CN105976388B - Sandstone partial size automatic testing method based on image procossing - Google Patents
Sandstone partial size automatic testing method based on image procossing Download PDFInfo
- Publication number
- CN105976388B CN105976388B CN201610340394.4A CN201610340394A CN105976388B CN 105976388 B CN105976388 B CN 105976388B CN 201610340394 A CN201610340394 A CN 201610340394A CN 105976388 B CN105976388 B CN 105976388B
- Authority
- CN
- China
- Prior art keywords
- image
- sandstone
- point
- concave point
- region
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of sandstone partial size automatic testing method based on image procossing, comprising the following steps: pass through camera and obtain sandstone image image_1；The structure of gravel and sand in image image_1 is extracted, image image_2 is obtained；Preliminary division is carried out to the sandstone in image image_2, and extracts concave point on each region obtained by division；The matching of concave point pair is carried out to each sandstone region image image_2, the sandstone for dividing adhesion obtain image image_3；In conjunction with image image_2 refinement information and concave point information, the sandstone of segmented image image_3 adhesion obtain image image_4；In conjunction with the profile information and concave point information of image image_2, the sandstone of segmented image image_4 adhesion obtain image image_5；The partial size of sandstone is obtained according to the sandstone image image_5 after segmentation.The present invention can efficiently, reliably detect sandstone partial size automatically, to improve the working efficiency of Practical Project.
Description
Technical field
The invention belongs to computer image processing technology field, especially a kind of sandstone partial size based on image procossing is automatic
Detection method.
Background technique
Due in the engineerings such as road construction, having certain requirement, too big or too small sandstone meeting to the partial size of sandstone
Road quality is caused to decline, therefore in the processs of construction such as road, building, an important step is that the sandstone that detection uses are
It is no to meet the requirements.
But existing detection technique is then measured to the sandstone sampled by being sampled to sandstone, from
And obtain the partial size of sandstone.Obviously, this method is very complicated and time-consuming, is unfavorable for improving working efficiency.Current
Shortcoming is compared in technical research in this regard, none can solve this problem in the method for high-efficient automatic.
Summary of the invention
It is an object of the invention to be to provide a kind of efficient, reliable sandstone partial size automatic testing method, it is based on computer
Image processing techniques is to improve the working efficiencies of the engineerings such as actual road construction.
The technical solution for realizing the aim of the invention is as follows: a kind of sandstone partial size based on the image procossing side of detection automatically
Method, comprising the following steps:
Step 1, sandstone image image_1 is obtained by camera；
Step 2, the structure of gravel and sand in image image_1 is extracted, image image_2 is obtained；
Step 3, Preliminary division is carried out to the sandstone in image image_2, and is extracted on each region obtained by division recessed
Point；
Step 4, the matching of concave point pair is carried out to each sandstone region image image_2, the sandstone for dividing adhesion obtain figure
As image_3；
Step 5, in conjunction with image image_2 refinement information and concave point information, the sandstone of segmented image image_3 adhesion are obtained
Image image_4；
Step 6, in conjunction with the profile information of image image_2 and concave point information, the sandstone of segmented image image_4 adhesion are obtained
To image image_5；
Step 7, the partial size of sandstone is obtained according to the sandstone image image_5 after segmentation.
Further, sandstone image image_1 is obtained by camera described in step 1, specifically:
11) original sandstone image is obtained by camera；
12) multiple images region is intercepted from original sandstone image；
13) sandstone image image_1 is obtained to the image-region progress gaussian filtering of step 12) acquisition.
Further, the structure of gravel and sand in extraction image image_1 described in step 2, obtains image image_2, specifically:
21) according to a preliminary estimate in image image_1 the structure of gravel and sand partial size l；
22) using the l estimated in step 21) as parameter, the morphology for extracting image image_1 builds index, that is, MBI
Characteristic pattern；
23) binaryzation is carried out to the MBI characteristic pattern obtained in step 22), remove the hole in image and carries out morphology
Opening operation, the image that obtains that treated, the structural element se1 that wherein opening operation uses is disc；
24) sandstone less than set threshold value th_1 are filtered out from treated image, obtain image image_2.
