CN105241811B - Multi-level focus adopts drawing method and system automatically - Google Patents
Multi-level focus adopts drawing method and system automatically Download PDFInfo
- Publication number
- CN105241811B CN105241811B CN201510641966.8A CN201510641966A CN105241811B CN 105241811 B CN105241811 B CN 105241811B CN 201510641966 A CN201510641966 A CN 201510641966A CN 105241811 B CN105241811 B CN 105241811B
- Authority
- CN
- China
- Prior art keywords
- image
- focusing
- length
- target
- layer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000001514 detection method Methods 0.000 claims abstract description 40
- 230000004927 fusion Effects 0.000 claims abstract description 13
- 239000000725 suspension Substances 0.000 claims description 52
- 238000012360 testing method Methods 0.000 claims description 7
- 235000013399 edible fruits Nutrition 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 4
- 238000007689 inspection Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000002203 pretreatment Methods 0.000 description 1
- 239000012898 sample dilution Substances 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Landscapes
- Microscoopes, Condenser (AREA)
- Automatic Focus Adjustment (AREA)
Abstract
The present invention provides a kind of focus at many levels and adopts drawing method and system automatically, tally is focused according to preset search range, search the first focusing surface, Image Acquisition is carried out to first focusing surface, obtain first object image, obtain focusing step-length, according to the focusing step-length, figure figure layer is adopted in setting, each image for adopting figure figure layer is acquired respectively, obtain the second target image group, by the image co-registration in the first object image and the second target image group, remove redundancy object and unintelligible target, and to remaining Motion parameters and classification, obtain sample object classification and Detection result.In whole process, different target image is obtained using multi-focusing, and fusion realization is carried out to different target image and accurately adopts figure compared with multisample to impure, it can be accurately to target identification in sample and counting.
Description
Technical field
The present invention relates to image acquisition technology fields, and drawing method and system are adopted automatically more particularly to multi-level focusing.
Background technology
In modern medical techniques field, in order to check that patient condition usually requires to carry out a variety of item detections, in these inspections
Have in survey project and be greatly related to the collecting sample detection from patient, such as the detection of common stool.
Before carrying out visible component detection to these samples, generally requires and use the equipment for producing pattern detection liquid to sample
The operation for being diluted, filtering and sampling is to obtain the suspension detected for visible component.Then suspension is filled with tally
In, microscope is amplified sample in tally, and image collecting device carries out Image Acquisition, pattern recognition device pair to sample
The image of acquisition carries out target identification, to obtain testing result.
Above-mentioned sample is there are many type, and impurity is more in the suspension (such as stool sampl suspension) of some samples, various mesh
Target density and size are different, and target to be detected is randomly dispersed in suspension.After suspension is filled in tally, mesh to be checked
Mark cannot sink to tally bottom completely, sink to the target of tally bottom completely, due to of different sizes, focussing plane also differs
It causes.When image collecting device is focused bat figure to sample to be tested, a part of target can only be focused, it is other to be not located at same plane
Target can not complete to focus.In the picture for causing shooting, partial target is clear, and partial target is fuzzy.Some targets can not clap
It takes the photograph, target identification can not be completed well and counted.
Invention content
Based on this, it is necessary to compared with multisample can not accurately adopt figure and ask to impure for figure mode is generally adopted automatically
Topic provides a kind of focus at many levels and adopts drawing method and system automatically, and realization accurately adopts figure to impure compared with multisample, with accurate right
Target identification and counting in sample.
A kind of focus at many levels adopts drawing method, including step automatically：
Tally is focused according to preset search range, searches the first focusing surface, image is carried out to first focusing surface
Acquisition obtains first object image, wherein the tally is filled with sample suspensions；
Obtain focusing step-length；
According to the focusing step-length, figure figure layer is adopted in setting, is acquired each image for adopting figure figure layer respectively, is obtained second
Target image group；
Acquired target image is merged, removes redundancy object and unintelligible target, and know automatically to remaining target
Not and classify, obtains sample object classification and Detection result, wherein the acquired target image includes the first object
Image in image and the second target image group.
