CN108062770A - The accurate positioning method at micropore center in a kind of microwell plate picture taken pictures naturally - Google Patents
The accurate positioning method at micropore center in a kind of microwell plate picture taken pictures naturally Download PDFInfo
- Publication number
- CN108062770A CN108062770A CN201711007652.8A CN201711007652A CN108062770A CN 108062770 A CN108062770 A CN 108062770A CN 201711007652 A CN201711007652 A CN 201711007652A CN 108062770 A CN108062770 A CN 108062770A
- Authority
- CN
- China
- Prior art keywords
- microwell plate
- center
- image
- circle
- hole
- 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
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000012360 testing method Methods 0.000 claims description 32
- 239000000284 extract Substances 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 238000012937 correction Methods 0.000 claims description 4
- 238000013459 approach Methods 0.000 claims description 3
- 239000000945 filler Substances 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 2
- 238000001514 detection method Methods 0.000 description 16
- 230000006870 function Effects 0.000 description 7
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 239000003153 chemical reaction reagent Substances 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000004807 localization Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- DEYFWGXTPWNADC-UHFFFAOYSA-N 6-[2-phenylethyl(propyl)amino]-5,6,7,8-tetrahydronaphthalen-1-ol Chemical compound C1CC2=C(O)C=CC=C2CC1N(CCC)CCC1=CC=CC=C1 DEYFWGXTPWNADC-UHFFFAOYSA-N 0.000 description 1
- 241001212149 Cathetus Species 0.000 description 1
- 241001465754 Metazoa Species 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000007705 chemical test Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000009432 framing Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000002203 pretreatment Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000009466 transformation 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
- G06T7/60—Analysis of geometric attributes
- G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
-
- 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
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to a kind of accurate positioning methods at micropore center in microwell plate picture taken pictures naturally, by how to be accurately positioned each hole center of microwell plate, the square boundary for how positioning microwell plate and the hole center that microwell plate how is accurately positioned according to the square boundary having good positioning are decomposed into.The present invention positions square boundary using image and fitting a straight line, not only efficiently precisely, also the hole centralized positioning of microwell plate has been served vital, the hole center to position microwell plate provides the guarantee of accuracy.And it is greatly improved for the work efficiency of hospital and Lab workers.
Description
Technical field
The present invention relates to micropores in the technical field of framing more particularly to a kind of microwell plate picture taken pictures naturally
The accurate positioning method at center.
Background technology
Microwell plate is a kind of multiduty detection breadboard, in sides such as clinical trial, chemical test, the reagent detections of animal doctor
Face is widely used.
And with the development of artificial intelligence, the staff of clinical trial to the experiment demand of intelligent automaticization increasingly
It is high.Wherein, realize that microwell plate Weigh sensor can accelerate the manual intelligent of clinical trial, liberate experimental work personnel significantly
Labour, promote society development.Further, the hole center of circle for being accurately positioned microwell plate can be with the reality in rapid extraction circular hole
Reagent is tested, by empirical numerical, convenient for the study of machine and the intelligent recognition of machine, makes the intelligent automatic chemical conversion of clinical trial
For possibility.
As it can be seen that how to be accurately positioned microwell plate micropore center is particularly important.
Known detection localization method has:
The Quick rectangle detection method of the big order of magnitude image of high-resolution of Patent No. 201110223911.7:It is realized
Step includes pretreatment and obtains cross point set, improve PPHT detection of straight lines, along linear search and delete all boundary points on straight line
Cross point set, searching five steps of accurate matched parallel lines and definite rectangle extreme coordinates etc. where pixel.Wherein in the first step
Pretreatment in realize the order of magnitude for reducing coordinate, the pixel value of the n of each boundary point four adjacent pixel coordinate pixels is put
It is the first numerical value to obtain a cross point set, so as to solve the detection under the big order of magnitude image of high-resolution.As for second step,
Using the peak value of progressive probability Hough transformation algorithm detection Hough, each straight line corresponding with boundary point pixel set is obtained.The
Three steps are then that Xun Huan finds the non-zero pixels points that inquiry includes on this straight line, get that maximum straight line.Then,
The Euclidean length and slope for the straight line that the 3rd step obtains are calculated, a pair of parallel and equal length straight line is found, reasonably recognizes
It is the border of rectangle for this.And look for whether a pair of parallel line for having from the Relative vertical of endpoint, composition one is complete
Whole rectangle.Although the method can solve the detection problem of high-resolution big data grade image;Accuracy is high, and suitable for unknown square
Shape size and direction;But it needs to calculate in advance in pretreatment and obtains cross point set, less efficient;Detection method needs accurate
A pair of of Euclidean length a pair of parallel line identical with slope is found, for the rectangle restricted application system in real image, and
It is computationally intensive.
