CN107527354A - A kind of region growing method based on composite diagram - Google Patents
A kind of region growing method based on composite diagram Download PDFInfo
- Publication number
- CN107527354A CN107527354A CN201710547605.6A CN201710547605A CN107527354A CN 107527354 A CN107527354 A CN 107527354A CN 201710547605 A CN201710547605 A CN 201710547605A CN 107527354 A CN107527354 A CN 107527354A
- Authority
- CN
- China
- Prior art keywords
- crack
- composite diagram
- seeds
- image
- 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.)
- Granted
Links
- 239000002131 composite material Substances 0.000 title claims abstract description 65
- 238000010586 diagram Methods 0.000 title claims abstract description 65
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000001914 filtration Methods 0.000 claims abstract description 24
- 230000000877 morphologic effect Effects 0.000 claims abstract description 18
- 239000012535 impurity Substances 0.000 claims abstract description 9
- 238000004364 calculation method Methods 0.000 claims description 7
- 230000001154 acute effect Effects 0.000 claims description 6
- 238000005260 corrosion Methods 0.000 claims description 4
- 230000007797 corrosion Effects 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 4
- 241001632422 Radiola linoides Species 0.000 claims 1
- 230000015572 biosynthetic process Effects 0.000 claims 1
- 238000003786 synthesis reaction Methods 0.000 claims 1
- 241001632427 Radiola Species 0.000 description 3
- 241001270131 Agaricus moelleri Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—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
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
- G06T7/155—Segmentation; Edge detection involving morphological operators
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of region growing method based on composite diagram:Step 1, gray proces, filtering process and thresholding are carried out to original fracture image to handle, obtains composite diagram;Step 2, Morphological scale-space and filtration treatment are carried out to composite diagram, obtains the drawing of seeds in target crack;Step 3, according to the minimum enclosed rectangle of all connected regions in the drawing of seeds in target crack, grown after selecting the growth district in the drawing of seeds to target crack, the crack pattern picture after being grown.The present invention is grown according to the mode that drawing of seeds and composite diagram separate, growth position is determined by drawing of seeds and acts on the composite diagram of relevant position, judged whether to grow and then returned in drawing of seeds by composite diagram and grown, avoid the other positions in region and grow impurity information, so that the crack information after growth becomes apparent from completely, accuracy is high;Due to judging that growth is in composite diagram rather than artwork, the traversal small complexity in region is low, and treatment effeciency is high.
Description
Technical field
The invention belongs to field of road, is related to a kind of image processing techniques, specifically a kind of area based on composite diagram
Domain growing method.
Background technology
In the processing application to image, scientific research personnel only feels to some in image or some parts under many circumstances
Interest, these parts are referred to as target or prospect, and they often correspond to the area for having different from other background properties in image
Domain.For more preferable research and analysis target, it is necessary to which they are therefrom split, be then possible to further carry out effectively
Utilization.
Image segmentation is a kind of important image processing techniques, and its application is very extensive, is almost related to
The field that all relevant image procossings occur, it not only occurs in the lab, more closely bound up with our daily life.Area
Domain growth is exactly a kind of method of region segmentation, and the method for region growing, which generally requires divided region, has certain specific
Common trait, by the way that whether the point constantly compared around seed point meets fusion standard so as to reaching the growth of relevant range.It is existing
Some growing methods are typically the direct growth in artwork, and the point that many need not grow usually is considered as into seed under this mode
Point, add the task amount of growth so that run time increase, efficiency step-down.
The content of the invention
In view of the above-mentioned drawbacks of the prior art or insufficient, it is an object of the present invention to provide one kind to be based on composite diagram
Region growing method.
To achieve these goals, the present invention, which adopts the following technical scheme that, is achieved:
A kind of region growing method based on composite diagram, the processing step of this method are as follows:
Step 1, gray proces, filtering process and thresholding processing are carried out successively to original fracture image, obtains thresholding
Image after processing, the non-crack area of image obtained after thresholding is white (pixel 255), by the coordinate pair of non-crack area
255 should be put to the relevant position in original fracture gradation of image figure, and by the position pixel, obtained image is as composite diagram;
Step 2, Morphological scale-space and filtration treatment are carried out successively to composite diagram, obtains the drawing of seeds in target crack;
Step 3, according to the minimum enclosed rectangle of all connected regions in the drawing of seeds in target crack, select to target crack
Drawing of seeds in growth district after grown, the crack pattern picture after being grown.
Further, the step 1 specifically includes following sub-step:
Step 11, the target crack of fracture picture first carries out gray processing processing, obtains gray level image;
Step 12, gray level image is divided into multiple size identical block of pixels, each block of pixels is filtered respectively
Processing uses mean filter or median filter method, to remove invalid interference impurity, so as to obtain the image after filtering process;
Step 13, thresholding processing is carried out to the image after filtering process, obtains the image after thresholding processing.
Step 14, the non-crack area coordinate of the image after thresholding is corresponded in the gray level image of step 11, by phase
Answer position coordinates pixel to put 255, obtain composite diagram.
Further, the step 2 specifically includes following sub-step:
Step 21, to composite diagram carry out Morphological scale-space, the Morphological scale-space specifically successively carry out expansion process and
Corrosion treatment, obtain the image after Morphological scale-space;
Step 22, the crack in the image after Morphological scale-space is filtered;Filtering rule:Filter out boundary rectangle
Length-width ratio is less than 1/3 crack, meanwhile, the average of all connected region areas, filtering in the image after computation of morphology processing
Fall the crack that area is less than the average, obtain the drawing of seeds in target crack.
Further, the step 3 specifically includes following sub-step:
Step 31, target crack is calculated for convenience, is obtained length, width, the positional information in crack first, is extracted target
The minimum enclosed rectangle of all connected regions in the drawing of seeds in crack;
Step 32, by the angle direction in crack, determine that growth district is in the position of minimum enclosed rectangle, be specifically:Such as
Shown in Fig. 1, if acute angle, then growth district is in the top in the upper right corner of boundary rectangle and the lower section in the lower left corner;It is if blunt
Angle, then growth district is in the left side in the upper left corner of boundary rectangle and the right side in the lower right corner;;Obtain detection figure;
Step 33, in order to ensure the crack of no careless omission, the starting point for the growth district that step 32 is obtained is as seed point
The composite diagram relevant position corresponded in step 14, judge whether to have in the range of the n*n centered on seed point undetected
Crack, it is specifically:
According to seed point coordinates, the connected domain of respective coordinates in composite diagram is found, calculates all gray scales in the connected region
Average value of the value less than 255 gray-value pixel points;Travel through each picture that current minimum enclosed rectangle is not belonging in the n*n scopes
Element, if the absolute difference of some pixel value and average value takes 2~6 not less than constant K, K, then it is assumed that the pixel does not split
Stitch information;If less than K, then it is assumed that the pixel is crack information, so as to obtain new crack;According to all new cracks
Length, width, positional information, extract the minimum enclosed rectangle of each connected domain;(for a discrete pixel, not by it
As crack information;) perform step 32;If not detecting new crack, step 34 is carried out;Step 34, in order to avoid life
More impurity information are grown, all seed points that step 33 is obtained correspond to corresponding positions in the composite diagram that step 13 obtains
Put, whether calculating is grown, and result of calculation is then grown if growth in the drawing of seeds in target crack;Otherwise, do not give birth to
It is long;Crack pattern picture after then being grown.
Further, the search n*n scopes in the step 33, wherein n determination process are as follows:
The distance in each crack crack adjacent thereto in the drawing of seeds in the target crack that calculation procedure 22 obtains, statistics are all
Two-by-two fracture interval from, calculate all distance values average value and intermediate value sum 1/2, be the detection figure convolution n.(such as
Fruit result is integer and is even number, then n takes the larger odd number adjacent with the even number;If n is decimal and integer part is strange
Count, then n round numbers part;If integer part is even number, n takes the larger odd number adjacent with the even number.)
Further, the interstitial distance of calculating refers to calculate the distance between the center of mass point of crack.
Further, the step 34 specifically includes following sub-step:
What whether the calculating grew comprises the following steps that:The drawing of seeds in the target crack obtained due to step 22 is by step
Rapid 14 obtained composite diagrams handle what is obtained step by step, so part connected region includes the seed in target crack in composite diagram
The seed point of figure.According to the seed point coordinates of the drawing of seeds in target crack, the connected domain of respective coordinates in composite diagram is found, is calculated
All average values for being less than 255 gray-value pixel points in the connected region, if the pixel in composite diagram with seed point corresponding points
Value and the absolute difference of the average value take 2~6 not less than constant K, K, then it is assumed that the growth district is not crack area, will
The pixel value that seed point is corresponded in the drawing of seeds in target crack is set to 0, i.e., does not grow;If less than K, then it is assumed that the growth district is
Crack area, the pixel value that the drawing of seeds in target crack is corresponded to seed point put 1, i.e., are grown using the seed point as starting point.
