CN109781747A - Cytologic specimen print dyeing effect detection method and system based on machine vision - Google Patents
Cytologic specimen print dyeing effect detection method and system based on machine vision Download PDFInfo
- Publication number
- CN109781747A CN109781747A CN201910164297.8A CN201910164297A CN109781747A CN 109781747 A CN109781747 A CN 109781747A CN 201910164297 A CN201910164297 A CN 201910164297A CN 109781747 A CN109781747 A CN 109781747A
- Authority
- CN
- China
- Prior art keywords
- dyeing
- image
- cell
- cytologic specimen
- 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
- 238000004043 dyeing Methods 0.000 title claims abstract description 181
- 230000000694 effects Effects 0.000 title claims abstract description 74
- 238000001514 detection method Methods 0.000 title claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 25
- 239000010893 paper waste Substances 0.000 claims abstract 3
- 238000000034 method Methods 0.000 claims description 37
- 230000008569 process Effects 0.000 claims description 19
- 239000011521 glass Substances 0.000 claims description 16
- 101100117236 Drosophila melanogaster speck gene Proteins 0.000 claims description 12
- 235000013399 edible fruits Nutrition 0.000 claims description 10
- 238000003860 storage Methods 0.000 claims description 8
- 230000018044 dehydration Effects 0.000 claims description 7
- 238000006297 dehydration reaction Methods 0.000 claims description 7
- 238000004321 preservation Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000004140 cleaning Methods 0.000 claims description 5
- 238000002203 pretreatment Methods 0.000 claims description 4
- 238000005903 acid hydrolysis reaction Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 2
- 230000009467 reduction Effects 0.000 claims description 2
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 description 17
- OKKJLVBELUTLKV-UHFFFAOYSA-N Methanol Chemical compound OC OKKJLVBELUTLKV-UHFFFAOYSA-N 0.000 description 9
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 8
- 239000012153 distilled water Substances 0.000 description 7
- QTBSBXVTEAMEQO-UHFFFAOYSA-N Acetic acid Chemical compound CC(O)=O QTBSBXVTEAMEQO-UHFFFAOYSA-N 0.000 description 6
- WSFSSNUMVMOOMR-UHFFFAOYSA-N Formaldehyde Chemical compound O=C WSFSSNUMVMOOMR-UHFFFAOYSA-N 0.000 description 6
- VEXZGXHMUGYJMC-UHFFFAOYSA-N Hydrochloric acid Chemical compound Cl VEXZGXHMUGYJMC-UHFFFAOYSA-N 0.000 description 6
- DKGAVHZHDRPRBM-UHFFFAOYSA-N Tert-Butanol Chemical compound CC(C)(C)O DKGAVHZHDRPRBM-UHFFFAOYSA-N 0.000 description 6
- 238000004062 sedimentation Methods 0.000 description 6
- 229960000935 dehydrated alcohol Drugs 0.000 description 5
- 238000004040 coloring Methods 0.000 description 4
- 229960004756 ethanol Drugs 0.000 description 4
- 235000019441 ethanol Nutrition 0.000 description 4
- 230000011218 segmentation Effects 0.000 description 4
- 229960000583 acetic acid Drugs 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 239000012362 glacial acetic acid Substances 0.000 description 3
- 230000007812 deficiency Effects 0.000 description 2
- 238000002791 soaking Methods 0.000 description 2
- 238000010186 staining Methods 0.000 description 2
- ANRHNWWPFJCPAZ-UHFFFAOYSA-M thionine Chemical compound [Cl-].C1=CC(N)=CC2=[S+]C3=CC(N)=CC=C3N=C21 ANRHNWWPFJCPAZ-UHFFFAOYSA-M 0.000 description 2
- 238000005406 washing Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000009514 concussion Effects 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- WSFSSNUMVMOOMR-NJFSPNSNSA-N methanone Chemical compound O=[14CH2] WSFSSNUMVMOOMR-NJFSPNSNSA-N 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000391 smoking effect Effects 0.000 description 1
- 239000012128 staining reagent Substances 0.000 description 1
Landscapes
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention discloses cytologic specimen print dyeing effect detection methods and system based on machine vision, comprising: image processing step: Image Acquisition, carries out image procossing to cytologic specimen print, identifies coloration result；If dyeing effect is good, dye uniformity identification step is entered；If dyeing effect is poor, the time of second of dyeing is just determined according to processing result image；If continuously three times dyeing effect be all it is poor, be considered as dyeing failure, using cytologic specimen print as waste paper handle；Dye uniformity identification step: being calculated using uniformity algorithm, dye uniformity caused by identification cell is agglomerating；Cell mass position identification step: the physical location alignment algorithm adhered to using cytologic specimen slide cell calculates cell mass position, identifies cell mass position.
Description
Technical field
This disclosure relates to which cytologic specimen production field, contaminates more particularly to the cytologic specimen print based on machine vision
Color effect detection method and system.
Background technique
The statement of this part is only to refer to background technique relevant to the disclosure, not necessarily constitutes the prior art.
With the fast development of artificial intelligence, the intelligence of all trades and professions is more and more obvious.Before being related to cell observation
There is deficiencies in intelligence degree for preparation.In a large amount of observations point of some needs such as hospital, biology or medical research institute
The place of cell is analysed, the process of cell dyeing film-making needs manually to be operated completely, not only wastes a large amount of manpower in this way
Financial resources, and cell sample may be caused to pollute or damage because of some maloperations during operation.