Further, Preliminary division is carried out to the sandstone in image image_2 described in step 3, and each obtained by division
Concave point is extracted on region, specific as follows:
31) according to the connectivity of region in image image_2, Preliminary division is carried out to sandstone region in image image_2；
32) concave point is extracted from the edge in each sandstone region, wherein the extracting method of concave point are as follows: traverse sandstone clockwise
Point in edges of regions, forerunner's point of current point P, P are M, subsequent point is N, if point M, P, N composition angle less than 120 °,
Point P is then regarded as concave point；Point intermediate in this concave point group is represented concave point group by the concave point group continuously distributed for concave point；
33) filtration step 32) obtain concave point, filtering rule are as follows:, will be apart from small by clock-wise order in edges of regions
One of concave point is removed in two concave points of set threshold value th_2, obtains concave point collection Pits.
Further, the matching of concave point pair is carried out described in step 4 to each sandstone region image image_2, divides adhesion
Sandstone obtain image image_3, specifically:
41) concave point on sandstone region in image image_2 is matched, matching rule are as follows: two concave point P1,
P2, forerunner's point are respectively M1, M2, and subsequent point is respectively N1, N2, if the line of P1, P2 are also cross angle M1P1N1 and angle
Concave point P1, P2 are then considered as successful match by M2P2N2 interior zone；If a concave point has matched a number of other concave points, choose
Two concave points are apart from the smallest concave point pair；
42) for the concave point pair of successful match, by the concave point to the cut-off rule for being connected to the segmentation sandstone region, general
It is removed with successful concave point to from concave point collection pits；
43) all concave points in image image_2 in all sandstone regions are traversed, step 41)~42 are repeated), to image
Image_2 is split；
44) all sandstone regions are traversed, if the sandstone region divided is less than set threshold value th_3, by this point
Secant removal, finally obtains and utilizes the matched sandstone segmentation result image image_3 of concave point.
Further, image image_2 refinement information and concave point information are combined described in step 5, segmented image image_3 is viscous
Sandstone even obtain image image_4, specific as follows:
51) image image_2 is refined using ZS parallel thinning algorithm, the image I_thing after being refined, figure
As each thinning lines both correspond to a sandstone region in image_2 in I_thing；
52) to the thinning lines in the edge and corresponding I_thing in sandstone region, the point in thinning lines is traversed, if refinement
Exist on line point C with concave point P at a distance from less than set threshold value th_4 when, then concave point P is as the foundation divided；
53) concave point P is obtained from step 52), the point in the thinning lines that same sandstone region is belonged to around concave point P is recorded
Get off, forms point set set1；
54) regression equation of thinning lines near concave point P is obtained according to point set set1, the specific method is as follows:
Regression equation isIts parameter can be acquired by following formula:
Wherein, (xi,yi) it is point in point set set, indicate the coordinate in image；N is the size of point set set；
55) it crosses concave point P and makees the cut-off rule that slope is k1, sandstone region is split, in which:
Concave point P is removed from concave point collection pits；
56) all sandstone regions in image image_2 are traversed, step 52)~55 are repeated), image image_3 is divided
It cuts；
57) all sandstone regions are traversed, if the sandstone region divided is less than set threshold value th_3, by this point
Secant removal, finally obtains the sandstone segmentation result image image_4 using image thinning information.
Further, image image_1, the profile information of image image_2 and concave point information are combined described in step 6, point
The sandstone for cutting image image_4 adhesion obtain image image_5, specifically:
61) edge edge extracting method is used, and extracts contour images from image image_1 using canny operator
Then BW1 and image image_2 is carried out logic and operation by BW1, obtain the contour images BW2 in sandstone region；
62) to the concave point P in sandstone region, look for whether that there are the contour lines in same sandstone region around P, if depositing
Then using concave point P as the foundation of segmentation；
63) contour line around concave point P and concave point P has been obtained from step 62), establishes the set s put on contour line
Et2 obtains the regression equation of contour lineSpecific method is identical as step 54)；
64) it crosses concave point P and makees the cut-off rule that slope is k2, sandstone region is split, in which:
Concave point P is removed from concave point collection pits；
65) all sandstone regions in image image_2 are traversed, step 62)~64 are repeated), image image_4 is divided
It cuts；
66) all sandstone regions are traversed, if the sandstone region divided is less than set threshold value th_3, by this point
Secant removal, finally obtains the sandstone segmentation result image image_5 using image outline information.