Compared with existing growing method, beneficial effects of the present invention are as follows:
1st, in method of the invention, the mode separated according to drawing of seeds and composite diagram is grown, and life is determined by drawing of seeds
Long position and the composite diagram for acting on relevant position, are judged whether to grow and then are returned in drawing of seeds and grown by composite diagram,
This avoid the other positions in region to grow impurity information so that the crack information after growing becomes apparent from, is complete
It is whole, so as to reach higher accuracy.
2nd, due to judging that growth is in composite diagram rather than artwork, the region for allowing for traversing graph picture diminish, complexity
It is low, handle the more efficient of image.
Brief description of the drawings
Fig. 1 is the flow chart of the method for the present invention.Wherein, in the case of (a), (b) represent that crack and horizontal direction are obtuse angle
Horizontal fracture and vertical crack;(c), (d) represents that crack is horizontal fracture in the case of acute angle with horizontal direction and split vertically
Seam.
Fig. 2 is original crack pattern piece in embodiments of the invention.
Fig. 3 is the composite diagram after thresholding processing in embodiments of the invention.
Fig. 4 is the drawing of seeds in the embodiment of the present invention.
Fig. 5 is the design sketch after growth.
The present invention is further explained below in conjunction with the drawings and specific embodiments.
Embodiment
The invention provides a kind of region growing method based on composite diagram, the processing step of this method are as follows:
Step 1, gray proces, filtering process and thresholding processing are carried out successively to original fracture image, obtains thresholding
Image after processing, non-crack area is white (pixel 255) after thresholding, and the coordinate of non-crack area is corresponded into original split
Relevant position in gradation of image figure is stitched, and the position pixel is put 255, obtained image is as composite diagram.It is specifically divided into as follows
Sub-step:
Step 11, the target crack of fracture picture first carries out gray processing processing, obtains gray level image;
Step 12, gray level image is divided into multiple size identical block of pixels, each block of pixels is filtered respectively
Processing uses mean filter or median filter method, to remove invalid interference impurity, so as to obtain the image after filtering process;
Step 13, thresholding processing is carried out to the image after filtering process, obtains the figure after thresholding processing;
Step 14, non-crack area (pixel 255) coordinate of the figure after thresholding is corresponded into the gray level image of step 11
In, relevant position coordinate pixel is put 255, obtains composite diagram.
Step 2, Morphological scale-space and filtration treatment are carried out successively to composite diagram, obtains the drawing of seeds in target crack.Specifically
It is divided into following sub-step:
Step 21, to composite diagram carry out Morphological scale-space, the Morphological scale-space specifically successively carry out expansion process and
Corrosion treatment, obtain the image after Morphological scale-space;
Step 22, the crack in the image after Morphological scale-space is filtered;Filtering rule:Filter out boundary rectangle
Length-width ratio is less than 1/3 crack, meanwhile, the average of all connected region areas, filtering in the image after computation of morphology processing
Fall the crack that area is less than the average, obtain the drawing of seeds in target crack.
Step 3, according to the minimum enclosed rectangle of all connected regions in the drawing of seeds in target crack, select to target crack
Drawing of seeds in growth district after grown, the crack pattern picture after being grown.Specifically include following sub-step:
Step 31, target crack is calculated for convenience, is obtained length, width, the positional information in crack first, is extracted target
The minimum enclosed rectangle of all connected regions in the drawing of seeds in crack;
Step 32, by the angle direction (i.e. crack and horizontal direction are acute angle or obtuse angle) in crack, growth district is determined
Position in boundary rectangle, it is specifically:As shown in Fig. 1 (c), (d), if acute angle, then growth district is in boundary rectangle
The top in the upper right corner and the lower section in the lower left corner;Such as Fig. 1 (a), (b), if obtuse angle, then growth district is in the upper left of boundary rectangle
The left side at angle and the right side in the lower right corner;(in Fig. 1, the small frame at rectangle diagonal both ends represents the position of region growing, and lines represent
The locality information in crack;The width of growth district is generally two fracture lengths, and length is 100 pixels);Detected
Figure;
Step 33, in order to ensure the crack of no careless omission, the starting point (extreme point) for the growth district that step 32 is obtained is made
The composite diagram relevant position corresponded to for seed point in step 14, judge whether have not in the range of the n*n centered on seed point
The crack detected, it is specifically:According to seed point coordinates, the connected domain of respective coordinates in composite diagram is found, calculates the connected domain
The average value of pixel of the interior all gray values less than 255;Travel through in the n*n scopes and be not belonging to the every of current minimum enclosed rectangle
Individual pixel, if the absolute difference of some pixel value and average value takes 2~6 not less than constant K, K, then it is assumed that the pixel is not
It is crack information;If less than K, then it is assumed that the pixel is crack information, so as to obtain new crack;Split according to all new
Length, width, the positional information of seam, extract the minimum enclosed rectangle of each connected domain;(for a discrete pixel, no
As crack information;) perform step 32;If not detecting new crack, step 34 is carried out;
Wherein, n determination:Each crack crack adjacent thereto in the drawing of seeds in the target crack that calculation procedure 22 obtains
Distance, (it is to seek the distance between the center of mass point of crack to calculate interstitial distance method here), count all fracture intervals two-by-two from,
Calculate all distance values average value and intermediate value sum 1/2, be the detection figure convolution n.(if n is even number n=n+1;
If n is decimal, using the principle that rounds up.)
Step 34, in order to avoid growing more impurity information, all seed points that step 33 is obtained correspond to step
Whether relevant position in rapid 14 obtained composite diagrams, calculating grow, and result of calculation is if growth, then in the seed in target crack
Grown in figure;Otherwise, do not grow;Crack pattern picture after then being grown.
What whether above-mentioned calculating grew comprises the following steps that:The drawing of seeds in the target crack obtained due to step 22 is by step
Rapid 14 obtained composite diagrams handle what is obtained step by step, so part connected region includes the seed in target crack in composite diagram
The seed point of figure.According to the seed point coordinates of the drawing of seeds in target crack, the connected domain of respective coordinates in composite diagram is found, is calculated
All average values for being less than 255 gray-value pixel points in the connected region, if the pixel in composite diagram with seed point corresponding points
Value and the absolute difference of the average value are not less than constant K (K takes 2~6), then it is assumed that the growth district is not crack area, will
The pixel value that seed point is corresponded in the drawing of seeds in target crack is set to 0, i.e., does not grow;If less than K, then it is assumed that the growth district is
Crack area, the pixel value that the drawing of seeds in target crack is corresponded to seed point put 1, i.e., are grown using the seed point as starting point;
Finally give the crack pattern picture after growth.
Embodiment:
The region growing method based on connected region that the present embodiment provides, comprises the following steps:
Step 1, gray proces, filtering process and thresholding processing are carried out successively to the original fracture image shown in Fig. 2, is obtained
Image to after thresholding processing, non-crack area is white (pixel 255) after thresholding, should by the coordinate pair of non-crack area
The relevant position into original fracture gradation of image figure, and the position pixel is put 255, obtained image is as composite diagram..Specifically
It is divided into following sub-step:
Step 11, the target crack of fracture picture first carries out gray processing processing, obtains gray level image;
Step 12, gray level image is divided into 10 × 10 block of pixels, mean filter processing is carried out to each block of pixels, obtained
Image after to filtering process;
Step 13, thresholding processing is carried out to the image after filtering process, schemed after obtaining thresholding processing.
Step 14, non-crack area (pixel 255) coordinate of the figure after thresholding is corresponded into the gray level image of step 11
In, relevant position coordinate pixel is put 255, obtains composite diagram.(referring to Fig. 3);
Step 2, Morphological scale-space and filtration treatment are carried out successively to composite diagram, obtains the drawing of seeds in target crack.Specifically
It is divided into following sub-step:
Step 21, expansion process and corrosion treatment are carried out successively to the composite diagram after thresholding processing, obtained at morphology
Image after reason;
Step 22, the crack in the image after Morphological scale-space is filtered, filtering rule is to filter out boundary rectangle
Length-width ratio be less than 1/3 crack, meanwhile, calculate the average of all connected region areas, filter out splitting for the following area of average
Seam, obtains the drawing of seeds in target crack.