In the implementation of the present invention, following technical problem exists in the prior art in inventor:
(1): in the prior art, fine evaluation, Bu Nengtong cannot be carried out by image procossing to dyeing effect quality
The result of image procossing is crossed further to instruct the time of secondary dyeing；
(2): in the prior art, cannot it is agglomerating to cell caused by dyeing problem of non-uniform identified and handled；
(3): in the prior art, cannot be to cell settlement during, deviation existing for sedimentation location and normal place carry out
Precisely identification.
Summary of the invention
In order to solve the deficiencies in the prior art, present disclose provides the cytologic specimen prints based on machine vision to dye effect
Fruit detection method and system；
In a first aspect, present disclose provides the cytologic specimen print dyeing effect detection methods based on machine vision；
Cytologic specimen print dyeing effect detection method based on machine vision, comprising:
Image processing step: Image Acquisition carries out image procossing to cytologic specimen print, identifies coloration result；If
Dyeing effect is good, enters dye uniformity identification step；If dyeing effect is poor, second just is determined according to processing result image
The time of secondary dyeing；If continuously three times dyeing effect be all it is poor, be considered as dyeing failure, using cytologic specimen print as give up
Piece processing；
Dye uniformity identification step: being calculated using uniformity algorithm, even dyeing caused by identification cell is agglomerating
Degree；
Cell mass position identification step: the physical location alignment algorithm adhered to using cytologic specimen slide cell, to thin
Born of the same parents' cumularsharolith, which is set, to be calculated, and identifies cell mass position.
As a kind of possible implementation, described image processing step is specifically included:
Step (21): the image of acquisition cytologic specimen print pre-processes image, divides the image into foreground picture
Picture and background image；
Step (22): carrying out graphics template matching to pretreated image, judge whether successful match,
If successful match, it is good to be treated as coloration result；If it fails to match, carries out second and dye, and according to preceding
The average gray value of scape image determines the duration of second of dyeing；After second is dyed, it is again introduced into step (21)；
If the image of second of dyeing successful match in graphics template matching process, it is good to be treated as coloration result；Such as
Fruit matching fails again, then carries out third time dyeing；
If the image of third time dyeing successful match in graphics template matching process, it is good to be treated as coloration result；Such as
It fails to match for the dyeing of fruit third time, then is considered as dyeing failure, terminates.
As a kind of possible implementation, step (22) specific steps are as follows:
Positive round is found in foreground image by template matching；The positive round is standard round；A diameter of setting value；
If positive round can be found in foreground image, illustrate that dyeing effect is preferable；If can not find positive round, illustrate dyeing effect
Fruit is poor；It carries out second to dye, and determines the duration of second of dyeing according to the average gray value of foreground image；Assuming that prospect
The average gray value of image is n, and dyeing total duration is T；
Secondary dyeing duration t are as follows:
T=T*0.3* (256-n)；
After second is dyed, it is again introduced into step (21)；If the image of second of dyeing is matched in graphics template
It is good to be treated as coloration result for successful match in the process；If matching fails again, third time dyeing is carried out；And according to secondary
The calculation of duration determines the duration of third time dyeing after dyeing；
Similarly, if the image of third time dyeing successful match in graphics template matching process, is treated as coloration result
It is good；If it fails to match for third time dyeing, it is considered as dyeing failure, terminates.
As a kind of possible implementation, the dye uniformity identification step is specifically included:
Step (31): the average gray value g1 of foreground image area-of-interest is calculated；
Step (32): the highest gray value g2 of foreground image area-of-interest is calculated；
Step (33): it calculates large size packed cell degree G:G=(g1-g2)/256；
Step (34): large-scale packed cell degree G is compared with given threshold, judges whether dyeing is uniform, if greatly
In given threshold, it is considered as dyeing unevenly, image is stored in dyeing speck database if uneven；Otherwise it is considered as contaminating
Color is uniform, terminates.
As a kind of possible implementation, cell mass position identification step is specifically included:
Step (41): to foreground image, fitting circle is formed using least square method；
Step (42): the distance of positioning fitting circle center location to slide left margin, right margin and lower boundary；
Step (43): calculating the left and right accuracy of cell attachment position, if left and right accuracy is greater than given threshold, table
Show that cell attachment position right-left error is big, just the dyeing speck database that slide information preservation is big to cell position deviation
In；If left and right accuracy is less than given threshold, then it represents that the accurate zero deflection of cell attachment position left and right directions terminates；
Step (44): calculating the accuracy up and down of cell attachment position, if accuracy is greater than given threshold, table up and down
Show that cell attachment position up and down direction deviation is big, just the dyeing speck database that slide information preservation is big to cell position deviation
In；If upper and lower accuracy is less than given threshold, then it represents that the accurate zero deflection of cell attachment position up and down direction terminates.
Second aspect, the disclosure additionally provide the cytologic specimen print dyeing effect detection system based on machine vision；
Cytologic specimen print dyeing effect detection system based on machine vision, comprising:
Image processing module: Image Acquisition carries out image procossing to cytologic specimen print, identifies coloration result；If
Dyeing effect is good, enters dye uniformity identification step；If dyeing effect is poor, second just is determined according to processing result image
The time of secondary dyeing；If continuously three times dyeing effect be all it is poor, be considered as dyeing failure, using cytologic specimen print as give up
Piece processing；
Dye uniformity identification module: being calculated using uniformity algorithm, even dyeing caused by identification cell is agglomerating
Degree；
Cell mass location identification module: the physical location alignment algorithm adhered to using cytologic specimen slide cell, to thin
Born of the same parents' cumularsharolith, which is set, to be calculated, and identifies cell mass position.
The third aspect, the disclosure additionally provide a kind of electronic equipment；
A kind of electronic equipment, comprising: the meter that memory, processor and storage execute on a memory and on a processor
The instruction of calculation machine, when the computer instruction is run by processor, completes step described in the method for above-mentioned first aspect.