The rectangular target object detecting method of Patent No. 201610514128.9:It is special that implementation step includes extraction sample
Sign obtains root wave filter, obtaining widget wave filter, model inspection and carries out gradient search calculating.It relies on the side of machine learning
Method as negative sample by the use of the picture for not containing object, it is necessary to form training set in advance.Extract sample characteristics:Respectively containing
With do not contain the picture of rectangular target object as positive sample and negative sample, and be divided into multiple cell factories, obtain every piece
Histograms of oriented gradients.And second step is then scanned samples picture, and all pieces of histograms of oriented gradients is extracted and tied
It closes, forms final feature vector and be input in support vector machines, obtain root wave filter.3rd step, according to the position of root wave filter
The position for finding component wave filter is put, optimal location is then filtered out by the height of score, by the feature of component wave filter
It is input to training in hidden variable support vector machines.Be finally based on search edge gradient computational algorithm judge detection block whether
Outside objective contour edge, the position of detection block is then adjusted.Although this method detection accurate rate is higher, it needs big spirogram
Piece is trained in advance as sample, and the various steps used in method are more complicated compared with other detection methods.
In addition to above two method, there are the image processing apparatus and hough transform of Patent No. 201210439013.X
Method, the image processing apparatus of Patent No. 201210439005.5 and line detection method, Patent No.
201610944459.6 the tool wear monitoring method based on minimum enclosed rectangle etc..But these detection methods are not achieved pre-
The effect thought.
The content of the invention
It is an object of the invention to overcome the deficiencies of the prior art and provide a kind of registration rate height, detection efficiency are high
Naturally the accurate positioning method at micropore center in the microwell plate picture taken pictures.
To achieve the above object, technical solution provided by the present invention is:First position the square boundary of microwell plate, Ran Hougen
According to the hole center for the square boundary positioning microwell plate having good positioning.
Wherein, the square boundary for positioning microwell plate comprises the following steps:
S1, the image that microwell plate is shot under natural lighting;
S2, gray level image, Gassian low-pass filter and binary image processing, the feature of prominent image are carried out to image
Information eliminates the irrelevant information of image, so as to the feature of more easily capture images, is more convenient next operation and judgement.
Moreover, the horizontal pixel point number XMax and longitudinal pixel number YMax of image are obtained in advance.
S3, small area connected domain is deleted;
S4, main connected domain burr is deleted, the irrelevant point that step S2 and S3 are carried over excludes, and at the same time obtaining
The datum mark of fitting a straight line, so as to the formulation boundary line of next step;
S5, fitting a straight line and rectangle correction;
The hole center of positioning microwell plate comprises the following steps:
S6, fuzzy positioning holes center:The center in each hole is determined according to the step S5 rectangular areas being obtained;
S7, it is accurately positioned hole center.
Further, step S2 image preprocessings include gray level image, Gassian low-pass filter and binary image.
Further, step S3 deletes small area connected domain and includes mainly including traveling through and recording eight connectivity region, calculate
The area of each connected region, sorting from small to large by area to all connected domains connects with according to corresponding proportion deletion small area
Domain.It is specific as follows:
S3-1, traversal simultaneously record eight connectivity region:The function (bwlabeln and regionprops) carried using MATLAB
To calculate and record the unicom in small object region point number and the corresponding position of each connectivity points;
S3-2, the area for calculating each connected domain:Since the size of a connected domain can be by the sum of its connectivity points
Instead of, therefore draw by the numbers of the connectivity points for counting each connected domain the area of each connected domain;
S3-3, sort from small to large by area to all connected domains;
S3-4, small area connected domain is deleted in proportion.As the case may be, the 50%~80% of all connected domains is deleted
Zonule eliminates the connected region of small object area, and realization further eliminates noise, excludes unrelated pixel.