Step 3, according to the minimum enclosed rectangle of all connected regions in the drawing of seeds in target crack, select to target crack
Drawing of seeds in growth district after grown, the crack pattern picture after being grown.Specifically include following sub-step:
Step 31, the minimum enclosed rectangle of all connected regions in the drawing of seeds in target crack is extracted;
Step 32, by the angle direction (i.e. crack is in acute angle or obtuse angle with horizontal direction) in crack, vitellarium is determined
Domain is in the position of boundary rectangle, so as to find out two orientation for being adapted to growth, i.e. two extreme points (as shown in Figure 1).
Step 33, in order to ensure the crack of no careless omission, the starting point (extreme point) for the growth district that step 32 is obtained is made
The composite diagram relevant position corresponded to for seed point in step 14, judge whether have not in the range of the n*n centered on seed point
The crack detected;According to length, width, the positional information in all new cracks, the minimum external square of each connected domain is extracted
Shape;(for a discrete pixel, not as crack information;) perform step 32;If new split is not detected
Seam, carry out step 34;
Step 34, in order to avoid growing more impurity information, all seed points that step 33 is obtained correspond to step
Whether relevant position in rapid 13 obtained composite diagrams, calculating grow, and result of calculation is if growth, then in the seed in target crack
Grown in figure;Otherwise, do not grow;Crack pattern picture after then being grown.
All extreme point positions are traveled through, it is determined that corresponding to the position in composite diagram, judge whether to grow by connected region,
Then correspond to drawing of seeds relevant position to be grown, so as to obtain the crack pattern picture after growth as shown in Figure 4.
From embodiment as can be seen that the advantage of the invention is that:
1st, the mode separated according to drawing of seeds and composite diagram is grown, and is determined growth position by drawing of seeds and is acted on phase
The composite diagram of position is answered, is judged whether to grow and then is returned in drawing of seeds by composite diagram and grown, so as to avoid in region
Interior other positions grow impurity information so that the crack information after growing becomes apparent from, completely (such as Fig. 4), so as to reach
Higher accuracy is arrived.
2nd, due to judging that growth is in composite diagram rather than artwork, the region for allowing for traversing graph picture diminish, complexity
It is low so as to handle the more efficient of image.
Claims (7)
1. a kind of region growing method based on composite diagram, it is characterised in that the processing step of this method is as follows:
Step 1, gray proces, filtering process and thresholding processing are carried out successively to original fracture image, obtains thresholding processing
Image afterwards, the non-crack area of the image obtained after thresholding processing is white, and the coordinate of non-crack area is corresponded into original
Beginning crack pattern puts 255 as the relevant position in gray-scale map, and by the pixel of relevant position in gray-scale map, and obtained image is as conjunction
Cheng Tu;
Step 2, Morphological scale-space and filtration treatment are carried out successively to composite diagram, obtains the drawing of seeds in target crack;
Step 3, according to the minimum enclosed rectangle of all connected regions in the drawing of seeds in target crack, the kind to target crack is selected
Grown after growth district in subgraph, the crack pattern picture after being grown.
2. the region growing method based on composite diagram as claimed in claim 1, it is characterised in that the step 1 specifically includes
Following sub-step:
Step 11, the target crack of fracture picture first carries out gray processing processing, obtains gray level image;
Step 12, gray level image is divided into multiple size identical block of pixels, processing is filtered respectively to each block of pixels
Using mean filter or median filter method, to remove invalid interference impurity, so as to obtain the image after filtering process;
Step 13, thresholding processing is carried out to the image after filtering process, obtains the figure after thresholding processing;
Step 14, the non-crack area coordinate of the figure after thresholding is corresponded in the gray level image of step 11, by relevant position
Coordinate pixel puts 255, obtains composite diagram.
3. the region growing method based on composite diagram as claimed in claim 1, it is characterised in that the step 2 specifically includes
Following sub-step:
Step 21, Morphological scale-space is carried out to composite diagram, the Morphological scale-space specifically carries out expansion process and corrosion successively
Processing, obtains the image after Morphological scale-space;
Step 22, the crack in the image after Morphological scale-space is filtered;Filtering rule:Filter out the length and width of boundary rectangle
Than the crack less than 1/3, meanwhile, the average of all connected region areas, filters out face in the image after computation of morphology processing
Product obtains the drawing of seeds in target crack less than the crack of the average.