Fourth aspect, the disclosure additionally provide a kind of computer readable storage medium；
A kind of computer readable storage medium, is stored thereon with computer instruction, and the computer instruction is transported by processor
When row, step described in the method for above-mentioned first aspect is completed.
Compared with prior art, the beneficial effect of the disclosure is:
With improved automatic threshold segmentation algorithm, original image is divided into foreground image and background image, then,
The duration that second of dyeing is determined according to the average gray value of foreground image, saves the time of secondary dyeing；
It proposes and unevenly proposes the Uniformity Analysis algorithm of coloring effect for the agglomerating caused dyeing of cell；
During cell settlement, the dyeing speck of deviation existing for sedimentation location and normal place proposes cell
Specimen cell mass positional accuracy alignment algorithm is learned, both dyeing specks are identified in dyeing course by algorithm
Come, and prompt user in program interface, is about to defect information certainly and is saved in corresponding defect database.To greatly save
Manpower and material resources realize the dyeing course that cytologic specimen print is completed with machine vision auxiliary mechanical arm.
Detailed description of the invention
The accompanying drawings constituting a part of this application is used to provide further understanding of the present application, and the application's shows
Meaning property embodiment and its explanation are not constituted an undue limitation on the present application for explaining the application.
Fig. 1 is dyeing overall flow figure of the invention；
Fig. 2 is that algorithm flow chart is evaluated in dyeing of the invention；
Fig. 3 is the Uniformity Analysis algorithm flow chart of coloring effect；
Fig. 4 is cytologic specimen slide cell mass positional accuracy alignment algorithm flow chart of the invention.
Specific embodiment
It is noted that described further below be all exemplary, it is intended to provide further instruction to the application.Unless another
It indicates, all technical and scientific terms used herein has usual with the application person of an ordinary skill in the technical field
The identical meanings of understanding.
It should be noted that term used herein above is merely to describe specific embodiment, and be not intended to restricted root
According to the illustrative embodiments of the application.As used herein, unless the context clearly indicates otherwise, otherwise singular
Also it is intended to include plural form, additionally, it should be understood that, when in the present specification using term "comprising" and/or " packet
Include " when, indicate existing characteristics, step, operation, device, component and/or their combination.
Embodiment one: the cytologic specimen print dyeing effect detection method based on machine vision is provided；
As shown in Figure 1, the cytologic specimen print dyeing effect detection method based on machine vision, comprising:
Image processing step: Image Acquisition carries out image procossing to cytologic specimen print, identifies coloration result；If
Dyeing effect is good, enters dye uniformity identification step；If dyeing effect is poor, second just is determined according to processing result image
The time of secondary dyeing；If continuously three times dyeing effect be all it is poor, be considered as dyeing failure, using cytologic specimen print as give up
Piece processing；
Dye uniformity identification step: being calculated using uniformity algorithm, even dyeing caused by identification cell is agglomerating
Degree；
Cell mass position identification step: the physical location alignment algorithm adhered to using cytologic specimen slide cell, to thin
Born of the same parents' cumularsharolith, which is set, to be calculated, and identifies cell mass position.
It as one embodiment, further include pre-treatment step before described image processing step；
The pre-treatment step, specifically includes: cytologic specimen print being placed in glass frame, mechanical arm grabs slide
Glass frame is placed into BS fixer M minutes by frame, is placed into acid hydrolysis solution after then cleaning glass frame N minutes；Then will
It is placed into dyeing liquor L minutes after glass frame cleaning；Take out glass frame.
As one embodiment, after the identification step of the cell mass position, further includes: dehydration；
The dehydration, specifically includes: after cleaning to glass frame, carrying out serial dehydration to print, dyeing is completed.
The dye vat that water smoking successively uses is the graded ethanol of various concentration.It is respectively as follows:
50% ethyl alcohol 500ml (250ml dehydrated alcohol, 250ml distilled water),
75% ethyl alcohol 500ml (375ml dehydrated alcohol, 125ml distilled water),
95% ethyl alcohol 500ml (475ml dehydrated alcohol, 25ml distilled water),
Dehydrated alcohol 500ml (500ml dehydrated alcohol).
In addition to absolute alcohol, remaining gradient respectively impregnates 1.5min, and soaked in absolute ethyl alcohol two minutes.Realize serial dehydration.
As one embodiment,
The BS fixer be specifically by volume ratio be 80% methanol, 15% formaldehyde and 5% glacial acetic acid configuration and
At；BS fixer (BS fixing solution): it is mainly mixed by methanol, formaldehyde, glacial acetic acid etc., is mainly used for cell
The dyeing of core is fixed, and the dyeing for being particularly suitable for nucleus DNA is fixed.Its main contents dyed is that its first staining jar is put
BS fixer (methanol (80%) formaldehyde (15%) glacial acetic acid (5%) configures) is set, fixes 30 minutes wherein.
It is that 58% distilled water and 42% hydrochloric acid configure that the acid hydrolysis solution, which is specifically by volume ratio,；Mechanical arm is by glass
Horse is integrally taken to sink and carries out soaking and washing.Second staining jar is 5N hydrochloric acid (distilled water (58%) concentrated hydrochloric acid (42%)),
Carry out acidolysis 25min.After mechanical arm glass frame is integrally taken to sink carry out soaking and washing.
The dyeing liquor be specifically by volume ratio be 1% thionine, 1% tert-butyl alcohol and 98% distilled water configuration and
At.Third dye vat is that staining reagent (thionine (1%) tert-butyl alcohol (1%) distilled water (98%)) places 1H.