Further, step S4 deletes main connected domain burr and is surveyed successively from the upper and lower, left and right four direction of image to use
Pilot approaches actual boundary line;Single direction approximation actual boundary line is as follows:
S4-1, test point is obtained:Using the parameter (XMax and YMax) being passed among pre-treatment step, according to image
Feature, according to α %'s (α % are the extraction number of test point and the ratio of image length or width pixel size required for imagining)
Ratio come it is random obtain the transverse test point that number is XMax* α %, and with the coordinate value of each test point under data structure records
(x, 0);
S4-2, judge whether current test point is filling point (whether pixel point value is " 1 "), if " 1 ", be defined as touching
It hits;Otherwise, non-collision state is defined as, transverse test point advances to picture centre direction;
If S4-3, current test point collision, centered on current test point, formulate an a length of XMax* β %, width is
Then the rectangular area of YMax* β % calculates the number N of filler pixels point in the region;If the number of N is more than or equal to M, (M is
Determine whether the critical value of burr), then it is assumed that the test point extremely closes on actual boundary line, halts, and enters
Step S4-4;On the contrary, when N is less than M, then it is assumed that collision area is irrelevant point, is eliminated, and skips the region, after
It is continuous to advance, into next test point, return to step S4-2;If current test point is not collided, continue before picture centre direction
Into into next test point, the pixel in region being all changed to " 0 ", while test point coordinate is changed to (x, y+
YMax* β %), return to step S4-2;
S4-4, record test point coordinate, are included the point set for formulating boundary line.
Further, step S5 fitting a straight lines, which are found intersection, is as follows:
S5-1, four groups of datum marks up and down are obtained:Obtain the point set obtained by step S4;
S5-2, dot interlace line is carried out to every group of datum mark:Since adjacent 2 points of straight line variations done are excessive, take
It is separated by the rule of a point to take at 2 points, then calculates straight line line;
Filling points on S5-3, statistics every fitting a straight line of each group:
The filling points that S5-4, every group of basis count select the most straight line of filling points, and the straight line are made
For the prescription to boundary line;
S5-5, four straight line simultaneous linear equation groups being obtained according to step S5-4 acquire four endpoints of rectangular shaped rim,
And it records.
S5-6, four vertex being obtained according to step S5-5, utilize the rectangle the rectification function of opencv
GetPerspectiveTransform () and warpPerspective () carries out perspective transform, by the correction of microwell plate picture simultaneously
It extracts.
Further, step S6 obscures positioning holes center and is as follows:
S6-1, the length and width for confirming rectangular area in image:Coordinate difference between four endpoints being obtained using step S5
It is different, the rectangular area length of side in image is obtained, judges the length and width of rectangular area;
S6-2, basis survey the length ratio relation between measured microwell plate frame and microwell plate border in advance, to image
Each boundary line of rectangle carries out equal proportion cutting, and records the seat of cut point;
S6-3, fitting a straight line is carried out according to corresponding two cut points of upper and lower correspondence or left and right, left and right directions and upper is obtained
The intersection point of two straight lines in lower direction, is the center of circle, and the proportionate relationship of the length of side and circle hole radius further according to microwell plate is obtained half
Footpath;
S6-4, the coordinate value at each hole center of picture centre region and corresponding radius are recorded one by one respectively.
Further, step S7 is accurately positioned hole center and is as follows:
S7-1, the position according to radius and the center of circle, one by one extract each circular hole, the target area as processing;
S7-2, using Hough function houghCricle (), the circular of relevant radii size is carried out to target area and is identified;
S7-3, all circles identified by average value and known radius and central coordinate of circle screening step S7-2,
The center of circle of new definition is deviateed the circle that the former center of circle is more than 10% pixel value and is removed, and the X in the center of circle, Y-coordinate and radius are put down
Mean value computation;
S7-4, the hole central coordinate of circle after adjustment is added the relative position of target area in the picture, drawn each in image
The hole centre coordinate of a circular hole and the value of relevant radii, i.e., the position of each hole center in the picture.