Further, the partial size of sandstone is obtained described in step 7 according to the sandstone image image_5 after segmentation, specific as follows:
71) for the sandstone region in image image_5, the point on region contour is traversed, finds two apart from maximum
Point, using the distance of the two points as the partial size in the sandstone region；
72) each sandstone region in image image_5 is traversed, step 71) is repeated, obtains the grain in each sandstone region
Diameter forms the set len_set of partial size；
73) the maximum value len_max of partial size, the minimum value len_min of partial size, the average value len_avg of partial size are asked, specifically
Calculation method it is as follows:
Wherein, n is the size of len_set, and len_set (i) is i-th of element in len_set.
Compared with prior art, the present invention its remarkable advantage is: (1) advantage of computer high speed processing is utilized, in conjunction with
Image processing techniques, realizes sandstone partial size and detects automatically；(2) efficiently, reliably sandstone partial size can be examined automatically
It surveys, improves the working efficiency of the engineerings such as actual road construction.
Detailed description of the invention
Fig. 1 is that the present invention is based on the flow charts of the sandstone partial size automatic testing method of image procossing.
Fig. 2 is the matched schematic diagram of concave point in the present invention.
Fig. 3 is the sample graph after extracting.
Fig. 4 is the result figure after extracting bright structures in image.
Fig. 5 is the result figure for extracting concave point.
Fig. 6 is the result figure being split using the matching of concave point pair.
Fig. 7 is the result figure divided in conjunction with refinement information and concave point information.
Fig. 8 is the result figure divided in conjunction with profile information and concave point information.
Fig. 9 is the final segmentation result and partial size schematic diagram of sandstone.
Specific embodiment
The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
In conjunction with Fig. 1, the present invention is based on the sandstone partial size automatic testing methods of image procossing, comprising the following steps:
Step 1, sandstone image image_1 is obtained by camera；
Step 2, the structure of gravel and sand in image image_1 is extracted, image image_2 is obtained；
Step 3, Preliminary division is carried out to the sandstone in image image_2, and is extracted on each region obtained by division recessed
Point；
Step 4, the matching of concave point pair is carried out to each sandstone region image image_2, the sandstone for dividing adhesion obtain figure
As image_3；
Step 5, in conjunction with image image_2 refinement information and concave point information, the sandstone of segmented image image_3 adhesion are obtained
Image image_4；
Step 6, in conjunction with the profile information of image image_2 and concave point information, the sandstone of segmented image image_4 adhesion are obtained
To image image_5；
Step 7, the partial size of sandstone is obtained according to the sandstone image image_5 after segmentation.
In the sandstone partial size automatic testing method based on image procossing, sand is obtained by camera described in step 1
Stone image image_1, specifically:
11) original sandstone image is obtained by camera；
12) multiple images region is intercepted from original sandstone image；
13) sandstone image image_1 is obtained to the image-region progress gaussian filtering of step 12) acquisition.
In the sandstone partial size automatic testing method based on image procossing, in extraction image image_1 described in step 2
The structure of gravel and sand, obtain image image_2, specifically:
21) according to a preliminary estimate in image image_1 the structure of gravel and sand partial size l, wherein used in estimation method be make respectively
With m item is horizontal at equal intervals and n item vertical straight line is intersected with sandstone in image at equal intervals, the line segment in the structure of gravel and sand,
It is then the average value of all line segments to the estimation l of the sandstone partial size；
22) using the l estimated in step 21) as parameter, the morphology for extracting image image_1 builds index, that is, MBI
Characteristic pattern, wherein the basic thought of MBI is the pass by establishing between interior of building feature (such as brightness, size and contrast)
System and morphology operations accord with the relationship between attribute (such as reconstruction, granularity and direction) to extract structure more bright in image；
23) binaryzation is carried out to the MBI characteristic pattern obtained in step 22), remove the hole in image and carries out morphology
Opening operation, the image that obtains that treated, the structural element se1 that wherein opening operation uses is disc；
24) sandstone less than set threshold value th_1 are filtered out from treated image, obtain image image_2, wherein t
The size of h_1 and the requirement of user are related, can rule of thumb or expertise is determined.