As one embodiment, as shown in Fig. 2, described image processing step, specifically includes:
Step (21): the disclosure is using monocular vision come to simply being rinsed after dyeing and be sent to camera lens by mechanical arm
Preceding cytologic specimen print carries out front and takes pictures, and acquires the image of cytologic specimen print, after taking pictures, is located in advance to image
Reason, divides the image into foreground image and background image；Extract foreground image；
Step (22): carrying out graphics template matching to pretreated image, judge whether successful match,
If successful match, it is good to be treated as coloration result；If it fails to match, carries out second and dye, and according to preceding
The average gray value of scape image determines the duration of second of dyeing；After second is dyed, it is again introduced into step (21)；
If the image of second of dyeing successful match in graphics template matching process, it is good to be treated as coloration result；Such as
Fruit matching fails again, then carries out third time dyeing；
If the image of third time dyeing successful match in graphics template matching process, it is good to be treated as coloration result；Such as
It fails to match for the dyeing of fruit third time, then is considered as dyeing failure, terminates.
By multiple comparative experiments, we assert that the time of general secondary dyeing is no more than percent the 50 of total duration.Such as
Fruit is too many beyond the time, and judging from the experimental results, reason is substantially all for during the infall process of film-making or collection of specimens
Caused by cell quantity is insufficient.So we select to redye selection of time for the first time in this way can be with for percent the 30. of total duration
Solve the problems, such as most of because dyeing effect caused by dyeing time is insufficient is poor.
The time of secondary dyeing is determined according to the average gray value of prospect.Gray level image amounts to 256 gray levels, it is assumed that
The average gray of prospect is n, and dyeing total duration is T, secondary dyeing duration are as follows: t=T*0.3* (256-n).Secondary dyeing is completed
After continue to judge that dyeing effect, highest are redyed three times, three times after effect it is still undesirable, system be automatically labeled as " dyeing
Effect is poor, it is proposed that reforms sample print ".We select to redye three times.
It is as one embodiment, described that pretreatment specific steps are carried out to image are as follows:
Convert images into gray level image；Pass through the area-of-interest of one rectangle of setting position acquisition of information；By right
Area-of-interest does noise reduction and mean filter processing；
Using maximum variance between clusters threshold value, the foreground image of gray level image and background image are carried out by threshold value
Segmentation, the gray level image that gray value is less than threshold value is foreground image, and the gray level image that gray value is greater than threshold value is background image, is mentioned
Foreground image is taken out, the preprocessing process to image is completed.
As one embodiment, the maximum variance between clusters: note t is the segmentation threshold of foreground image and background image,
It is w0, average gray u0 that foreground image pixel number, which accounts for image scaled,；It is w1 that background image pixels points, which account for image scaled, is put down
Equal gray scale is u1,
The then overall average gray scale of image are as follows:
U=w0*u0+w1*u1； (1)
The difference of foreground image and background images:
G=w0* (u0-u) * (u0-u)+w1* (u1-u) * (u1-u)；
=w0*w1* (u0-u1) * (u0-u1)； (2)
Formula (2) is formula of variance.
When variance g maximum, it is believed that the difference of foreground image and background image is maximum at this time, and gray scale t at this time is best
Threshold value；
The gray level image of threshold value t be will be less than as foreground image, will be above the gray level image of threshold value t as background image,
The gray value that will be above threshold value t is all determined as 255, as white.
To also save its grayscale information while realizing adaptive threshold fuzziness.Number is provided for subsequent experimental
According to support.
As one embodiment, step (22) specific steps are as follows:
Positive round is found in foreground image by template matching；The positive round is standard round；A diameter of setting value；
By the template matching to positive round, qualitatively to determine the quality of dyeing effect: if can be looked in foreground image
To positive round, illustrate that dyeing effect is preferable；If can not find positive round, illustrate that dyeing effect is poor；Second is carried out to dye, and according to
The average gray value of foreground image determines the duration of second of dyeing；Assuming that the average gray value of foreground image is n, when dyeing total
A length of T；
Secondary dyeing duration t are as follows:
T=T*0.3* (256-n)；
After second is dyed, it is again introduced into step (21)；If the image of second of dyeing is matched in graphics template
It is good to be treated as coloration result for successful match in the process；If matching fails again, third time dyeing is carried out；And according to secondary
The calculation of duration determines the duration of third time dyeing after dyeing；
Similarly, if the image of third time dyeing successful match in graphics template matching process, is treated as coloration result
It is good；If it fails to match for third time dyeing, it is considered as dyeing failure, terminates.
Further, this method, which is aimed at, makes assessment to cytologic specimen print dyeing effect, by imitating to dyeing
The quantization of fruit is to determine operation performed by next step.It is namely carried out turning gray level image according to image, in gray level image upper-level threshold
Value segmentation, a suitable threshold value is determined by improved auto-thresholding algorithm, prospect and background segment are opened
Come.Find a positive round in the foreground by template matching.If we can find positive round at template matching, illustrate dyeing effect
Preferably.If can not find positive round, illustrate that dyeing effect is poor.The sample poor to dyeing effect carries out secondary dyeing, multiple every time
Dye is no more than percent the 30 of standard duration.The time of secondary dyeing is determined according to the average gray value of prospect.Gray level image
Amount to 256 gray levels, it is assumed that the average gray of prospect is n, and dyeing total duration is T, secondary dyeing duration are as follows: t=T*0.3*
(256-n).Continuing to judge that dyeing effect, highest are redyed three times after the completion of secondary dyeing, effect is still undesirable later three times,
System is automatically labeled as " dyeing effect is poor, it is proposed that reforms sample print ".