A kind of focus at many levels adopts drawing system automatically, including：
First object image collection module, for being focused to tally according to preset search range, the first focusing surface of lookup,
Image Acquisition is carried out to first focusing surface, obtains first object image, wherein the tally is filled with sample suspensions；
Focusing step-length acquisition module, for obtaining focusing step-length；
Second target image group acquisition module, for according to the focusing step-length, setting to be adopted figure figure layer, acquired respectively each
The image for adopting figure figure layer obtains the second target image group；
Fusion Module removes redundancy object and unintelligible target, and to surplus for merging acquired target image
Remaining Motion parameters and classification obtain sample object classification and Detection result, wherein the acquired target image includes
Image in the first object image and the second target image group.
The present invention focuses adopt drawing method and system automatically at many levels, is focused, is searched to tally according to preset search range
First focusing surface carries out Image Acquisition to first focusing surface, obtains first object image, focusing step-length is obtained, according to institute
Focusing step-length is stated, figure figure layer is adopted in setting, is acquired each image for adopting figure figure layer respectively, is obtained the second target image group, will
The first object image and the image co-registration in the second target image group, remove redundancy object and unintelligible target,
And to remaining Motion parameters and classification, obtain sample object classification and Detection result.In whole process, using multi-focusing
Different target image is obtained, and fusion realization is carried out to different target image and accurately adopts figure, Neng Gouzhun compared with multisample to impure
Really to target identification in sample and counting.
Description of the drawings
Fig. 1, which is that the present invention is multi-level, focuses the flow diagram for adopting drawing method one embodiment automatically；
Fig. 2, which is that the present invention is multi-level, focuses the flow diagram for adopting second embodiment of drawing method automatically；
Fig. 3, which is that the present invention is multi-level, focuses the structural schematic diagram for adopting drawing system one embodiment automatically；
Fig. 4, which is that the present invention is multi-level, focuses the structural schematic diagram for adopting second embodiment of drawing system automatically.
Specific implementation mode
As shown in Figure 1, a kind of multi-level focus adopts drawing method, including step automatically：
S200：Tally is focused according to preset search range, the first focusing surface is searched, first focusing surface is carried out
Image Acquisition obtains first object image, wherein the tally is filled with sample suspensions.
Preset search range is to be based on the preset value range of detection demand, the lookup for the first focusing surface,
We can be by focusing automatically equipment, the microscope that focuses automatically can be used for example to search the first focusing surface, carry out
Image Acquisition obtains first object image.It is non-essential in a wherein concrete operations example, execute step S200 into
Some preparations of row, such as acquired from patient firstly the need of by sample, later by Sample Dilution, suspension is made, then will
Sample suspensions are filled with tally, and tally is placed under microscope, are amplified to sample suspensions on tally by microscope.On
The focusing stated can upwards be carried out since tally bottom, can also from the surface of sample suspensions to tally bottom into
Row.
S400：Obtain focusing step-length.
Various ways acquisition, such as type that can be according to current detection sample suspensions and inspection may be used in focusing step-length
The calculating such as survey demand obtain, and can also be to directly read default focusing step-length to obtain, default focusing step-length is to be based on historical experience
Data are preset.
Step S400 is specially in one of the embodiments,：Detect the image ginseng of all kinds of targets in first object image
Number determines focusing step-length according to the image parameter of all kinds of targets in first object image.
The image parameter of target can specifically include target sharpness, target morphology and target sizes, according to these figures
As parameter, focusing step-length is obtained.More specifically, target sharpness, target morphology and target sizes can be directed to carry out
Weighted calculation obtains focusing step-length further according to weighing computation results and default detection demand.Using weighted calculation mode, examine
Consider influence of the different images parameter to step-length of focusing, it is clear to be obtained in subsequent operation so as to accurately obtain focusing step-length
Clear target image.
Step S400 is specially in one of the embodiments,：Read preset focusing step-length.Preset focusing step-length is
It is preset, the size of each target, default detection demand and historical experience in the foundation first object image of setting
Data.
S600：According to the focusing step-length, figure figure layer is adopted in setting, is acquired each image for adopting figure figure layer respectively, is obtained
Obtain the second target image group.
As the foregoing description, the density of various targets is in different size in sample suspensions, target can in suspension different positions
It sets suspension or is settled in bottom, if shooting single figure layer image merely, can lead to have target shooting unintelligible or shooting
Less than.In this regard, herein, according to the focusing step-length, figure figure layer is adopted in setting, and adopting figure figure layer can be provided with multiple, adopt respectively
Collect each image for adopting figure figure layer, obtains the second target image group.