4. the region growing method based on composite diagram as claimed in claim 1, it is characterised in that the step 3 specifically includes
Following sub-step:
Step 31, length, width, the positional information in crack are obtained first, extract all connections in the drawing of seeds in target crack
The minimum enclosed rectangle in region;
Step 32, by the angle direction in crack, determine that growth district is in the position of minimum enclosed rectangle, be specifically:If
Acute angle, then growth district is in the top in the upper right corner of boundary rectangle and the lower section in the lower left corner;If obtuse angle, then growth district exists
The left side in the upper left corner of boundary rectangle and the right side in the lower right corner;Obtain detection figure;
Step 33, in order to ensure the crack of no careless omission, the starting point for the growth district that step 32 is obtained is corresponding as seed point
To the composite diagram relevant position of step 14, judge whether there is undetected crack in the range of the n*n centered on seed point, have
Body is as follows:
According to seed point coordinates, the connected domain of respective coordinates in composite diagram is found, all gray values in the connected domain is calculated and is less than
The average value of 255 pixel;Each pixel that current minimum enclosed rectangle is not belonging in the n*n scopes is traveled through, if some
The absolute difference of pixel value and average value takes 2~6 not less than constant K, K, then it is assumed that the pixel is not crack information;If
Less than K, then it is assumed that the pixel is crack information, so as to obtain new crack;According to the length in all new cracks, width,
Positional information, extract the minimum enclosed rectangle of each connected domain;(for a discrete pixel, split not as new
Stitch information;) perform step 32;If not detecting new crack, step 34 is carried out;
Step 34, all seed points step 33 obtained correspond to relevant position in the composite diagram that step 13 obtains, and calculating is
No growth, result of calculation are then grown if growth in the drawing of seeds in target crack;Otherwise, do not grow;Then given birth to
Crack pattern picture after length.
5. the region growing method based on composite diagram as claimed in claim 4, it is characterised in that the n*n in the step 33
The determination process of scope, wherein n is as follows:
The distance in each crack crack adjacent thereto in the drawing of seeds in the target crack that calculation procedure 22 obtains, statistics are all two-by-two
Fracture interval from, calculate all distance values average value and intermediate value sum 1/2, i.e. n value.
6. the region growing method based on composite diagram as claimed in claim 5, it is characterised in that it is described calculate it is interstitial away from
From refer to calculate crack center of mass point between with a distance from.
7. the region growing method based on composite diagram as claimed in claim 4, it is characterised in that the step 34 specifically includes
Following sub-step:
What whether the calculating grew comprises the following steps that:According to the seed point coordinates of the drawing of seeds in target crack, synthesis is found
The connected domain of respective coordinates in figure, all average values for being less than 255 gray-value pixel points in the connected domain are calculated, if composite diagram
In with the pixel value of seed point corresponding points and the absolute difference of the average value take 2~6 not less than constant K, K, then it is assumed that the life
Long region is not crack area, and the pixel value that seed point is corresponded in the drawing of seeds in target crack is set to 0, i.e., not grown;It is if small
In K, then it is assumed that the growth district is crack area, and the pixel value that the drawing of seeds in target crack is corresponded to seed point puts 1, i.e., with this
Seed point is that starting point carries out region growing.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710547605.6A CN107527354B (en) | 2017-07-06 | 2017-07-06 | A kind of region growing method based on composite diagram |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710547605.6A CN107527354B (en) | 2017-07-06 | 2017-07-06 | A kind of region growing method based on composite diagram |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107527354A true CN107527354A (en) | 2017-12-29 |
CN107527354B CN107527354B (en) | 2019-10-22 |
Family
ID=60748873
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710547605.6A Expired - Fee Related CN107527354B (en) | 2017-07-06 | 2017-07-06 | A kind of region growing method based on composite diagram |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107527354B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110163842A (en) * | 2019-04-15 | 2019-08-23 | 深圳高速工程检测有限公司 | Building cracks detection method, device, computer equipment and storage medium |