This programme principle is as follows:
By how to be accurately positioned each hole center of microwell plate, be decomposed into the square boundary that how to position microwell plate and how
Square boundary according to having good positioning is accurately positioned the hole center of microwell plate.
Position microwell plate square boundary when, using image processing techniques simplification figure picture, and delete the noise in image and
Then burr draws square boundary by the method that fitting a straight line detects, and the rectangular area to navigating to carries out rectangle perspective
Conversion extracts the microwell plate rectangular area picture of standard.
When being accurately positioned the hole center of microwell plate, since the rectangular area of microwell plate is standard, and actual ratio will not
Change, the equal proportion point of each boundary line is obtained according to the ratio of the frame of microwell plate and the length of side, in the square for positioning and having corrected
Equal proportion straight-line intersection is calculated in shape, i.e., the hole center in each hole so as to fulfill the fuzzy positioning of microwell plate, recycles suddenly
Husband function houghCricle () adjusts accordingly the position in each hole, reduce round orifices orientation to square boundary position according to
Lai Xing, the positioning for making the center of circle is more flexible, and position is more accurate, so as to which the hole center for solving the problems, such as microwell plate is accurately positioned.
Compared with prior art, using image and fitting a straight line, square boundary is positioned, not only efficiently precisely, also
The hole centralized positioning of microwell plate is served vital, the hole center to position microwell plate provides the guarantor of accuracy
Card.And it is greatly improved for the work efficiency of hospital and Lab workers.
Description of the drawings
Fig. 1 is the flow chart of the accurate positioning method at micropore center in a kind of microwell plate picture taken pictures naturally of the present invention;
Fig. 2 is the mold schematic diagram of microwell plate;
Fig. 3 is the flow chart that small area connected domain is deleted in the present invention;
Fig. 4 is the flow chart that main connected domain burr is deleted in the present invention;
Fig. 5 is the flow chart that cathetus of the present invention fitting is found intersection;
Fig. 6 is the flow chart that positioning holes center is obscured in the present invention;
Fig. 7 is the flow chart that hole center is accurately positioned in the present invention.
Specific embodiment
With reference to specific embodiment, the invention will be further described:
Referring to shown in attached drawing 1, micropore center is accurate in a kind of microwell plate picture taken pictures naturally described in the present embodiment
Localization method first positions the square boundary of microwell plate, then according to the hole center for the square boundary positioning microwell plate having good positioning;
Wherein, the square boundary for positioning microwell plate comprises the following steps:
S1, the image that microwell plate is shot under natural lighting;
S2, gray processing, this low-pass filtering and binary conversion treatment are carried out to image;
S3, small area connected domain is deleted;As shown in figure 3, it is as follows:
S3-1, traversal simultaneously record eight connectivity region:The function (bwlabeln and regionprops) carried using MATLAB
To calculate and record the unicom in small object region point number and the corresponding position of each connectivity points;
S3-2, the area for calculating each connected domain;
S3-3, sort from small to large by area to all connected domains;
S3-4, small area connected domain is deleted in proportion.
S4, main connected domain burr is deleted;It needs to approach from image upper and lower, left and right four direction using test point successively respectively
Actual boundary line, as shown in figure 4, single direction approximation actual boundary line is as follows:
S4-1, test point is obtained;
S4-2, judge whether current test point collides;
If S4-3, current test point collision, centered on current test point, formulate an a length of XMax* β %, width is
Then the rectangular area of YMax* β % calculates the number N of filler pixels point in the region;If the number of N is more than or equal to M,
It halts, and enters step S4-4;On the contrary, when N is less than M, then it is assumed that collision area is irrelevant point, is eliminated, and
And the region is skipped, it moves on, into next test point, return to step S4-2;If current test point is not collided, continue to
Picture centre direction is advanced, into next test point, return to step S4-2;
S4-4, record test point coordinate, are included the point set for formulating boundary line.
S5, fitting a straight line are found intersection;As shown in figure 5, it is as follows:
S5-1, four groups of datum marks up and down are obtained;
S5-2, dot interlace line is carried out to each group datum mark, avoids adjacent 2 points of straight line variations done excessive;
Filling points on S5-3, statistics every fitting a straight line of each group;
S5-4, counted according to the filling of statistics, filling be then obtained and counts most straight line, and using the straight line as
The prescription to boundary line;
S5-5, four straight line simultaneous linear equation groups being obtained according to step S5-4 acquire four endpoints of rectangular shaped rim,
And it records.
S5-6, four vertex being obtained according to step S5-5, utilize the rectangle the rectification function of opencv
GetPerspectiveTransform () and warpPerspective () carries out perspective transform, by the correction of microwell plate picture simultaneously
It extracts.
The hole center of positioning microwell plate comprises the following steps:
S6, fuzzy positioning holes center;As shown in fig. 6, it is as follows:
S6-1, the length and width for confirming rectangular area in image;
S6-2, basis survey the length ratio relation between measured microwell plate frame and microwell plate border in advance, to image
Each boundary line of rectangle carries out equal proportion cutting, and records the seat of cut point;
S6-3, fitting a straight line is carried out according to corresponding two cut points of upper and lower correspondence or left and right, left and right directions and upper is obtained
The intersection point of two straight lines in lower direction, is the center of circle, and the proportionate relationship of the length of side and circle hole radius further according to microwell plate is obtained half
Footpath;
S6-4, the coordinate value at each hole center of image target area and corresponding radius are recorded one by one respectively.
S7, it is accurately positioned hole center;As shown in fig. 7, it is as follows:
S7-1, the position according to radius and the center of circle, one by one extract each circular hole, the target area as processing;
S7-2, circular identification is carried out using Hough function pair target area;
S7-3, all circles identified by average value and known radius and central coordinate of circle screening step S7-2,
The center of circle of new definition is deviateed the circle that the former center of circle is more than 10% pixel value and is removed, and the X in the center of circle, Y-coordinate and radius are put down
Mean value computation;
S7-4, the hole central coordinate of circle after adjustment is added the relative position of target area in the picture, drawn each in image
The hole centre coordinate of a circular hole and the value of relevant radii.
The present embodiment utilizes image and fitting a straight line, and square boundary is positioned, not only efficiently precisely, also to microwell plate
Hole centralized positioning served vital, the hole center to position microwell plate provides the guarantee of accuracy.It is and right
It is greatly improved in the work efficiency of hospital and Lab workers.
The examples of implementation of the above are only the preferred embodiments of the invention, and the implementation model of the present invention is not limited with this
It encloses, therefore the variation that all shape, principles according to the present invention are made, it should all cover within the scope of the present invention.
Claims (7)
1. a kind of accurate positioning method at micropore center in microwell plate picture taken pictures naturally, it is characterised in that:First position micropore
The square boundary of plate, then according to the hole center for the square boundary positioning microwell plate having good positioning;
Wherein, the square boundary for positioning microwell plate comprises the following steps:
S1, the image that microwell plate is shot under natural lighting;
S2, image is pre-processed;
S3, small area connected domain is deleted;
S4, main connected domain burr is deleted;
S5, fitting a straight line simultaneously correct rectangle;
The hole center of positioning microwell plate comprises the following steps:
S6, fuzzy positioning holes center;
S7, it is accurately positioned hole center.
2. the accurate positioning method at micropore center in a kind of microwell plate picture taken pictures naturally according to claim 1,
It is characterized in that:The step S2 image preprocessings include gray level image, Gassian low-pass filter and binary image.
3. the accurate positioning method at micropore center in a kind of microwell plate picture taken pictures naturally according to claim 1,
It is characterized in that:Small area connected domain is deleted described in step S3, is as follows:
S3-1, traversal simultaneously record eight connectivity region;
S3-2, the area for calculating each connected domain;
S3-3, sort from small to large by area to all connected domains;
S3-4, small area connected domain is deleted in proportion.
4. the accurate positioning method at micropore center in a kind of microwell plate picture taken pictures naturally according to claim 1,
It is characterized in that:The step S4 deletes main connected domain burr, respectively successively from image upper and lower, left and right four direction using test
Point approaches actual boundary line;
Single direction approximation actual boundary line is as follows:
S4-1, test point is obtained;
S4-2, judge whether current test point collides;
If S4-3, current test point collision, centered on current test point, formulate an a length of XMax* β %, width YMax*
Then the rectangular area of β % calculates the number N of filler pixels point in the region;If the number of N is more than or equal to M, before stopping
Into, and enter step S4-4;On the contrary, when N is less than M, then it is assumed that collision area is irrelevant point, is eliminated, and skipped
The region, moves on, into next test point, return to step S4-2;If current test point is not collided, continue into image
Heart direction is advanced, into next test point, return to step S4-2;
S4-4, record test point coordinate, are included the point set for formulating boundary line.
5. the accurate positioning method at micropore center in a kind of microwell plate picture taken pictures naturally according to claim 1,
It is characterized in that:The step S5 fitting a straight lines are simultaneously corrected rectangle and are as follows:
S5-1, four groups of datum marks up and down are obtained;
S5-2, dot interlace line is carried out to every group of datum mark;
Filling points on S5-3, statistics every fitting a straight line of each group;
S5-4, every group according to the filling points counted, select filling and count most straight line, and using the straight line as being somebody's turn to do
Prescription to boundary line;
S5-5, four straight line simultaneous linear equation groups being obtained according to step S5-4, acquire four endpoints of rectangular shaped rim, and remember
Record is got off.
S5-6, four vertex being obtained according to step S5-5, utilize the rectangle the rectification function of opencv
GetPerspectiveTransform () and warpPerspective () carries out perspective transform, by the correction of microwell plate picture simultaneously
It extracts.
6. the accurate positioning method at micropore center in a kind of microwell plate picture taken pictures naturally according to claim 1,
It is characterized in that:The step S6 obscures positioning holes center and is as follows:
S6-1, the length and width for confirming rectangular area in image;
S6-2, basis survey the length ratio relation between measured microwell plate frame and microwell plate border in advance, to image rectangle
Each boundary line carry out equal proportion cutting, and record the coordinate of cut point;
S6-3, fitting a straight line is carried out according to corresponding two cut points of upper and lower correspondence or left and right, left and right directions and upper and lower is obtained
To two straight lines intersection point, be the center of circle, radius is obtained in the proportionate relationship of the length of side and circle hole radius further according to microwell plate;
S6-4, the coordinate value at each hole center of image target area and corresponding radius are recorded respectively.
7. the accurate positioning method at micropore center in a kind of microwell plate picture taken pictures naturally according to claim 1,
It is characterized in that:The step S7 is accurately positioned hole center and is as follows:
S7-1, the position according to radius and the center of circle, one by one extract each circular hole, the target area as processing;
S7-2, circular identification is carried out using Hough function pair target area;
S7-3, all circles identified by average value and known radius and central coordinate of circle screening step S7-2, new fixed
The center of circle of position is deviateed the circle that the former center of circle is more than 10% pixel value and is removed, and carries out average value to the X in the center of circle, Y-coordinate and radius
It calculates;
S7-4, the hole central coordinate of circle after adjustment is added the relative position of target area in the picture, draws each circle in image
The hole centre coordinate in hole and the value of relevant radii.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711007652.8A CN108062770B (en) | 2017-10-25 | 2017-10-25 | Method for accurately positioning center of micropore in picture of micropore plate by natural photographing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711007652.8A CN108062770B (en) | 2017-10-25 | 2017-10-25 | Method for accurately positioning center of micropore in picture of micropore plate by natural photographing |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108062770A true CN108062770A (en) | 2018-05-22 |
CN108062770B CN108062770B (en) | 2021-08-17 |
Family
ID=62137030
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711007652.8A Active CN108062770B (en) | 2017-10-25 | 2017-10-25 | Method for accurately positioning center of micropore in picture of micropore plate by natural photographing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108062770B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111723734A (en) * | 2020-06-19 | 2020-09-29 | 哈尔滨理工大学 | Hole site identification method based on machine vision |
CN115540746A (en) * | 2022-01-06 | 2022-12-30 | 深圳荣耀智能机器有限公司 | Method for detecting position degree of side hole of structural part |
CN116298241A (en) * | 2023-01-09 | 2023-06-23 | 浙江宝太智能科技有限公司 | Hole site positioning method for chemiluminescent immunoassay analyzer detection system |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101206116A (en) * | 2007-12-07 | 2008-06-25 | 北京机械工业学院 | Goal spot global automatic positioning method |
CN104463159A (en) * | 2014-12-31 | 2015-03-25 | 北京释码大华科技有限公司 | Image processing method and device of iris positioning |
CN105469084A (en) * | 2015-11-20 | 2016-04-06 | 中国科学院苏州生物医学工程技术研究所 | Rapid extraction method and system for target central point |
WO2017056130A1 (en) * | 2015-09-30 | 2017-04-06 | Nec Corporation | Image processing device, image processing method and storage medium |
-
2017
- 2017-10-25 CN CN201711007652.8A patent/CN108062770B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101206116A (en) * | 2007-12-07 | 2008-06-25 | 北京机械工业学院 | Goal spot global automatic positioning method |
CN104463159A (en) * | 2014-12-31 | 2015-03-25 | 北京释码大华科技有限公司 | Image processing method and device of iris positioning |
WO2017056130A1 (en) * | 2015-09-30 | 2017-04-06 | Nec Corporation | Image processing device, image processing method and storage medium |
CN105469084A (en) * | 2015-11-20 | 2016-04-06 | 中国科学院苏州生物医学工程技术研究所 | Rapid extraction method and system for target central point |
Non-Patent Citations (2)
Title |
---|
于明鑫 等: "基于灰度信息和支持向量机的人眼检测方法", 《工程科学学报》 * |
王拯洲 等: "基于圆拟合的非完整圆激光光斑中心检测算法", 《激光与红外》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111723734A (en) * | 2020-06-19 | 2020-09-29 | 哈尔滨理工大学 | Hole site identification method based on machine vision |
CN115540746A (en) * | 2022-01-06 | 2022-12-30 | 深圳荣耀智能机器有限公司 | Method for detecting position degree of side hole of structural part |
CN116298241A (en) * | 2023-01-09 | 2023-06-23 | 浙江宝太智能科技有限公司 | Hole site positioning method for chemiluminescent immunoassay analyzer detection system |
Also Published As
Publication number | Publication date |
---|---|
CN108062770B (en) | 2021-08-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106407981B (en) | A kind of licence plate recognition method, apparatus and system | |
CN106529424B (en) | A kind of logo detection recognition method and system based on selective search algorithm | |
De Charette et al. | Real time visual traffic lights recognition based on spot light detection and adaptive traffic lights templates | |
CN105913093B (en) | A kind of template matching method for Text region processing | |
CN102704215B (en) | Automatic cutting method of embroidery cloth based on combination of DST file parsing and machine vision | |
CN105574527B (en) | A kind of quick object detecting method based on local feature learning | |
CN103310194B (en) | Pedestrian based on crown pixel gradient direction in a video shoulder detection method | |
CN107389701A (en) | A kind of PCB visual defects automatic checkout system and method based on image | |
CN106296653A (en) | Brain CT image hemorrhagic areas dividing method based on semi-supervised learning and system | |
CN106056597B (en) | Object visible detection method and device | |
CN110232379A (en) | A kind of vehicle attitude detection method and system | |
CN108062770A (en) | The accurate positioning method at micropore center in a kind of microwell plate picture taken pictures naturally | |
CN105404857A (en) | Infrared-based night intelligent vehicle front pedestrian detection method | |
CN103778411B (en) | Circle detection method and device based on raster image division | |
CN110097596A (en) | A kind of object detection system based on opencv | |
CN108257124A (en) | A kind of white blood cell count(WBC) method and system based on image | |
CN106504262A (en) | A kind of small tiles intelligent locating method of multiple features fusion | |
CN110222625A (en) | A kind of method and system that the identifying rows in video monitoring image are humanized | |
CN102930251A (en) | Two-dimensional collection data recording and discriminating device and method | |
CN105261006A (en) | Medical image segmentation algorithm based on Fourier transform | |
CN103699876B (en) | Method and device for identifying vehicle number based on linear array CCD (Charge Coupled Device) images | |
Ma et al. | An intelligent object detection and measurement system based on trinocular vision | |
CN103455826A (en) | Efficient matching kernel body detection method based on rapid robustness characteristics | |
CN108229447A (en) | A kind of high beam detection method based on video flowing | |
Xu et al. | License plate recognition system based on deep learning |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |