean of pixels with gray levels 0 to d;
(1.2) finding an optimal segmentation threshold T * So that a crack region G 1 And a background region G 2 The sum of the information entropies of (a);
Figure QLYQS_3
(1.3) all pixel points of the image G are subjected to threshold value T * Dividing, and classifying pixel points smaller than a threshold value as a crack region G 1 Setting the corresponding gray value as 0, and classifying the rest pixel points as background areas G 2 The corresponding gray value is set to 1;
(1.4) formation of crack region G by the number method 1 To form a feasible grid, the background region G 2 Converting into an infeasible grid, thereby generating a grid set corresponding to the binary image as a search space;
(2) Through the base line y 0 Determines a target search area G in the search space s ；
(2.1) determining a baseline position; by curve y = Asin (ω x + β) + y at the crack 0 Up searching point pair s 1 ＝(x 1 ,y 1 ) And s 2 ＝(x 2 ,y 2 ) To satisfy x 1 -x 2 = T/2, where A is the amplitude of the crack curve, β is the initial phase, π is the angular velocity, y 0 Is the baseline position, and T is the period of the fracture curve;
then the midpoint s of such a point pair is satisfied 0 ＝(x 0 ,y 0 ) Must fall at base line y 0 Then utilizing a voting accumulation mechanism to make the systemCalculating the ordinate information of all the middle points, thereby determining the base line y 0 The position of (a);
(2.2) taking the base line y 0 Cutting off the region to obtain a target search region G s ；
(3) Dividing the target search region G s ；
Searching for a target area G s Dividing the wafer into L different regions along the longitudinal direction, and ensuring each sub-region G in the dividing process sl The inner crack area is still left-right through, wherein L =1,2, ...
(4) Simultaneously aiming each subarea G by utilizing ant colony algorithm sl Performing path search to complete real-time extraction of image cracks;
(4.1) setting the number of ant colony ants as K; the number of pixel points in the l sub-area is n; any two pixel points in the first sub-region
Figure QLYQS_6
And
Figure QLYQS_9
is a distance of
Figure QLYQS_11
And i is not equal to j; any two pixel points in the first sub-region
Figure QLYQS_5
And
Figure QLYQS_8
initial pheromone concentration of
Figure QLYQS_12
Any two pixel points in the first sub-area at the moment t
Figure QLYQS_14
And with
Figure QLYQS_4
Has a pheromone concentration of
Figure QLYQS_7
Randomly selecting one pixel point at each of the left side and the right side in the first sub-area
Figure QLYQS_10
And
Figure QLYQS_13
as a starting point and an end point of a path search;
(4.2) calculating slave pixel points of the ant k in the ith sub-area at the moment t
Figure QLYQS_15
Transfer to pixel
Figure QLYQS_16
Transition probability of
Figure QLYQS_17
Figure QLYQS_18
/>
Wherein the content of the first and second substances,
Figure QLYQS_19
representing the current time t in the first sub-region
Figure QLYQS_20
Transfer to pixel
Figure QLYQS_21
The desired degree of (d); all k For ant k, the set of pixel points to be searched, r represents allow k Any one pixel point; a is a pheromone importance factor; b is a heuristic function importance factor;
(4.3) Ant k from the beginning in the first subregion by roulette selection
Figure QLYQS_22
Starting to go to the next pixel point according to the transfer probability until the terminal point is reached>
Figure QLYQS_23
Then updating pheromone concentration between all traversed pixel points>
Figure QLYQS_24
Figure QLYQS_25
Figure QLYQS_26
Wherein ρ is the volatilization rate of the pheromone,
Figure QLYQS_27
for ant k to follow pixel point in the first sub-area
Figure QLYQS_28
To pixel point
Figure QLYQS_29
The concentration of the pheromone of (a),
Figure QLYQS_30
all ants are driven to pixel points in the first sub-area
Figure QLYQS_31
To the pixel point
Figure QLYQS_32
Q is a constant;
(4.4) when the number of iterations reaches the maximum value, the path search ends, each sub-region G sl And obtaining a specific crack search path, and then splicing all sub-regions of the obtained specific crack search path to finish the crack extraction of the logging image.
CN202011537010.0A 2020-12-23 2020-12-23 Image crack real-time extraction method suitable for ultrasonic imaging logging Active CN112581456B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011537010.0A CN112581456B (en) 2020-12-23 2020-12-23 Image crack real-time extraction method suitable for ultrasonic imaging logging

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011537010.0A CN112581456B (en) 2020-12-23 2020-12-23 Image crack real-time extraction method suitable for ultrasonic imaging logging

Publications (2)

Publication Number Publication Date
CN112581456A CN112581456A (en) 2021-03-30
CN112581456B true CN112581456B (en) 2023-04-18

Family

ID=75139443

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011537010.0A Active CN112581456B (en) 2020-12-23 2020-12-23 Image crack real-time extraction method suitable for ultrasonic imaging logging

Country Status (1)

Country Link
CN (1) CN112581456B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB2608149B (en) * 2021-06-24 2024-03-27 Darkvision Tech Inc Machine learning model for measuring perforations in a tubular

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101377850A (en) * 2008-09-27 2009-03-04 北京航空航天大学 Method of multi-formwork image segmentation based on ant colony clustering
CN109313247A (en) * 2016-05-31 2019-02-05 上海联影医疗科技有限公司 System and method for removing gibbs artifact in medical image system
CN109543622A (en) * 2018-11-26 2019-03-29 长春工程学院 A kind of electric transmission line isolator image partition method
CN109993721A (en) * 2019-04-04 2019-07-09 电子科技大学成都学院 A kind of image enchancing method based on clustering algorithm and ant group algorithm

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101377850A (en) * 2008-09-27 2009-03-04 北京航空航天大学 Method of multi-formwork image segmentation based on ant colony clustering
CN109313247A (en) * 2016-05-31 2019-02-05 上海联影医疗科技有限公司 System and method for removing gibbs artifact in medical image system
CN109543622A (en) * 2018-11-26 2019-03-29 长春工程学院 A kind of electric transmission line isolator image partition method
CN109993721A (en) * 2019-04-04 2019-07-09 电子科技大学成都学院 A kind of image enchancing method based on clustering algorithm and ant group algorithm

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
王小燕 ; 许建荣 ; .图像分割技术在血管图像中的应用.中国介入影像与治疗学.(第01期), *

Also Published As

Publication number Publication date
CN112581456A (en) 2021-03-30

Similar Documents

Publication Publication Date Title
Liu et al. Detection and localization of rebar in concrete by deep learning using ground penetrating radar
CN109343022B (en) Method for estimating interlayer soil water content
CN112462346A (en) Ground penetrating radar roadbed defect target detection method based on convolutional neural network
CN112232392B (en) Data interpretation and identification method for three-dimensional ground penetrating radar
CN111476088B (en) Asphalt pavement water damage identification model construction method, identification method and system
CN111025286B (en) Ground penetrating radar map self-adaptive selection method for water damage detection
AU2013276319B2 (en) Methods for determining material and/or subsurface composition
CN106446919A (en) Quick detection method for ground penetrating radar hyperbolic curve target
CN113504306B (en) Steel rail defect detection method based on ultrasonic phased array low-rank matrix recovery
CN116597365A (en) Underground disease object identification method based on neural network
CN112581456B (en) Image crack real-time extraction method suitable for ultrasonic imaging logging
JP5719075B1 (en) Cavity thickness exploration method
CN103616390A (en) Non-destructive detection method for cemented filling body roof fracture state
JP5629840B1 (en) Cavity thickness exploration method
CN108830140B (en) Volcanic lithology identification method based on electric imaging logging fractal dimension
CN113359101A (en) Underground target detection method, system and computer storage medium
Chen et al. A novel image-based approach for interactive characterization of rock fracture spacing in a tunnel face
CN116184401A (en) System and method for engineering quality inspection
CN113468804B (en) Underground pipeline identification method based on matrix bundles and deep neural network
JP6903196B1 (en) Road surface area detection device, road surface area detection system, vehicle and road surface area detection method
CN113988142B (en) Tunnel lining cavity acoustic identification method based on convolutional neural network
CN114487110A (en) Floor tile hollowing automatic detection method and device
CN115223044A (en) End-to-end three-dimensional ground penetrating radar target identification method and system based on deep learning
JPH0572332A (en) Detecting method for invisible object
Guo et al. Research on tunnel lining image target recognition method based on YOLOv3

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