In the sandstone partial size automatic testing method based on image procossing, in image image_2 described in step 3
Sandstone carry out Preliminary division, and extract concave point on each region obtained by division, specific as follows:
31) according to the connectivity of region in image image_2, Preliminary division is carried out to sandstone region in image image_2；
32) concave point is extracted from the edge in each sandstone region, wherein the extracting method of concave point are as follows: traverse sandstone clockwise
Point in edges of regions, forerunner's point of current point P, P are M, subsequent point is N, if point M, P, N composition angle less than 120 °,
Point P is then regarded as concave point；Point intermediate in this concave point group is represented concave point group by the concave point group continuously distributed for concave point；
33) filtration step 32) obtain concave point, filtering rule are as follows:, will be apart from small by clock-wise order in edges of regions
One of concave point is removed in two concave points of set threshold value th_2, obtains concave point collection Pits, wherein the size and use of th_2
The requirement at family is related, can rule of thumb or expertise is determined.
It is each to image image_2 described in step 4 in the sandstone partial size automatic testing method based on image procossing
Sandstone region carries out the matching of concave point pair, and the sandstone for dividing adhesion obtain image image_3, specifically:
41) Fig. 2 is combined, the concave point on sandstone region in image image_2 is matched, matching rule are as follows: two
Concave point P1, P2, forerunner's point are respectively M1, M2, and subsequent point is respectively N1, N2, if the line of P1, P2 are also cross angle M1P1N1
With angle M2P2N2 interior zone, then concave point P1, P2 are considered as successful match；If a concave point has matched a number of other concave points,
Two concave points are chosen apart from the smallest concave point pair；
42) for the concave point pair of successful match, by the concave point to the cut-off rule for being connected to the segmentation sandstone region, general
It is removed with successful concave point to from concave point collection pits；
43) all concave points in image image_2 in all sandstone regions are traversed, step 41)~42 are repeated), to image
Image_2 is split；
44) all sandstone regions are traversed, if the sandstone region divided is less than set threshold value th_3, by this point
Secant removal, finally obtains and utilizes the matched sandstone segmentation result image image_3 of concave point.The wherein size of th_3 and user
It, can rule of thumb or expertise is determined it is required that related.
In the sandstone partial size automatic testing method based on image procossing, combine image image_2 thin described in step 5
Change information and concave point information, the sandstone of segmented image image_3 adhesion obtain image image_4, specific as follows:
51) image image_2 is refined using ZS (ZHANG and SUEN) parallel thinning algorithm, after obtaining refinement
Image I_thing, each thinning lines both correspond to a sandstone region in image_2 in image I_thing, wherein ZS
Algorithm is a classical thinning algorithm, the basic principle is that carrying out skeletal extraction to a secondary bianry image, is deleted unwanted
Profile point only retains its skeletal point；
52) to the thinning lines in the edge and corresponding I_thing in sandstone region, the point in thinning lines is traversed, if refinement
Exist on line point C with concave point P at a distance from less than set threshold value th_4 when, then concave point P is as the foundation divided, wherein th_4
Size it is related with the requirement of user, can rule of thumb or expertise is determined；
53) concave point P is obtained from step 52), the point in the thinning lines that same sandstone region is belonged to around concave point P is recorded
Get off, forms point set set1；
54) regression equation of thinning lines near concave point P is obtained according to point set set1, the specific method is as follows:
Regression equation isIts parameter can be acquired by following formula:
Wherein, (xi,yi) it is point in point set set, indicate the coordinate in image；N is the size of point set set；
55) it crosses concave point P and makees the cut-off rule that slope is k1, sandstone region is split, in which:
Concave point P is removed from concave point collection pits；
56) all sandstone regions in image image_2 are traversed, step 52)~55 are repeated), image image_3 is divided
It cuts；
57) all sandstone regions are traversed, if the sandstone region divided is less than set threshold value th_3, by this point
Secant removal, finally obtains the sandstone segmentation result image image_4 using image thinning information.The wherein size and step of th_3
The rapid value 44) set is equal.