S800：Acquired target image is merged, removes redundancy object and unintelligible target, and certainly to remaining target
Dynamic identification and classification, obtain sample object classification and Detection result, wherein acquired target image includes the first object
Image in image and the second target image group.
That there are certain targets is unintelligible for image in first object image and the second target image group, certain targets are multiple
Repeated acquisition, in this regard, herein by image (the acquired target figure in first object image and the second target image group
Picture) fusion, redundancy object and unintelligible target are removed, and to remaining Motion parameters and classification, obtain sample object
Classification and Detection result.Related software may be used for above-mentioned removal redundancy object and unintelligible target or image real time transfer is set
It is standby to be operated.For Motion parameters and classification, the trained nerve based on morphological image feature can be utilized
Network classifier is handled.
The present invention focuses adopt drawing method automatically at many levels, is focused to tally according to preset search range, it is poly- to search first
Focal plane carries out Image Acquisition to first focusing surface, obtains first object image, focusing step-length is obtained, according to the focusing
Figure figure layer is adopted in step-length, setting, is acquired each image for adopting figure figure layer respectively, is obtained the second target image group, by described the
One target image and the image co-registration in the second target image group, remove redundancy object and unintelligible target, and to surplus
Remaining Motion parameters and classification obtain sample object classification and Detection result.In whole process, obtained not using multi-focusing
Same target image, and fusion realization is carried out to different target image and accurately adopts figure compared with multisample to impure, it can be accurately to sample
Target identification and counting in this.
Described according to the focusing step-length in one of the embodiments, figure figure layer is adopted in setting, is acquired respectively each described
The step of adopting the image of figure figure layer, obtaining the second target image group specifically includes：
The position on the basis of the position of first focusing surface is extended the default areas Cai Tu with upper and lower both direction respectively
Between, it is described it is default adopt in figure section, according to the focusing step-length, figure figure layer is adopted in setting, wherein described default to adopt figure section small
In tally bottom to the section on the sample suspensions surface；
Each image for adopting figure figure layer is acquired, the second target image group is obtained.
In the present embodiment, the position of the first focusing surface is obtained first, later on the basis of the first focusing surface position,
Nearby extended preset of both direction adopts figure section up and down for one focusing surface position, is walked according to focusing in default adopt in figure section
Long, figure figure layer is adopted in setting, acquires each image for adopting figure figure layer, obtains the second target image group.For example, we can first obtain
First focusing surface position, later on the basis of the first focusing surface position, both direction is according to tune above and below the first focusing surface position
Burnt step-length respectively setting 5 adopt figure figure layer, i.e., at this time obtain 10 images for adopting figure figure layer, the second target image group i.e. include this 10
A image for adopting figure figure layer.It should be pointed out that in the present embodiment, default figure section of adopting is the good of advance planning setting,
Much smaller than tally bottom to the section on sample suspensions surface.More specifically, it is one smaller to preset and adopt figure section
Value adopts figure figure layer in the setting of the neighbouring position up and down of the first focusing surface position.
Described according to the focusing step-length in one of the embodiments, figure figure layer is adopted in setting, is acquired respectively each described
The image of figure figure layer is adopted, obtaining the second target image group step further includes later：
Read default second focusing step-length, wherein the default second focusing step-length determination process is specifically, acquisition is filled with
The first position of tally bottom target focus layer is fallen in the sample suspensions of tally and is suspended in tally top target
The second position of focus layer, the corresponding first position of the multiple sample suspensions of repeated acquisition and the second position, statistics obtain
First position group and second position group are obtained, according to the first position group and the second position group, obtains described default second
Focusing step-length；
In the station acquisition image from first focusing surface at a distance of the default second focusing step-length, third target is obtained
Image；
It is described to merge acquired target image, redundancy object and unintelligible target are removed, and certainly to remaining target
It is dynamic to identify and be specially the step of classifying, obtain sample object classification and Detection result：
By in the first object image, the second target image group image and the third target image melt
It closes, removes redundancy object and unintelligible target, and to remaining Motion parameters and classification, obtain sample object classification inspection
Survey result.