CN114332079A (en) * | 2022-03-04 | 2022-04-12 | 宁波昌亚新材料科技股份有限公司 | Plastic lunch box crack detection method, device and medium based on image processing |
CN114418921A (en) * | 2020-10-13 | 2022-04-29 | 南京鑫鼎云科技有限公司 | Industrial image crack detection method |
CN115880248A (en) * | 2022-12-13 | 2023-03-31 | 哈尔滨耐是智能科技有限公司 | Surface scratch defect identification method and visual detection equipment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104504388A (en) * | 2014-12-17 | 2015-04-08 | 长安大学 | Pavement crack identification and feature extraction algorithm and system |
CN104573689A (en) * | 2013-10-11 | 2015-04-29 | 江南大学 | Method for automatically extracting area-of-interest from hyperspectral image of green soybeans based on iteration threshold value |
WO2016172827A1 (en) * | 2015-04-27 | 2016-11-03 | 武汉武大卓越科技有限责任公司 | Stepwise-refinement pavement crack detection method |
-
2017
- 2017-07-06 CN CN201710547605.6A patent/CN107527354B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104573689A (en) * | 2013-10-11 | 2015-04-29 | 江南大学 | Method for automatically extracting area-of-interest from hyperspectral image of green soybeans based on iteration threshold value |
CN104504388A (en) * | 2014-12-17 | 2015-04-08 | 长安大学 | Pavement crack identification and feature extraction algorithm and system |
WO2016172827A1 (en) * | 2015-04-27 | 2016-11-03 | 武汉武大卓越科技有限责任公司 | Stepwise-refinement pavement crack detection method |
Non-Patent Citations (1)
Title |
---|
王聪雅: ""基于图像识别处理的桥梁底面裂缝检测方法的研究"", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110163842A (en) * | 2019-04-15 | 2019-08-23 | 深圳高速工程检测有限公司 | Building cracks detection method, device, computer equipment and storage medium |
CN110163842B (en) * | 2019-04-15 | 2021-06-25 | 深圳高速工程检测有限公司 | Building crack detection method and device, computer equipment and storage medium |
CN114418921A (en) * | 2020-10-13 | 2022-04-29 | 南京鑫鼎云科技有限公司 | Industrial image crack detection method |
CN114332079A (en) * | 2022-03-04 | 2022-04-12 | 宁波昌亚新材料科技股份有限公司 | Plastic lunch box crack detection method, device and medium based on image processing |
CN115880248A (en) * | 2022-12-13 | 2023-03-31 | 哈尔滨耐是智能科技有限公司 | Surface scratch defect identification method and visual detection equipment |
CN115880248B (en) * | 2022-12-13 | 2024-02-09 | 哈尔滨耐是智能科技有限公司 | Surface scratch defect identification method and visual detection equipment |
Also Published As
Publication number | Publication date |
---|---|
CN107527354B (en) | 2019-10-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106846344B (en) | A kind of image segmentation optimal identification method based on the complete degree in edge | |
CN107527354A (en) | A kind of region growing method based on composite diagram | |
CN111753577B (en) | Apple identification and positioning method in automatic picking robot | |
CN103473764B (en) | A kind of remote sensing image object variations detection method | |
CN103426164B (en) | Based on scallop size computing method and the scallop sorting system of Opencv graphical analysis | |
CN103699900B (en) | Building horizontal vector profile automatic batch extracting method in satellite image | |
CN105493141B (en) | Unstructured road border detection | |
CN110782477A (en) | Moving target rapid detection method based on sequence image and computer vision system | |
CN105354847A (en) | Fruit surface defect detection method based on adaptive segmentation of sliding comparison window | |
CN103136752B (en) | Image magnification method based on edge extraction | |
WO2013091369A1 (en) | Multi-target segmentation and tracking method based on depth image | |
CN112395984B (en) | Method for detecting seedling guide line of unmanned agricultural machine | |
CN103577828B (en) | A kind of Approach for road detection based on edge feature | |
CN103177257A (en) | Image identification method and image classification method for coprinus comatus | |
Feng et al. | A separating method of adjacent apples based on machine vision and chain code information | |
CN110111283A (en) | The reminding method and system of infrared suspected target under a kind of complex background | |
CN105719306A (en) | Rapid building extraction method from high-resolution remote sensing image | |
CN109886267A (en) | A kind of soft image conspicuousness detection method based on optimal feature selection | |
CN106157266A (en) | A kind of orchard fruit image acquiring method | |
CN103914848A (en) | Farmland image segmentation method based on wavelet transform and maximum between-cluster variance | |
CN107154044A (en) | A kind of dividing method of Chinese meal food image | |
CN109684941A (en) | One kind picking region partitioning method based on MATLAB image procossing litchi fruits | |
CN103729621B (en) | Plant leaf image automatic recognition method based on leaf skeleton model | |
CN101739667B (en) | Non-downsampling contourlet transformation-based method for enhancing remote sensing image road | |
CN104637066A (en) | Method for extracting binary image quick skeleton based on sequential refining |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20191022 |