It is to be understood that we dyeing course be added to based on machine vision to cytologic specimen print dyeing effect
Identification, if recognition effect is preferable, so that it may assert that dyeing is completed, if dyeing effect is poor, we can calculate second
Dyeing time.Fortune in this way, realizes the closed-loop control to the dyeing effect of cytologic specimen print, to reach preferable
Dyeing effect.
As one embodiment, in dyeing, because the cell in sample print may occur during cell settlement
The case where being unevenly distributed, the situation will lead to that sample print is uneven by dyeing after stain effect, by camera
It is uneven for the grey value profile in target identification region in embodiment to image after Image Acquisition.
As shown in figure 3, the dye uniformity identification step, specifically includes:
Step (31): the average gray value g1 of foreground image area-of-interest is calculated；
Step (32): the highest gray value g2 of foreground image area-of-interest is calculated；
Step (33): it calculates large size packed cell degree G:G=(g1-g2)/256；
In the work such as concussion, the sedimentation of early period, sometimes because operation is lack of standardization, it will lead to cell mass and be difficult to uniformly beat
It dissipates, reflection is exactly large-scale packed cell, cytologic specimen of this large size packed cell in the later period into cytologic specimen print
It is difficult in the identification of print, stick to each other between them is stacked with, it is difficult to divide, the diagosis effect in image later period
Fruit, it is large-scale packed cell degree that we, which define G, then G=(g1-g2)/256.
Step (34): large-scale packed cell degree G is compared with given threshold, judges whether dyeing is uniform, if greatly
In given threshold, it is considered as dyeing unevenly, image is stored in dyeing speck database if uneven；Otherwise it is considered as contaminating
Color is uniform, terminates.
Through overtesting, when G is more than 0.2, we are it can be assumed that large-scale packed cell degree is higher, i.e. the agglomerating degree of cell
Height may will affect the diagosis effect in later period, be marked in a program, and the image information of the piece is stored in the agglomerating degree of cell
High database.
It is to be understood that it is attached by being deposited in specified region that the problem can be divided into cell during cell settlement
Effect is uneven or negligible amounts and in infall process because the problem of operation, cause cell on cell specimen slide attached
Position and normal bit be equipped with both problems of biggish deviation.For the non-uniform situation of adhesion effect after sedimentation, we are added
Identification to uneven situation is coloured, by the Uniformity Analysis algorithm of coloring effect come qualitative analysis coloring effect.
As one embodiment, during for cell settlement, the dyeing of deviation existing for sedimentation location and normal place
Defect, we have proposed cytologic specimen slide cell mass positional accuracy alignment algorithms:
As shown in figure 4, cell mass position identification step, specifically includes:
Step (41): to foreground image, fitting circle is formed using least square method；
Step (42): the distance of positioning fitting circle center location to slide left margin, right margin and lower boundary；
Step (43): calculating the left and right accuracy of cell attachment position, if left and right accuracy is greater than given threshold, table
Show that cell attachment position right-left error is big, just the dyeing speck database that slide information preservation is big to cell position deviation
In；If left and right accuracy is less than given threshold, then it represents that the accurate zero deflection of cell attachment position left and right directions terminates；
Step (44): calculating the accuracy up and down of cell attachment position, if accuracy is greater than given threshold, table up and down
Show that cell attachment position up and down direction deviation is big, just the dyeing speck database that slide information preservation is big to cell position deviation
In；If upper and lower accuracy is less than given threshold, then it represents that the accurate zero deflection of cell attachment position up and down direction terminates.
As one embodiment, step (42) specific steps are as follows:
The fitting circle center of circle to the upright position of slide left margin be h1, the fitting circle center of circle to the upright position of slide right margin
For h2, the distance in the fitting circle center of circle to slide lower boundary is h3；
As one embodiment, the specific steps of the left and right accuracy of cell attachment position are calculated are as follows:
The left and right accuracy A1 of cell attachment position:
A1=(h1-h2)/(h1+h2) * 100%；
As one embodiment, the specific steps of the accuracy up and down of cell attachment position are calculated are as follows:
The accuracy A2 up and down of cell attachment position:
A2=(h3-20)/20*100%.
, may be because of the maloperation of operator in infall process, the physical location and ideal that cause cell to adhere to
There are biggish deviations for position.On the basis of dyeing successfully, we can detect fitting circle by image procossing, by several
We can find the center of circle of fitting circle to what form, then by edge detection algorithm, we can navigate to the left and right of glass slide
Two boundaries and lower boundary.It is h1 that we, which define the fitting center of circle to the upright position of left margin, and right margin is arrived in the fitting center of circle
Upright position is h2, we define cell attachment position or so accuracy be A1, A1=(h1-h2)/(h1+h2) * 100%, when
When A1<0.15, it is believed that cell attachment position is substantially accurate in left and right directions, and as A1>0.15, program prompts cell attachment
Point position right-left error is larger, and the piece information is passed to the dyeing speck database of cell attachment position deviation.
Similarly, the distance that we define the fitting center of circle to lower boundary is h3, arrives lower boundary by multiple multi-standard position
The measurement of vertical range, it is believed that 20mm is optimum position, we define cell attachment position, and accuracy is A2, A2=up and down
(h3-20)/20*100%, when A2<0.15, it is believed that cell attachment position is substantially accurate in above-below direction, works as A1>0.15
When, program prompts cell attachment sites position up and down direction deviation larger, and the piece information is passed to cell attachment position deviation
Dyeing speck database.
It is to be understood that we are thin using cytologic specimen slide for the position of the cell mass on cytologic specimen slide
The physical location accuracy alignment algorithm of born of the same parents' attachment, by the center location after Image Acquisition by pretreated fitting circle
Positioning, and normal place comparison, to determine the offset in the center of circle.Offset can assert that sedimentation location is deposited beyond prescribed limit
In deviation, it is labeled in program and is stored in database.We are directed to medically common decanter type film-making mode, system
Can the effect of piece be standard circular, we identify circle, by recognize geometric circular, qualitatively to analyze dyeing
Effect.
Embodiment two: the cytologic specimen print dyeing effect detection system based on machine vision is provided；
Cytologic specimen print dyeing effect detection system based on machine vision, comprising:
Image processing module: Image Acquisition carries out image procossing to cytologic specimen print, identifies coloration result；If
Dyeing effect is good, enters dye uniformity identification step；If dyeing effect is poor, second just is determined according to processing result image
The time of secondary dyeing；If continuously three times dyeing effect be all it is poor, be considered as dyeing failure, using cytologic specimen print as give up
Piece processing；
Dye uniformity identification module: being calculated using uniformity algorithm, even dyeing caused by identification cell is agglomerating
Degree；
Cell mass location identification module: the physical location alignment algorithm adhered to using cytologic specimen slide cell, to thin
Born of the same parents' cumularsharolith, which is set, to be calculated, and identifies cell mass position.
Embodiment three: a kind of electronic equipment is provided；
A kind of electronic equipment, comprising: the meter that memory, processor and storage execute on a memory and on a processor
The instruction of calculation machine, when the computer instruction is run by processor, completes step described in method in above-mentioned first embodiment.
Example IV: a kind of computer readable storage medium is provided；
A kind of computer readable storage medium, is stored thereon with computer instruction, and the computer instruction is transported by processor
When row, step described in method in above-mentioned first embodiment is completed.
Can exclude it is nearly all because dyeing duration not enough caused by dyeing effect it is poor.To guarantee dyeing effect
Formedness and consistency provide high-quality sample for the pathology detection in later period.For the cytologic specimen of final dyeing effect difference
Print, we remind user in a program and automatically save the record.The series print can be arranged to by unification by mechanical arm useless
Section, no longer progress next step operation.
The foregoing is merely preferred embodiment of the present application, are not intended to limit this application, for the skill of this field
For art personnel, various changes and changes are possible in this application.Within the spirit and principles of this application, made any to repair
Change, equivalent replacement, improvement etc., should be included within the scope of protection of this application.
Claims (10)
1. the cytologic specimen print dyeing effect detection method based on machine vision, characterized in that include:
Image processing step: Image Acquisition carries out image procossing to cytologic specimen print, identifies coloration result；If dyeing
Effect is good, enters dye uniformity identification step；If dyeing effect is poor, just determine to contaminate for second according to processing result image
The time of color；If continuously three times dyeing effect be all it is poor, be considered as dyeing fail, using cytologic specimen print as waste paper at
Reason；
Dye uniformity identification step: being calculated using uniformity algorithm, dye uniformity caused by identification cell is agglomerating；
Cell mass position identification step: the physical location alignment algorithm adhered to using cytologic specimen slide cell, to cell mass
Position is calculated, and identifies cell mass position.
2. the method as described in claim 1, characterized in that further include pre-treatment step before described image processing step；
The pre-treatment step, specifically includes: cytologic specimen print is placed in glass frame, mechanical arm grabs glass frame,
Glass frame is placed into BS fixer M minutes, is placed into after then cleaning glass frame in acid hydrolysis solution N minutes；Then by glass
It is placed into dyeing liquor L minutes after horse cleaning；Take out glass frame；
After the identification step of the cell mass position, further includes: dehydration；The dehydration, specifically includes: to glass frame
After being cleaned, serial dehydration is carried out to print, dyeing is completed.
3. the method as described in claim 1, characterized in that described image processing step specifically includes:
Step (21): acquisition cytologic specimen print image, image is pre-processed, divide the image into foreground image and
Background image；
Step (22): carrying out graphics template matching to pretreated image, judge whether successful match,
If successful match, it is good to be treated as coloration result；If it fails to match, carries out second and dye, and according to foreground picture
The average gray value of picture determines the duration of second of dyeing；After second is dyed, it is again introduced into step (21)；
If the image of second of dyeing successful match in graphics template matching process, it is good to be treated as coloration result；If
With failing again, then third time dyeing is carried out；
If the image of third time dyeing successful match in graphics template matching process, it is good to be treated as coloration result；If the
It fails to match for dyeing three times, then is considered as dyeing failure, terminates.
4. method as claimed in claim 3, characterized in that described to carry out pretreatment specific steps to image are as follows:
Convert images into gray level image；Pass through the area-of-interest of one rectangle of setting position acquisition of information；By emerging to sense
Do noise reduction and mean filter processing in interesting region；
Using maximum variance between clusters threshold value, the foreground image of gray level image and background image are divided by threshold value
It cuts, the gray level image that gray value is less than threshold value is foreground image, and the gray level image that gray value is greater than threshold value is background image, is extracted
Foreground image out completes the preprocessing process to image.
5. the method as described in claim 1, characterized in that step (22) specific steps are as follows:
Positive round is found in foreground image by template matching；The positive round is standard round；A diameter of setting value；
If positive round can be found in foreground image, illustrate that dyeing effect is preferable；If can not find positive round, illustrate dyeing effect compared with
Difference；It carries out second to dye, and determines the duration of second of dyeing according to the average gray value of foreground image；Assuming that foreground image
Average gray value be n, dyeing total duration be T；
Secondary dyeing duration t are as follows:
T=T*0.3* (256-n)；
After second is dyed, it is again introduced into step (21)；If the image of second of dyeing is in graphics template matching process
It is good to be treated as coloration result for middle successful match；If matching fails again, third time dyeing is carried out；And according to secondary dyeing
The calculation of duration determines the duration of third time dyeing afterwards；
Similarly, if the image of third time dyeing successful match in graphics template matching process, it is good to be treated as coloration result；Such as
It fails to match for the dyeing of fruit third time, then is considered as dyeing failure, terminates.
6. method as claimed in claim 4, characterized in that the dye uniformity identification step specifically includes:
Step (31): the average gray value g1 of foreground image area-of-interest is calculated；
Step (32): the highest gray value g2 of foreground image area-of-interest is calculated；
Step (33): it calculates large size packed cell degree G:G=(g1-g2)/256；
Step (34): large-scale packed cell degree G is compared with given threshold, judges whether dyeing is uniform, if it is greater than setting
Determine threshold value, is considered as dyeing unevenly, image is stored in dyeing speck database if uneven；Otherwise it is equal to be considered as dyeing
It is even, terminate.
7. the method as described in claim 1, characterized in that cell mass position identification step specifically includes:
Step (41): to foreground image, fitting circle is formed using least square method；
Step (42): the distance of positioning fitting circle center location to slide left margin, right margin and lower boundary；
Step (43): calculating the left and right accuracy of cell attachment position, if left and right accuracy is greater than given threshold, then it represents that thin
Born of the same parents' attachment position right-left error is big, just by slide information preservation into the big dyeing speck database of cell position deviation；
If left and right accuracy is less than given threshold, then it represents that the accurate zero deflection of cell attachment position left and right directions terminates；
Step (44): calculating the accuracy up and down of cell attachment position, if accuracy is greater than given threshold up and down, then it represents that thin
Born of the same parents' attachment position up and down direction deviation is big, just by slide information preservation into the big dyeing speck database of cell position deviation；
If upper and lower accuracy is less than given threshold, then it represents that the accurate zero deflection of cell attachment position up and down direction terminates.
8. the cytologic specimen print dyeing effect detection system based on machine vision, characterized in that include:
Image processing module: Image Acquisition carries out image procossing to cytologic specimen print, identifies coloration result；If dyeing
Effect is good, enters dye uniformity identification step；If dyeing effect is poor, just determine to contaminate for second according to processing result image
The time of color；If continuously three times dyeing effect be all it is poor, be considered as dyeing fail, using cytologic specimen print as waste paper at
Reason；
Dye uniformity identification module: being calculated using uniformity algorithm, dye uniformity caused by identification cell is agglomerating；
Cell mass location identification module: the physical location alignment algorithm adhered to using cytologic specimen slide cell, to cell mass
Position is calculated, and identifies cell mass position.
9. a kind of electronic equipment, characterized in that include: memory, processor and storage on a memory and on a processor
The computer instruction of execution when the computer instruction is run by processor, is completed described in any one of claim 1-7 method
Step.
10. a kind of computer readable storage medium, characterized in that be stored thereon with computer instruction, the computer instruction quilt
When processor is run, step described in any one of claim 1-7 method is completed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910164297.8A CN109781747B (en) | 2019-03-05 | 2019-03-05 | Machine vision-based cytological specimen sample wafer dyeing effect detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910164297.8A CN109781747B (en) | 2019-03-05 | 2019-03-05 | Machine vision-based cytological specimen sample wafer dyeing effect detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109781747A true CN109781747A (en) | 2019-05-21 |
| CN109781747B CN109781747B (en) | 2024-05-10 |
Family
ID=66486704
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910164297.8A Active CN109781747B (en) | 2019-03-05 | 2019-03-05 | Machine vision-based cytological specimen sample wafer dyeing effect detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109781747B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110132695A (en) * | 2019-05-27 | 2019-08-16 | 福州迈新生物技术开发有限公司 | A kind of pathological staining system reduction human error is caused to dye abnormal method |
| CN110823666A (en) * | 2019-11-27 | 2020-02-21 | 李雄 | Application of hematoxylin in preparation of staining agent for reducing color fading of pathological conventional section after staining |
| CN113515077A (en) * | 2021-04-23 | 2021-10-19 | 重庆德方信息技术有限公司 | System and method for monitoring human body cell staining process |
| CN114235803A (en) * | 2021-12-14 | 2022-03-25 | 安图实验仪器(郑州)有限公司 | Decolorization determination method used after microbial staining |
| CN117147266A (en) * | 2023-10-31 | 2023-12-01 | 广州医科大学附属第一医院（广州呼吸中心） | Blood disease analysis system based on blood morphology special dyeing |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5625706A (en) * | 1995-05-31 | 1997-04-29 | Neopath, Inc. | Method and apparatus for continously monitoring and forecasting slide and specimen preparation for a biological specimen population |
| US20040253616A1 (en) * | 2003-06-12 | 2004-12-16 | Cytyc Corporation | Method and system of determining the stain quality of slides using a scatter plot distributions |
| US20130301898A1 (en) * | 2012-05-11 | 2013-11-14 | Rohit Jain | Method and Apparatus for Image Scoring and Analysis |
| WO2014181347A2 (en) * | 2013-04-23 | 2014-11-13 | Aditya Imaging Information Technologies | Method and system for processing a digital slide image of a stained tissue |
| US20160321512A1 (en) * | 2015-04-30 | 2016-11-03 | Ge Healthcare Bio-Sciences Corp. | Automated staining and segmentation quality control |
| CN106645153A (en) * | 2016-12-23 | 2017-05-10 | 麦克奥迪（厦门）医疗诊断系统有限公司 | Fast staining and detecting method for cell pathological sections |
| US20170178361A1 (en) * | 2015-12-18 | 2017-06-22 | Abbott Laboratories | Methods and Systems for Assessing Histological Stains |
-
2019
- 2019-03-05 CN CN201910164297.8A patent/CN109781747B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5625706A (en) * | 1995-05-31 | 1997-04-29 | Neopath, Inc. | Method and apparatus for continously monitoring and forecasting slide and specimen preparation for a biological specimen population |
| US20040253616A1 (en) * | 2003-06-12 | 2004-12-16 | Cytyc Corporation | Method and system of determining the stain quality of slides using a scatter plot distributions |
| US20130301898A1 (en) * | 2012-05-11 | 2013-11-14 | Rohit Jain | Method and Apparatus for Image Scoring and Analysis |
| WO2014181347A2 (en) * | 2013-04-23 | 2014-11-13 | Aditya Imaging Information Technologies | Method and system for processing a digital slide image of a stained tissue |
| US20160321512A1 (en) * | 2015-04-30 | 2016-11-03 | Ge Healthcare Bio-Sciences Corp. | Automated staining and segmentation quality control |
| US20170178361A1 (en) * | 2015-12-18 | 2017-06-22 | Abbott Laboratories | Methods and Systems for Assessing Histological Stains |
| CN109121437A (en) * | 2015-12-18 | 2019-01-01 | 雅培实验室 | For assessing the method and system of histological stain |
| CN106645153A (en) * | 2016-12-23 | 2017-05-10 | 麦克奥迪（厦门）医疗诊断系统有限公司 | Fast staining and detecting method for cell pathological sections |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110132695A (en) * | 2019-05-27 | 2019-08-16 | 福州迈新生物技术开发有限公司 | A kind of pathological staining system reduction human error is caused to dye abnormal method |
| CN110132695B (en) * | 2019-05-27 | 2022-01-11 | 福州迈新生物技术开发有限公司 | Method for reducing abnormal dyeing caused by human error by pathological dyeing system |
| CN110823666A (en) * | 2019-11-27 | 2020-02-21 | 李雄 | Application of hematoxylin in preparation of staining agent for reducing color fading of pathological conventional section after staining |
| CN113515077A (en) * | 2021-04-23 | 2021-10-19 | 重庆德方信息技术有限公司 | System and method for monitoring human body cell staining process |
| CN114235803A (en) * | 2021-12-14 | 2022-03-25 | 安图实验仪器(郑州)有限公司 | Decolorization determination method used after microbial staining |
| CN114235803B (en) * | 2021-12-14 | 2024-05-14 | 安图实验仪器(郑州)有限公司 | Decoloring judgment method for microorganism after dyeing |
| CN117147266A (en) * | 2023-10-31 | 2023-12-01 | 广州医科大学附属第一医院（广州呼吸中心） | Blood disease analysis system based on blood morphology special dyeing |
| CN117147266B (en) * | 2023-10-31 | 2024-02-13 | 广州医科大学附属第一医院（广州呼吸中心） | Blood disease analysis system based on blood morphology special dyeing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109781747B (en) | 2024-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109781747A (en) | Cytologic specimen print dyeing effect detection method and system based on machine vision | |
| CN108074243B (en) | Cell positioning method and cell segmentation method | |
| CN111709948B (en) | Method and device for detecting defects of container | |
| CN110705403A (en) | Cell sorting method, cell sorting device, cell sorting medium, and electronic apparatus | |
| CN113421230B (en) | Visual detection method for defects of vehicle-mounted liquid crystal display light guide plate based on target detection network | |
| CN110728681B (en) | Mura defect detection method and device | |
| JP6791245B2 (en) | Image processing device, image processing method and image processing program | |
| CN106295789A (en) | A kind of crop seed method of counting based on image procossing | |
| CN112990015B (en) | Automatic identification method and device for lesion cells and electronic equipment | |
| CN111476307B (en) | Lithium battery surface defect detection method based on depth field adaptation | |
| CN109002831A (en) | A kind of breast density classification method, system and device based on convolutional neural networks | |
| CN113256577A (en) | Cancer auxiliary analysis system and device based on HE staining pathological image | |
| CN112393880A (en) | Screen replacement detection method and device | |
| CN110135407A (en) | Sample mask method and computer storage medium | |
| CN215179046U (en) | Full-automatic liquid-based cell slide making, dyeing, mounting and reading integrated machine | |
| CN109584211A (en) | A kind of vision automatic testing method of animal oocyte polar body | |
| CN117309892B (en) | Defect detection method, device and system for blue film of battery and light source controller | |
| CN117256514B (en) | Method for removing unqualified hatching eggs for sorting system | |
| CN105092589A (en) | Detection method for defects of capsule head | |
| CN115701868B (en) | Domain self-adaptive enhancement method applicable to various visual tasks | |
| Dixon et al. | Measuring optical distortion in aircraft transparencies: a fully automated system for quantitative evaluation | |
| CN115908423A (en) | Fuzzy silk defect image detection method based on weak supervised learning | |
| CN118052814B (en) | AI technology-based full-automatic specimen pretreatment system and method | |
| De Vitis et al. | A technique to reduce the processing time of defect detection in glass tubes | |
| CN114882268A (en) | Deep learning-based paddy damage detection method and system and storage medium |
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 | ||
| CP03 | Change of name, title or address |
Address after: 250353 University Road, Changqing District, Ji'nan, Shandong Province, No. 3501 Patentee after: Qilu University of Technology (Shandong Academy of Sciences) Country or region after: China Address before: 250353 University Road, Changqing District, Ji'nan, Shandong Province, No. 3501 Patentee before: Qilu University of Technology Country or region before: China |
|
| CP03 | Change of name, title or address |