Default second focusing step-length is different from default focusing step-length above-mentioned, and specifically, default second focusing step-length is
According to the preset value that test of many times determines, the experiment process is specifically, first, acquisition is filled in the sample suspensions of tally and settles
In the first position of tally bottom target focus layer and the second position for being suspended in tally top target focus layer, later,
The corresponding first position of the multiple sample suspensions of repeated acquisition and the second position, statistics obtain first position group and second position group,
Finally, according to first position group and second position group, default second focusing step-length, during above-mentioned experiment, first position are obtained
The automatic type of focusing may be used with the second position to determine, multiple sample suspensions of selection can be the sample suspensions of same type
(such as be blood sample suspension or be stool sampl suspension).It can be counted after obtaining first position group and second position group
Calculate the average value of the average value of first position and the second position in the group of the second position in the group of first position, then putting down first position
The average value of mean value and the second position subtracts each other and takes absolute value, and obtains default second focusing step-length.Obtain the second default focusing step
After length on the first focusing surface, image is acquired according to default second focusing step-length, third target image is obtained, has adopted at this time
Collection obtains first object image, the second target image group and third target image, is carrying out subsequent image fusing stage, will
Three kinds of images are merged, and remove redundancy object and unintelligible target, and to remaining Motion parameters and classification, obtain
Obtain sample object classification and Detection result.Consider that target density is different again, target is suspended in sample suspensions different location and is easy shadow
Ring that adopt partial target in figure image unintelligible or the problem of can not take so that the sample obtained after final fusion treatment
Classification and Detection result is more accurate.
Described according to the focusing step-length in one of the embodiments, figure figure layer is adopted in setting, is acquired respectively each described
The image of figure figure layer is adopted, the second target image group step is obtained and specifically includes：
The position on the basis of tally bottom, in the tally bottom to sample suspensions surface section, according to
Figure figure layer is adopted in the focusing step-length, at equal intervals setting；
Each image for adopting figure figure layer is acquired respectively, obtains the second target image group.
In specific example, according to pre-treatment determine focusing step-length, on the basis of tally bottom from bottom to top (from
Tally bottom is to close to microscopical direction) at equal intervals setting adopt figure figure layer, acquire each described adopt figure figure layer respectively
Image obtains the second target image group.Focusing step-length can be understood as a fixed numerical value herein, be arrived in tally bottom
In the zone distance on suspension surface, sets multiple adopt figure figure layer at equal intervals.
As shown in Fig. 2, further including after step S200 in one of the embodiments,：
S300：Fog-level detection is carried out to all targets in the first object image, detects the first object figure
It whether there is unintelligible target image as in.
Fog-level detection is carried out to all targets in first object image, when there are unintelligible in first object image
When target image, illustrates not realizing to the classification and Detection of sample object based on first object image, need to focus again
The image of different figure layers is acquired to be handled, in this regard, entering step S400；When there is no unintelligible mesh in first object image
When logo image, illustrate to realize the classification and Detection to sample object according to first object image, to improve sample object classification
The efficiency of detection directly carries out Motion parameters and classification, obtains the operation of sample object classification and Detection result.
In addition, as shown in figure 3, adopting drawing method automatically based on above-mentioned multi-level focusing, the present invention also provides a kind of multi-level
Focusing adopts drawing system automatically, including：
First object image collection module 100 is searched first and is focused for being focused to tally according to preset search range
Face carries out Image Acquisition to first focusing surface, obtains first object image；
Step-length of focusing acquisition module 200, for obtaining focusing step-length；
Second target image group acquisition module 300, for according to the focusing step-length, setting to be adopted figure figure layer, acquired respectively
Each image for adopting figure figure layer, obtains the second target image group；
Fusion Module 400, for acquired target image to be merged, removal redundancy object and unintelligible target, and
To remaining Motion parameters and classification, sample object classification and Detection result is obtained, wherein the acquired target image
Including the image in the first object image and the second target image group.
The present invention focuses adopt drawing system automatically at many levels, and first object image collection module 100 is according to preset search range
Tally is focused, the first focusing surface is searched, Image Acquisition is carried out to first focusing surface, obtains first object image, is adjusted
Burnt step-length acquisition module 200 obtains focusing step-length, and the second target image group acquisition module 300 is according to the focusing step-length, setting
Figure figure layer is adopted, acquires each image for adopting figure figure layer respectively, obtains the second target image group, Fusion Module 400 will be described
First object image and the image co-registration in the second target image group, remove redundancy object and unintelligible target, and right
Remaining Motion parameters and classification obtain sample object classification and Detection result.In whole process, obtained using multi-focusing
Different target image, and fusion realization is carried out to different target image and accurately adopts figure compared with multisample to impure, it can be accurate right
Target identification and counting in sample.
The focusing step-length acquisition module 200 is specifically used in detection first object image in one of the embodiments,
The image parameter of all kinds of targets determines focusing step-length according to the image parameter of all kinds of targets in first object image.Wherein, scheme
As parameter includes target sharpness, target morphology and target sizes.
The focusing step-length acquisition module 200 is specifically used for reading preset focusing step-length in one of the embodiments,.
Preset focusing step-length is preset, the size of each target, default detection in the foundation first object image of setting
Demand and historical empirical data.
The second target image group acquisition module 300 specifically includes in one of the embodiments,：
First adopts figure figure layer setup unit, for the position on the basis of the position of first focusing surface, respectively with upper and lower
Both direction, which is extended to preset, adopts figure section, is adopted in figure section default, and according to the focusing step-length, figure figure layer is adopted in setting,
Wherein, the default section adopted figure section and be less than tally bottom to the sample suspensions surface；
First collecting unit obtains the second target image group for acquiring each image for adopting figure figure layer.
The multi-level focusing adopts drawing system and further includes automatically in one of the embodiments,：
Second focusing step-length acquisition module, for reading default second focusing step-length, wherein the default second focusing step
Long determination process is filled with first that tally bottom target focus layer is fallen in the sample suspensions of tally specifically, acquiring
The second position of tally top target focus layer is set and is suspended in, the multiple sample suspensions of repeated acquisition are described first corresponding
Set with the second position, statistics obtains first position group and second position group, according to the first position group and described second
Set of locations obtains the default second focusing step-length；
Third target image acquisition module, for from first focusing surface at a distance of the default second focusing step-length
Station acquisition image obtains third target image；
The Fusion Module is used for image in the first object image, the second target image group and described
Third target image merges, and removes redundancy object and unintelligible target, and to remaining Motion parameters and classification, obtain
Sample object classification and Detection result.
The second target image group acquisition module 300 specifically includes in one of the embodiments,：
Second adopts figure figure layer setup unit, for the position on the basis of tally bottom, in tally bottom to the sample
In the section of this suspension surface, according to the focusing step-length, figure figure layer is adopted in setting at equal intervals；
Second collecting unit obtains the second target image group for acquiring each image for adopting figure figure layer respectively.
As shown in figure 4, the multi-level focusing adopts drawing system and further includes automatically in one of the embodiments,：
Detection module 500, for carrying out fog-level detection to all targets in the first object image, described in detection
It whether there is unintelligible target image in first object image.
In order to further explain in detail it is of the invention multi-level focus adopt automatically drawing method and system technical solution and its
Below whole process will be discussed in detail using two examples by taking stool sampl as an example in the advantageous effect brought.
Example one：
Step 1：Make stool sampl suspension.
Step 2：Stool sampl suspension is filled with tally.
Step 3：The sample on tally is amplified automatically using microscope.
Step 4：It is focused automatically upwards since tally bottom surface according to preset search range, finds first
A focusing surface carries out Image Acquisition and obtains first object image and record the first focusing surface position.
Step 5：Fog-level is carried out to all targets in first object image to detect to obtain the first testing result, when the
There are when unintelligible target image in one target image, into next step；When there is no unintelligible targets in first object image
When image, Motion parameters and classification are directly carried out, obtains the operation of sample object classification and Detection result.
Step 6：Detect first object image in each target clarity, form and size, by this few class parameter into
Row weights and refers to testing requirements, determines the first focusing step-length.
Step 7：Respectively 5 sub-pictures of acquisition, the focal position per sub-picture are divided between focusing up and down near the first focusing surface
First focusing step-length.This 10 sub-picture is acquired respectively obtains the second image group.
Step 8：Read default second focusing step-length.
Step 9：Microscope is adjusted according to default second focusing step-length on the first focusing surface top, and acquires image, is obtained
Third image.
Step 10：Image in first object image and the second target image group and third image is merged, is removed
Then redundancy object and unintelligible target carry out automatic identification and classification to remaining target, to obtain sample object classification inspection
Survey result.
Example two：
Step 1：Make stool sampl suspension.
Step 2：Stool sampl suspension is filled with tally.
Step 3：The sample on tally is amplified automatically using microscope.
Step 4：Tally bottom focus automatically and Image Acquisition obtains first object image；
Step 5：Fog-level is carried out to all targets in first object image to detect to obtain the first testing result, when the
There are when unintelligible target image in one target image, into next step；When there is no unintelligible targets in first object image
When image, Motion parameters and classification are directly carried out, obtains the operation of sample object classification and Detection result.
Step 6：The default fixed focusing step-length L and adopt figure layer number n that microscope is adjusted are read, from tally bottom under
Collect specimen picture obtains second, third to the fixed focusing step-length L in supreme interval respectively ... until n-th layer target image, presets solid
The burnt step-length L that sets the tone is preset based on the size of each target, detection demand and historical empirical data in first object image
's.
Step 7：To first, second ... and the n-th target image carries out image automatic identification and differential counting, adds up and goes
Sample to be tested target classification testing result is obtained after falling repetition object count.
Several embodiments of the invention above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to the protection of the present invention
Range.Therefore, the protection domain of patent of the present invention should be determined by the appended claims.
Claims (8)
1. a kind of multi-level focus adopts drawing method automatically, which is characterized in that including step：
Tally is focused according to preset search range, searches the first focusing surface, Image Acquisition is carried out to first focusing surface,
Obtain first object image, wherein the tally is filled with sample suspensions, and focusing is since tally bottom to the sample
The surface of this suspension carries out, or is carried out from the surface of the sample suspensions to the tally bottom；
Obtain focusing step-length；
According to the focusing step-length, figure figure layer is adopted in setting, is acquired each image for adopting figure figure layer respectively, is obtained the second target
Image group；
Acquired target image is merged, removes redundancy object and unintelligible target, and to remaining Motion parameters with
And classification, obtain sample object classification and Detection result；
Described according to the focusing step-length, figure figure layer is adopted in setting, is acquired each image for adopting figure figure layer respectively, is obtained second
Further include after target image group step：
Read default second focusing step-length, wherein the default second focusing step-length determination process is specifically, acquisition is filled with counting
The first position of tally bottom target focus layer is fallen in the sample suspensions of plate and is suspended in the focusing of tally top target
The second position of layer, the corresponding first position of the multiple sample suspensions of repeated acquisition and the second position, statistics obtain the
One set of locations and second position group obtain default second focusing according to the first position group and the second position group
Step-length；
In the station acquisition image from first focusing surface at a distance of the default second focusing step-length, third target figure is obtained
Picture；
It is described to merge acquired target image, redundancy object and unintelligible target are removed, and know automatically to remaining target
And the step of classifying, obtain sample object classification and Detection result it is not specially：
By in the first object image, the second target image group image and the third target image fusion, go
Except redundancy object and unintelligible target, and to remaining Motion parameters and classification, obtain sample object classification and Detection knot
Fruit.
2. multi-level focusing according to claim 1 adopts drawing method automatically, which is characterized in that described to obtain step-length of focusing
Step is specially：
The image parameter for detecting all kinds of targets in the first object image, according to all kinds of targets in the first object image
Image parameter obtains focusing step-length.
3. multi-level focusing according to claim 1 or 2 adopts drawing method automatically, which is characterized in that described according to the tune
Figure figure layer is adopted in burnt step-length, setting, and the step of acquiring each image for adopting figure figure layer respectively, obtaining the second target image group has
Body includes：
The position on the basis of the position of first focusing surface is extended to preset with upper and lower both direction and adopts figure section respectively,
It is described it is default adopt in figure section, according to the focusing step-length, figure figure layer is adopted in setting, wherein described default to adopt figure section and be less than
The section on tally bottom to the sample suspensions surface；
Each image for adopting figure figure layer is acquired, the second target image group is obtained.
4. multi-level focusing according to claim 1 adopts drawing method automatically, which is characterized in that described to obtain step-length of focusing
Step is specially：
Read default focusing step-length.
5. multi-level focusing according to claim 1 or 4 adopts drawing method automatically, which is characterized in that described according to the tune
Figure figure layer is adopted in burnt step-length, setting, acquires each image for adopting figure figure layer respectively, it is specific to obtain the second target image group step
Including：
The position on the basis of tally bottom, in the tally bottom to sample suspensions surface section, according to described
Figure figure layer is adopted in focusing step-length, at equal intervals setting；
Each image for adopting figure figure layer is acquired respectively, obtains the second target image group.
6. a kind of multi-level focus adopts drawing system automatically, which is characterized in that including：
First object image collection module searches the first focusing surface, to institute for being focused to tally according to preset search range
It states the first focusing surface and carries out Image Acquisition, obtain first object image, wherein the tally was focused filled with sample suspensions
Journey is carried out since tally bottom to the surface of the sample suspensions, or from the surface of the sample suspensions to the tally
Bottom carries out；
Focusing step-length acquisition module, for obtaining focusing step-length；
Second target image group acquisition module acquires each described respectively for according to the focusing step-length, setting and adopting figure figure layer
The image of figure figure layer is adopted, the second target image group is obtained；
Fusion Module removes redundancy object and unintelligible target, and to remaining mesh for merging acquired target image
Automatic identification and classification are marked, sample object classification and Detection result is obtained；
Second focusing step-length acquisition module, for reading default second focusing step-length, wherein the default second focusing step-length is true
Determine process specifically, acquisition be filled with fallen in the sample suspensions of tally the first position of tally bottom target focus layer with
Be suspended in the second position of tally top target focus layer, the corresponding first position of the multiple sample suspensions of repeated acquisition with
The second position, statistics obtains first position group and second position group, according to the first position group and the second position
Group obtains the default second focusing step-length；
Third target image acquisition module, in the position for step-length of focusing at a distance of described default second from first focusing surface
Image is acquired, third target image is obtained；
The Fusion Module be used for by the first object image, the second target image group image and the third
Target image merges, and removes redundancy object and unintelligible target, and to remaining Motion parameters and classification, obtain sample
Target classification testing result.
7. multi-level focusing according to claim 6 adopts drawing system automatically, which is characterized in that the second target image group
Acquisition module specifically includes：
First adopts figure figure layer setup unit, for the position on the basis of the position of first focusing surface, respectively with upper and lower two
Direction, which is extended to preset, adopts figure section, is adopted in figure section in described preset, and according to the focusing step-length, figure figure layer is adopted in setting,
Wherein, the default section adopted figure section and be less than tally bottom to the sample suspensions surface；
First collecting unit obtains the second target image group for acquiring each image for adopting figure figure layer.
8. multi-level focusing according to claim 6 adopts drawing system automatically, which is characterized in that the second target image group
Acquisition module specifically includes：
Second adopts figure figure layer setup unit, for the position on the basis of tally bottom, in the tally bottom to the sample
In the section of this suspension surface, according to the focusing step-length, figure figure layer is adopted in setting at equal intervals；
Second collecting unit obtains the second target image group for acquiring each image for adopting figure figure layer respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510641966.8A CN105241811B (en) | 2015-09-30 | 2015-09-30 | Multi-level focus adopts drawing method and system automatically |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510641966.8A CN105241811B (en) | 2015-09-30 | 2015-09-30 | Multi-level focus adopts drawing method and system automatically |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105241811A CN105241811A (en) | 2016-01-13 |
| CN105241811B true CN105241811B (en) | 2018-10-23 |
Family
ID=55039533
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510641966.8A Active CN105241811B (en) | 2015-09-30 | 2015-09-30 | Multi-level focus adopts drawing method and system automatically |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105241811B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106019557A (en) * | 2016-08-17 | 2016-10-12 | 珠海科域生物工程有限公司 | Multi-layer photographing method and photographing system for microscopic examination |
| CN110763679A (en) * | 2019-10-31 | 2020-02-07 | 湖南爱威医疗科技有限公司 | Image acquisition method and device, microscope system and computer readable storage medium |
| CN110927158B (en) * | 2019-10-31 | 2022-07-05 | 湖南爱威医疗科技有限公司 | Image acquisition method and device, microscope system and computer readable storage medium |
| CN112241953B (en) * | 2020-10-22 | 2023-07-21 | 江苏美克医学技术有限公司 | Sample image fusion method and device based on multi-focus image fusion and HDR algorithm |
| CN112861675B (en) * | 2021-01-28 | 2024-08-20 | 四川沃文特生物技术有限公司 | Detection and identification method for fecal component |
| CN112986141B (en) * | 2021-02-05 | 2024-04-30 | 爱威科技股份有限公司 | Hierarchical image acquisition method, hierarchical image acquisition device, computer equipment and storage medium |
| CN112986239B (en) * | 2021-02-05 | 2024-06-07 | 爱威科技股份有限公司 | Hierarchical image acquisition method, hierarchical image acquisition device, computer equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6438261B1 (en) * | 1998-09-03 | 2002-08-20 | Green Vision Systems Ltd. | Method of in-situ focus-fusion multi-layer spectral imaging and analysis of particulate samples |
| CN101615289A (en) * | 2009-08-05 | 2009-12-30 | 北京优纳科技有限公司 | The three-dimensional acquisition of biopsy tissues and fusion method of multilayer images |
| CN101630061A (en) * | 2009-08-17 | 2010-01-20 | 公安部物证鉴定中心 | Optical confocal three-dimensional data acquisition system of tool traces and acquisition method thereof |
| CN202600251U (en) * | 2012-04-01 | 2012-12-12 | 潍坊学院 | Microscopic image fusion system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100808610B1 (en) * | 2006-06-13 | 2008-02-28 | 중앙대학교 산학협력단 | Digital Multi-focusing using Image Fusion |
-
2015
- 2015-09-30 CN CN201510641966.8A patent/CN105241811B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6438261B1 (en) * | 1998-09-03 | 2002-08-20 | Green Vision Systems Ltd. | Method of in-situ focus-fusion multi-layer spectral imaging and analysis of particulate samples |
| CN101615289A (en) * | 2009-08-05 | 2009-12-30 | 北京优纳科技有限公司 | The three-dimensional acquisition of biopsy tissues and fusion method of multilayer images |
| CN101630061A (en) * | 2009-08-17 | 2010-01-20 | 公安部物证鉴定中心 | Optical confocal three-dimensional data acquisition system of tool traces and acquisition method thereof |
| CN202600251U (en) * | 2012-04-01 | 2012-12-12 | 潍坊学院 | Microscopic image fusion system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105241811A (en) | 2016-01-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105241811B (en) | Multi-level focus adopts drawing method and system automatically | |
| US11774735B2 (en) | System and method for performing automated analysis of air samples | |
| JP3822242B2 (en) | Method and apparatus for evaluating slide and sample preparation quality | |
| KR102536011B1 (en) | System, method for training and applying a defect classifier on wafers with deeply stacked layers | |
| US9489562B2 (en) | Image processing method and apparatus | |
| JP3990981B2 (en) | Method and apparatus for inspecting a substrate | |
| CN106226157B (en) | Concrete structure member crevices automatic detection device and method | |
| RU2019115141A (en) | System and method for performing automated analysis of air samples | |
| US9811897B2 (en) | Defect observation method and defect observation device | |
| JP6479676B2 (en) | System and method for classification of particles in a fluid sample | |
| CN102637258B (en) | Method for creating online surface quality detection system defect library | |
| CN109142371A (en) | High density flexible exterior substrate defect detecting system and method based on deep learning | |
| CN106952258A (en) | A kind of bottle mouth defect detection method based on gradient orientation histogram | |
| JP2018504768A (en) | Wafer defect detection | |
| CN109117836A (en) | Text detection localization method and device under a kind of natural scene based on focal loss function | |
| CN107845090A (en) | A kind of silicon chip detection method and silicon chip detection means | |
| CN108491851A (en) | A kind of container lockhole based on machine vision is quick to be identified and suspender method for correcting error | |
| JPH10508709A (en) | Device for detecting air bubbles in cover slip adhesive | |
| JP2009250645A (en) | Defect reviewing method and device therefor | |
| CN108186051A (en) | A kind of image processing method and processing system of the automatic measurement fetus Double Tops electrical path length from ultrasonoscopy | |
| TW202219494A (en) | A defect detection method and a defect detection device | |
| CN108508023A (en) | The defect detecting system of end puller bolt is contacted in a kind of railway contact line | |
| JP3272036B2 (en) | Optical inspection method and device for defects on object surface | |
| CN114324382A (en) | Panel terminal cleanliness detection method and panel terminal cleanliness detection device | |
| JP2010151523A (en) | Method and device for analyzing particle image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 410000 Aiwei medical science and Technology Park, No.26 Fuling Road, bachelor street, Yuelu District, Changsha City, Hunan Province Patentee after: AVE SCIENCE & TECHNOLOGY Co.,Ltd. Country or region after: China Address before: 410013 B6, 27 Yuan Lu Yuan Yuan, Wen Xuan Road, Changsha hi tech Development Zone, Hunan. Patentee before: AVE SCIENCE & TECHNOLOGY Co.,Ltd. Country or region before: China |