In the sandstone partial size automatic testing method based on image procossing, described in step 6 combine image image_1,
The profile information and concave point information of image image_2, the sandstone of segmented image image_4 adhesion obtain image image_5, specifically
Are as follows:
61) edge edge extracting method is used, and extracts contour images from image image_1 using canny operator
Then BW1 and image image_2 is carried out logic and operation by BW1, obtain the contour images BW2 in sandstone region；
62) to the concave point P in sandstone region, look for whether that there are the contour lines in same sandstone region around P, if depositing
Then using concave point P as the foundation of segmentation；
63) contour line around concave point P and concave point P has been obtained from step 62), establishes the set s put on contour line
Et2 obtains the regression equation of contour lineSpecific method is identical as step 54)；
64) it crosses concave point P and makees the cut-off rule that slope is k2, sandstone region is split, in which:
Concave point P is removed from concave point collection pits；
65) all sandstone regions in image image_2 are traversed, step 62)~64 are repeated), image image_4 is divided
It cuts；
66) all sandstone regions are traversed, if the sandstone region divided is less than set threshold value th_3, by this point
Secant removal, finally obtains the sandstone segmentation result image image_5 using image outline information.The wherein size and step of th_3
The rapid value 44) set is equal.
In the sandstone partial size automatic testing method based on image procossing, according to the sandstone after segmentation described in step 7
Image image_5 obtains the partial size of sandstone, specific as follows:
71) for the sandstone region in image image_5, the point on region contour is traversed, finds two apart from maximum
Point, using the distance of the two points as the partial size in the sandstone region；
72) each sandstone region in image image_5 is traversed, step 71) is repeated, obtains the grain in each sandstone region
Diameter forms the set len_set of partial size；
73) the maximum value len_max of partial size, the minimum value len_min of partial size, the average value len_avg of partial size are asked, specifically
Calculation method it is as follows:
Wherein, n is the size of len_set, and len_set (i) is i-th of element in len_set.
Embodiment 1
Illustrate practicability of the invention followed by a specific implementation example:
The present invention uses the sandstone partial size automatic testing method based on Computer Image Processing, and it is to use that this, which implements example,
Industry is paved the way used sandstone, and the partial size of sandstone is extracted by using method proposed by the invention, the specific steps are as follows:
Step 1) obtains sandstone image information by camera.Example has used a camera, in practical applications
One group of multiple camera can be used to obtain more sandstone images.Then several images are intercepted in the original image of shooting
As sampling, Fig. 3 is the sampled images after extracting in region.
Step 2), extracts more bright sandstone from image, and used method is to utilize different size different directions
Linear structure operation is reconstructed to image.Exemplary method uses MBI algorithm to extract bright structures, and Fig. 4 is to extract
Result afterwards.
Step 3), extracts concave point from image, and method used in example is that the every bit on sandstone profile is found
Its forerunner's point and subsequent point, obtain the angle that forerunner's point, current point, subsequent point are constituted, and the point for angle less than 120 degree is regarded
For concave point, Fig. 5 is the result of concave point after extracting.Wherein circle is concave point, it can be seen that most of concave points are all to be located at two sand
The position of stone overlapping.
Step 4) carries out the matching of concave point pair, divides the sandstone of adhesion, and matched rule is the line point of two concave points
Not between the forerunner's point and subsequent point of two concave points, Fig. 2 is the matched schematic diagram of concave point, then for the two of successful match
A concave point connects to divide adhesion sandstone, and Fig. 6 is to be split using the matching of concave point pair as a result, wherein relatively thick
Line be current cut-off rule.
Step 5) divides the sandstone of adhesion in conjunction with image thinning information and concave point information, Fig. 7 is after segmentation as a result, its
In thicker line be current cut-off rule, thinner line be before cut-off rule.
Step 6) divides the sandstone of adhesion in conjunction with image outline information and concave point information, Fig. 8 is after segmentation as a result, its
In thicker line be current cut-off rule, thinner line be before cut-off rule.
Step 7), the partial size of sandstone is obtained to the sandstone image after segmentation, and specific method is to each sandstone region
Wherein partial size of the longest line segment as the sandstone is sought, Fig. 9 is that the final segmentation result of sandstone and the partial size of each sandstone show
It is intended to.In this example, the final result of the partial size of sandstone is as shown in table 1.
The size information of sandstone in table 1, image
| Maximum particle diameter | Minimum grain size | Average grain diameter | |
| Length (pixel) | 138.2208 | 31.7648 | 65.1812 |
In conclusion it is proposed that the sandstone partial size extraction method based on Computer Image Processing can very well by
The sandstone of adhesion are split in image, this means that the size information that can extract sandstone using image procossing very well.
Claims (7)
1. a kind of sandstone partial size automatic testing method based on image procossing, which comprises the following steps:
Step 1, sandstone image image_1 is obtained by camera；
Step 2, the structure of gravel and sand in image image_1 is extracted, image image_2 is obtained；
Step 3, Preliminary division is carried out to the sandstone in image image_2, and extracts concave point on each region obtained by division；
Step 4, the matching of concave point pair is carried out to each sandstone region image image_2, the sandstone for dividing adhesion obtain image
image_3；
Step 5, in conjunction with image image_2 refinement information and concave point information, the sandstone of segmented image image_3 adhesion obtain image
image_4；
Step 6, in conjunction with the profile information of image image_2 and concave point information, the sandstone of segmented image image_4 adhesion obtain figure
As image_5；
Step 7, the partial size of sandstone is obtained according to the sandstone image image_5 after segmentation；
Image image_2 refinement information and concave point information are combined described in step 5, the sandstone of segmented image image_3 adhesion obtain
Image image_4, specific as follows:
51) image image_2 is refined using ZS parallel thinning algorithm, the image I_thing after being refined, image I_
Each thinning lines both correspond to a sandstone region in image_2 in thing；
52) to the thinning lines in the edge and corresponding I_thing in sandstone region, the point in thinning lines is traversed, if in thinning lines
When being less than threshold value th_4 at a distance from concave point P in the presence of point C, then concave point P is as the foundation divided；
53) concave point P is obtained from step 52), the point in the thinning lines that same sandstone region is belonged to around concave point P is recorded
Come, forms point set set1；
54) regression equation of thinning lines near concave point P is obtained according to point set set1, the specific method is as follows:
Regression equation isIts parameter can be acquired by following formula:
Wherein, (xi,yi) it is point in point set set1, indicate the coordinate in image；N is the size of point set set1；
55) it crosses concave point P and makees the cut-off rule that slope is k1, sandstone region is split, in which:
Concave point P is removed from concave point collection pits；
56) all sandstone regions in image image_2 are traversed, step 52)~55 are repeated), image image_3 is split；
57) all sandstone regions are traversed, if the sandstone region divided is less than threshold value th_3, which is removed,
Finally obtain the sandstone segmentation result image image_4 using image thinning information.
2. the sandstone partial size automatic testing method according to claim 1 based on image procossing, which is characterized in that step 1
It is described that sandstone image image_1 is obtained by camera, specifically:
11) original sandstone image is obtained by camera；
12) multiple images region is intercepted from original sandstone image；
13) sandstone image image_1 is obtained to the image-region progress gaussian filtering of step 12) acquisition.
3. the sandstone partial size automatic testing method according to claim 1 based on image procossing, which is characterized in that step 2
The structure of gravel and sand extracted in image image_1, obtains image image_2, specifically:
21) according to a preliminary estimate in image image_1 the structure of gravel and sand partial size l；
22) using the l estimated in step 21) as parameter, the morphology for extracting image image_1 builds index, that is, MBI feature
Figure；
23) binaryzation is carried out to the MBI characteristic pattern obtained in step 22), removes the hole in image and carry out morphology and opens fortune
It calculates, the image that obtains that treated, the structural element se1 that wherein opening operation uses is disc；
24) sandstone less than threshold value th_1 are filtered out from treated image, obtain image image_2.
4. the sandstone partial size automatic testing method according to claim 1 based on image procossing, which is characterized in that step 3
The sandstone in image image_2 carry out Preliminary division, and extract concave point on each region obtained by division, specifically such as
Under:
31) according to the connectivity of region in image image_2, Preliminary division is carried out to sandstone region in image image_2；
32) concave point is extracted from the edge in each sandstone region, wherein the extracting method of concave point are as follows: traversal sandstone region clockwise
Point on edge, forerunner's point of current point P, P are M, subsequent point is N, if the angle of point M, P, N composition is incited somebody to action less than 120 °
Point P is regarded as concave point；Point intermediate in this concave point group is represented concave point group by the concave point group continuously distributed for concave point；
33) filtration step 32) concave point that obtains, filtering rule are as follows: in edges of regions, clock-wise order is pressed, it will be apart from less than threshold
One of concave point is removed in two concave points of value th_2, obtains concave point collection Pits.
5. the sandstone partial size automatic testing method according to claim 1 based on image procossing, which is characterized in that step 4
The matching that concave point pair is carried out to each sandstone region image image_2, the sandstone for dividing adhesion obtain image image_3,
Specifically:
41) concave point on sandstone region in image image_2 is matched, matching rule are as follows: two concave points P1, P2,
Forerunner's point is respectively M1, M2, and subsequent point is respectively N1, N2, if the line of P1, P2 are also cross in angle M1P1N1 and angle M2P2N2
Concave point P1, P2 are then considered as successful match by portion region；If a concave point has matched a number of other concave points, choose two concave points away from
From the smallest concave point pair；
42) for the concave point pair of successful match, by the concave point to being connected to the cut-off rule for dividing the sandstone region, will matching at
The concave point of function is removed to from concave point collection pits；
43) all concave points in image image_2 in all sandstone regions are traversed, step 41)~42 are repeated), to image
Image_2 is split；
44) all sandstone regions are traversed, if the sandstone region divided is less than threshold value th_3, which is removed,
It finally obtains and utilizes the matched sandstone segmentation result image image_3 of concave point.
6. the sandstone partial size automatic testing method according to claim 1 based on image procossing, which is characterized in that step 6
The profile information and concave point information of the combination image image_2, the sandstone of segmented image image_4 adhesion obtain image
Image_5, specifically:
61) edge extracting method is used, and extracts contour images BW1 from image image_1 using canny operator, then will
BW1 and image image_2 carries out logic and operation, obtains the contour images BW2 in sandstone region；
62) to the concave point P in sandstone region, look for whether that there are the contour lines in same sandstone region around P, and if it exists,
Then using concave point P as the foundation of segmentation；
63) contour line around concave point P and concave point P has been obtained from step 62), is established the set set2 put on contour line, is obtained
Obtain the regression equation of contour lineSpecific method is identical as step 54)；
64) it crosses concave point P and makees the cut-off rule that slope is k2, sandstone region is split, in which:
Concave point P is removed from concave point collection pits；
65) all sandstone regions in image image_2 are traversed, step 62)~64 are repeated), image image_4 is split；
66) all sandstone regions are traversed, if the sandstone region divided is less than threshold value th_3, which is removed,
Finally obtain the sandstone segmentation result image image_5 using image outline information.
7. the sandstone partial size automatic testing method according to claim 1 based on image procossing, which is characterized in that step 7
The sandstone image image_5 according to after segmentation obtains the partial size of sandstone, specific as follows:
71) for the sandstone region in image image_5, the point on region contour is traversed, two is found apart from maximum point, incites somebody to action
Partial size of the distance of the two points as the sandstone region；
72) each sandstone region in image image_5 is traversed, step 71) is repeated, obtains the partial size in each sandstone region, shape
At the set len_set of partial size；
73) the maximum value len_max of partial size, the minimum value len_min of partial size, the average value len_avg of partial size are asked, it is specific to count
Calculation method is as follows:
Wherein, n is the size of len_set, and len_set (i) is i-th of element in len_set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610340394.4A CN105976388B (en) | 2016-05-20 | 2016-05-20 | Sandstone partial size automatic testing method based on image procossing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610340394.4A CN105976388B (en) | 2016-05-20 | 2016-05-20 | Sandstone partial size automatic testing method based on image procossing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105976388A CN105976388A (en) | 2016-09-28 |
| CN105976388B true CN105976388B (en) | 2019-04-12 |
Family
ID=56956169
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610340394.4A Active CN105976388B (en) | 2016-05-20 | 2016-05-20 | Sandstone partial size automatic testing method based on image procossing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105976388B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101403743A (en) * | 2008-10-31 | 2009-04-08 | 广东威创视讯科技股份有限公司 | Automatic separating method for X type overlapping and adhering chromosome |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014049873A1 (en) * | 2012-09-28 | 2014-04-03 | 富士機械製造株式会社 | Device for correcting image processing data, and method for correcting image processing data |
-
2016
- 2016-05-20 CN CN201610340394.4A patent/CN105976388B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101403743A (en) * | 2008-10-31 | 2009-04-08 | 广东威创视讯科技股份有限公司 | Automatic separating method for X type overlapping and adhering chromosome |
Non-Patent Citations (1)
| Title |
|---|
| 基于显微图像的PTA粒径分布估计方法研究;韦冬冬;《中国优秀硕士学位论文全文数据库》;20100815;摘要,正文第9-10,16-23,55-56,62-64,67页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105976388A (en) | 2016-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103839268B (en) | Method for detecting fissure on surface of subway tunnel | |
| CN110399875A (en) | A kind of form of general use information extracting method based on deep learning and pixel projection | |
| WO2016172827A1 (en) | Stepwise-refinement pavement crack detection method | |
| CN105335743A (en) | Vehicle license plate recognition method | |
| Li et al. | 3D laser imaging and sparse points grouping for pavement crack detection | |
| CN103528534B (en) | A kind of electric power line ice-covering thickness detection method based on image monitoring | |
| CN103336946A (en) | Binocular stereoscopic vision based clustered tomato identification method | |
| CN106485708B (en) | A kind of round log method of counting based on image recognition | |
| CN113962997B (en) | Strip steel edge crack defect detection method and system based on image processing | |
| CN108460780A (en) | A kind of adhesion grain of rice image partition method based on background framework characteristic | |
| CN114399522A (en) | High-low threshold-based Canny operator edge detection method | |
| CN106446905A (en) | Surface crack texture extraction method based on fusion of seepage algorithm and adaptive Canny algorithm | |
| CN103295013A (en) | Pared area based single-image shadow detection method | |
| CN105718916A (en) | Lane line detection method based on Hough transform | |
| CN105719275A (en) | Parallel combination image defect segmentation method | |
| Herumurti et al. | Urban road extraction based on hough transform and region growing | |
| CN106778754A (en) | A kind of industrial ammeter digit recognition method of robust | |
| CN103279762A (en) | Judging method for common growth form of fruit under natural environment | |
| CN105869174A (en) | Sky scene image segmentation method | |
| CN108020554A (en) | A kind of steel strip surface defect recognition detection method | |
| CN111598897A (en) | Infrared image segmentation method based on Otsu and improved Bernsen | |
| CN103400113A (en) | Method for detecting pedestrian on expressway or in tunnel based on image processing | |
| CN102073872A (en) | Image-based method for identifying shape of parasite egg | |
| CN103971367A (en) | Hydrologic data image segmenting method | |
| CN105976388B (en) | Sandstone partial size automatic testing method based on image procossing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |