CN109191434A - Image detecting system and detection method in a kind of cell differentiation - Google Patents
Image detecting system and detection method in a kind of cell differentiation Download PDFInfo
- Publication number
- CN109191434A CN109191434A CN201810913993.XA CN201810913993A CN109191434A CN 109191434 A CN109191434 A CN 109191434A CN 201810913993 A CN201810913993 A CN 201810913993A CN 109191434 A CN109191434 A CN 109191434A
- Authority
- CN
- China
- Prior art keywords
- image
- cell
- module
- coefficient
- scattering
- 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.)
- Pending
Links
- 230000024245 cell differentiation Effects 0.000 title claims abstract description 45
- 238000001514 detection method Methods 0.000 title claims abstract description 17
- 108090000623 proteins and genes Proteins 0.000 claims abstract description 33
- 238000003709 image segmentation Methods 0.000 claims abstract description 24
- 230000011218 segmentation Effects 0.000 claims abstract description 20
- 230000004069 differentiation Effects 0.000 claims abstract description 18
- 239000003550 marker Substances 0.000 claims abstract description 17
- 230000008676 import Effects 0.000 claims abstract description 13
- 230000006698 induction Effects 0.000 claims abstract description 11
- 238000000034 method Methods 0.000 claims description 48
- 230000006870 function Effects 0.000 claims description 33
- 239000011159 matrix material Substances 0.000 claims description 30
- 230000014509 gene expression Effects 0.000 claims description 22
- 238000012545 processing Methods 0.000 claims description 21
- 238000010521 absorption reaction Methods 0.000 claims description 20
- 238000004422 calculation algorithm Methods 0.000 claims description 17
- 230000008569 process Effects 0.000 claims description 13
- 238000012549 training Methods 0.000 claims description 12
- 210000004209 hair Anatomy 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000005516 engineering process Methods 0.000 claims description 10
- 238000003384 imaging method Methods 0.000 claims description 10
- 238000005259 measurement Methods 0.000 claims description 10
- 230000005540 biological transmission Effects 0.000 claims description 9
- 238000001914 filtration Methods 0.000 claims description 8
- 239000007787 solid Substances 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 229910052704 radon Inorganic materials 0.000 claims description 6
- SYUHGPGVQRZVTB-UHFFFAOYSA-N radon atom Chemical compound [Rn] SYUHGPGVQRZVTB-UHFFFAOYSA-N 0.000 claims description 6
- 230000003044 adaptive effect Effects 0.000 claims description 4
- 230000005859 cell recognition Effects 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 230000004927 fusion Effects 0.000 claims description 4
- 230000008447 perception Effects 0.000 claims description 4
- 238000000527 sonication Methods 0.000 claims description 4
- 230000008901 benefit Effects 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000009792 diffusion process Methods 0.000 claims description 3
- 230000004907 flux Effects 0.000 claims description 3
- 230000001939 inductive effect Effects 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims description 3
- 230000035772 mutation Effects 0.000 claims description 3
- 230000000644 propagated effect Effects 0.000 claims description 3
- 239000004576 sand Substances 0.000 claims description 3
- 230000003595 spectral effect Effects 0.000 claims description 3
- 238000009827 uniform distribution Methods 0.000 claims description 3
- 230000008859 change Effects 0.000 claims 1
- 230000002708 enhancing effect Effects 0.000 claims 1
- 230000000149 penetrating effect Effects 0.000 claims 1
- 238000003745 diagnosis Methods 0.000 abstract description 4
- 238000005070 sampling Methods 0.000 abstract description 4
- 230000003862 health status Effects 0.000 abstract description 2
- 239000000523 sample Substances 0.000 description 25
- 230000000694 effects Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 2
- 239000012535 impurity Substances 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000005192 partition Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 108700039691 Genetic Promoter Regions Proteins 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 239000012472 biological sample Substances 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000032459 dedifferentiation Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006806 disease prevention Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 230000002427 irreversible effect Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 102000004169 proteins and genes Human genes 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30024—Cell structures in vitro; Tissue sections in vitro
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
- Investigating Or Analysing Biological Materials (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to cell differentiation technical fields, the image detecting system and detection method in a kind of cell differentiation are disclosed, the image detecting system in cell differentiation includes: marker gene import modul, induction differentiation module, image capture module, main control module, picture recognition module, image segmentation module, data memory module, display module.The present invention improves the accuracy and recognition speed of cell image recognition by picture recognition module, and intelligent operation and judgement may be implemented, and can quickly identify the class state of cell, and can helping doctor, rapidly whether diagnosis cell is in health status;The result obtained after the segmentation of piecemeal threshold value is screened by image segmentation module simultaneously, incomplete nucleus is sent into adaptivenon-uniform sampling and carries out secondary splitting, complete nuclei picture directly exports, and the result of result and piecemeal Threshold segmentation using adaptivenon-uniform sampling is then merged into final optimal output image.
Description
Technical field
The invention belongs in cell differentiation technical field more particularly to a kind of cell differentiation image detecting system and detection
Method.
Background technique
Cell differentiation (celldifferentiation) refers to that the cell of same source gradually produces morphosis, function
Can the different cell population of feature process, as a result, spatially cell generates difference, same cell in time
It is different with the state of its past.The essence of cell differentiation is the selective expression of genome over time and space, is passed through
Opening or closing for different genes expression, finally generates significant protein.Under normal circumstances, cell differentiation procedure is irreversible
's.However, under certain conditions, the cell broken up is also unstable, invertibity variation can also occur for gene expression pattern,
It is returned to its undifferentiated state, this process is known as dedifferenting (dedifferentiation).However, existing cell differentiation figure
As accuracy of identification error is big, speed is slow;The image iuntercellular overlapping of acquisition simultaneously;Cell edges boundary is fuzzy;Cell image ginseng
Enter impurity;Nucleus size shape texture is inconsistent, influences diagnosis judgement.
In conclusion problem of the existing technology is:
Existing cell differentiation image recognition precision error is big, speed is slow;The image iuntercellular overlapping of acquisition simultaneously;Cell side
Edge boundary is fuzzy;Cell image participates in impurity;Nucleus size shape texture is inconsistent, influences diagnosis judgement.
The image of the shooting of the image detecting system of cell is not clear enough in the prior art, to the further research band of experiment
It is inconvenient to come;Figure segmentation effect is bad, cannot effectively improve the segmentation effect of image;In the display of cell image, Bu Nengyou
The removal discrete noise image of effect influences the conclusion and summary of result of study so that watching cell image very unclear.
OPT is imaged since biological sample is without transparency process, there are problems that photon equilibrium state.Conventional OPT imaging
Algorithm only considers biological tissue to the absorption characteristic of photon, and the absorption coefficient that can only also rebuild photon cannot rebuild the scattering of photon
Coefficient.When being imaged using OPT technology, the influence of scattering be can not ignore, and can be blended in the absorption characteristic of photon
Together, so as to cause the reduction of traditional OPT imaging space resolution ratio and the inaccuracy of reconstructed results.
Summary of the invention
In view of the problems of the existing technology, the present invention provides the image detecting systems in a kind of cell differentiation.
The invention is realized in this way the image detecting method in a kind of cell differentiation, comprising:
The luminescent image as caused by the expression of light generating protein gene in cell is shot by image capture module;
The luminescent image generated by picture recognition module acquisition, identifies cell class status information;
In Image Acquisition, the figure of acquisition is improved using gray level image cross hairs region clarity theoretical model algorithm
The image of picture, the shooting for the image detecting system in cell differentiation is made of m × n pixel, grey scale pixel value matrix B
(I, J), wherein 0≤I≤m-1,0≤J≤n-1;
Cross hairs gray level image maximum gradation value Bmax, minimum gradation value Bmin, B is used in the 1/2 of gray scale difference valuedifIt indicates, it is as follows
Formula:
According to cross hairs gray level image sharpness computation theoretical model, if the clarity of gray level image is C, obtained cross
Line gray level image clarity improved model, such as following formula:
It is split by image of the image segmentation module to acquisition and filters out nucleus figure;During image is split, adopt
With FCM image segmentation algorithm, specifically have:
The determination of initialization: according to the requirement of image segmentation, the determination initialized to image is needed, and to needs
Parameter is initialized, and by the cluster centre of histogram;
The determination of the adaptivity of the factor, fitness, according to the fitness function of construction:
F=a/ (b+J);
Wherein, a, b are adjustable parameters;The objective function that J is;
Mutation operation: individual front and back variable quantity be 0.5r (t/T), data r be generated in defined section it is random
Number, T are the maximum algebra calculated;
Iterative calculation: new fuzzy membership matrix will be obtained by new cutting data, generate new cutting parameter, return
Iterative calculation is returned, the segmentation of image is completed in the termination until completing condition;
Pass through the image data information of data memory module storage acquisition;Pass through the cell image of display module display acquisition
Data information;In the cell image data information for showing acquisition, discrete noise image is removed using non-local mean filtering algorithm,
Discrete noise image v=v (i) | and i ∈ I } to the estimated value NL [v] (i) of a pixel i, calculate all pixels in image
Weighted average, w (i, j) be weight, 0≤w (i, j)≤1 and
Gray vector v (Ni) and v (Nj) similitude indicate pixel i and pixel j between similitude,
For square of the weighted euclidean distance in the region i, j, a indicates Gaussian kernel standard deviation, and a > 0, h are the coefficient that wave-path degree is considered in control, Z
(i) all areas similarity summation within the scope of picture search.
Further, the luminescent image generated by picture recognition module acquisition, identifies in cell class status information, need to be into
Row: biological tissue's attenuation coefficient is rebuild;Rebuild biological tissue scatters coefficient;Calculate the absorption coefficient of cell;
Before rebuilding biological tissue's attenuation coefficient, the transmission modeling of photon weak scattering biological tissue need to be carried out, comprising:
Indicate Γ-The photon of upper incidence is to positionAnd direction isWhen the distance propagated, then:
WhereinForThe radiancy at place indicates in unit solid angle, the unit time is interior, by perpendicular to unit side
To vectorUnit area on mean power flux density, dimension is W/ (m2.Sr);K0For the photon ballistic propagation of introducing
Operator,Indicate that biological tissue existsTotal attenuation coefficient at place,Indicate biological tissue
Absorption coefficient,Indicate scattering coefficient;
It re-defines:
Wherein K is the photon equilibrium state propagation operator introduced,For normalized scattering phase function, indicate photon from
Direction v' is scattered toThe probability in direction meetsD Ω ' expression unit direction vectorCorresponding solid angle
Infinitesimal;Define m0=K0gin,Then there is nn+1=Kmn(n >=0), thusTotal radiance at place are as follows:
Wherein mnIt indicates to scatter through n times and reachRadiancy component;When photon diffusion zone propagate when, the spectral radius of K
ρ (K) value is close to 1, when photon is in weak scattering regional spread, ρ (K) > > 1, in this case, as n → ∞,
Fast convergence;
Then, in output boundary Γ+The total amount of data g that upper description detector receivesout, i.e.,
To:
Wherein A is the matrix for describing photon transmission, A0、A1And A2Ballistic transport is described respectively, primary scattering transmits and multiple
Hop is scattered, g is defined0=A0gin, g1=A1ginRespectively indicate the ballistic transport component and primary scattering point in measurement numerical value
Amount, then know:
Incident light direction isIts direction is after primary scatteringThen in above formula aboutIntegral only one
There is value on special angle, takesThe value of coefficient k is by phase functionIt determines, together
Shi Dingyi Respectively indicate light
Attenuation after son scatters and before scattering, then have:
Further, calculating reconstruction attenuation coefficient includes:
Using the directional light of space uniform distribution to cell g in imaginginIt is irradiated, it is integrated by picture recognition module
CCD camera acquire the cell-free irradiation light blocked and measure incident intensity;It is right The right and left is the same as divided by ginAnd take negative logarithm, then:
Collect 360 degree of measurement data G0Afterwards, inverse Radon is realized using accurate efficient filter back-projection reconstruction algorithm
Transformation calculates attenuation coefficient, i.e. μt=FBP (G0);
The reconstruction scattering coefficient is by formulag1Contain OPT
The influence scattered in imaging, as the angle acquisition data g from a certain determination1When,WithScattering angle determine, coefficient k one
The constant of a determination;Both sides are the same as divided by kgin, then have:
Known by above formulaProlong for scattering coefficientThe weighting Radon in direction is converted, institute weighted value ω1(t) and ω2(t)
It is function related with attenuation coefficient, it willDiscretization is simultaneously expressed as follows with a matrix type:
Wμs=G1;
Wherein W indicates the weight matrix after discretization, μsAnd G1Scattering coefficient vector sum different angle is respectively indicated to measure
The AVHRR NDVI vector arrived establishes following objective function using the weighted least-squares criterion with penalty function:
Wherein the first item of expression formula is the approximate expression form of likelihood function, Section 2 R (μs) it is regular terms, usual root
It is constructed according to the prior information of image, β is regularization factors, and Matrix C is covariance matrix;With niIndicate ccd detector inspection
The scattered photon number measured, corresponding covariance matrix indicate are as follows:
Using optimal method to Φ (μs) objective function solve, that is, find out scattering coefficient:
μs=argmin Φ (μs);
It calculates absorption coefficient and utilizes calculated result, calculate the absorption coefficient of cell, utilize relational expression μt=μa+μsIt calculates thin
The absorption coefficient μ of born of the same parentsa。
Further, image-recognizing method includes:
1) cell image is pre-processed, using to the processing of cell image gray processing, image denoising sonication, image ash
Spend histogram equalization processing, image dividing processing;
2) sparse coefficient is established to pretreated cell image using compressed sensing technology:
Cell image is extracted by column first and constitutes cell image sample column vector;
The perception matrix A ∈ R that training sample cell image in database is constitutedm×nMeasurement square as compressed sensing
Battle array, wherein m is sample characteristics dimension, and n is sample size;
According to compressive sensing theory, to real-time collected cell image sample y ∈ Rm, by solving optimal l1Norm is come
Sparse coefficient x is constructed, solution mode is
3) cell recognition calculates the difference r of the linear weighted function of cell to be identified and every a kind of training sample celli(y), it selects
Classification of the generic of the smallest a kind of training sample of difference as test sample is selected, specific formula for calculation is;
In formula,Indicate extract cell sparse coefficient x to be identified with
The corresponding coefficient of i-th class cell image.
Further, image partition method further comprises:
(1) a width cell TCT image is inputted, carrying out gradation conversion, median filtering denoising and contrast to image enhances
Deng pretreatment;
(2) processing of piecemeal Threshold segmentation is carried out to all images, Threshold segmentation is carried out using OTSU method, is partitioned into prospect
With background;
(3) image shape test is carried out, contexts image distinctness is directly placed into result output set to be detected, does not lead to
Cross the carry out adaptive threshold fuzziness again of detection;
(4) result divided twice is merged, just obtains final ideal segmented image set.
Another object of the present invention is to provide a kind of computer program, the computer program operation cell differentiation
In image detecting method.
Another object of the present invention is to provide a kind of terminal, and the terminal at least carries the figure realized in the cell differentiation
As the controller of detection method.
Another object of the present invention is to provide a kind of computer readable storage medium, including instruction, when its on computers
When operation, so that computer executes the image detecting method in the cell differentiation.
Another object of the present invention is to provide a kind of cell differentiation for realizing the image detecting method in the cell differentiation
In image detecting system, the image detecting system in the cell differentiation includes:
Marker gene import modul, connect with main control module, for be imported with differentiation state detection marker gene open
The cell of the fusion of sub-area and light generating protein gene carries out culture processes;
Induction differentiation module, connect with main control module, for for the cell after 1 process of marker gene import modul,
It is cultivated under conditions of induction differentiation;
Image capture module is connect with main control module, for in cell as caused by the expression of light generating protein gene
Luminescent image is shot;
Main control module, with marker gene import modul, induction differentiation module, image capture module, picture recognition module, figure
As segmentation module, data memory module, display module connection, worked normally for controlling modules;
Picture recognition module is connect with main control module, for information such as image recognition cell class states to acquisition;
Image segmentation module is connect with main control module, filters out nucleus figure for being split to the image of acquisition;
Data memory module is connect with main control module, for storing the image data information of acquisition;
Display module is connect with main control module, for showing the cell image data information of acquisition.
Another object of the present invention is to provide a kind of cell differentiation image detection platform, and the cell differentiation image detection is flat
Platform at least carries the image detecting system in the cell differentiation.
Advantages of the present invention and good effect are as follows:
The present invention improves the accuracy and recognition speed of cell image recognition by picture recognition module, and intelligence may be implemented
Energyization operation and judgement, can quickly identify the class state of cell, can helping doctor, rapidly whether diagnosis cell is in
Health status, prevention of disease and treatment have preferable application value;Simultaneously by image segmentation module first with improved
Piecemeal threshold value partitioning algorithm obtains relatively rough segmentation result, then utilizes nuclear characteristics form test method, right
The result obtained after the segmentation of piecemeal threshold value is screened, and incomplete nucleus is sent into adaptivenon-uniform sampling and carries out secondary point
It cuts, complete nuclei picture directly exports, and then the result of result and piecemeal Threshold segmentation using adaptivenon-uniform sampling is closed
And at final optimal output image.
Institute of the invention is improved using gray level image cross hairs region clarity theoretical model algorithm, so that cell differentiation
In image detecting system shooting image it is relatively sharp, accurately;Figure divides module using FCM image segmentation algorithm point
Analysis, so that image segmentation is accurate, effectively improves the segmentation effect of image;Display module is adopted to remove discrete noise image
It is handled with non-local mean filtering algorithm, so that cell image is apparent.
The present invention proposes while rebuilding the effective ways of absorption coefficient and scattering coefficient in terms of transmission-type OPT imaging.
There are problems that weak scattering OPT imaging, in conjunction with the own characteristic of OPT image-forming data acquisition, construct corresponding mathematical model,
And the method by additionally measuring the data that one group keeps certain tilt angle with incident light, it is to ballistic transport component and once scattered
It penetrates transmission component to be separated, and then realizes the three-dimensional reconstruction of sample absorption coefficient and scattering coefficient, to both can effectively solve
Certainly existing scattering problems are imaged in OPT;It improves OPT image quality, while the information content that richer OPT technology provides, makes
Obtaining OPT technology can be from the texture characteristic of two angles description biologies of absorption coefficient and scattering coefficient.
Detailed description of the invention
Fig. 1 is the image detecting system structural block diagram in cell differentiation provided in an embodiment of the present invention.
In figure: 1, marker gene import modul;2, induction differentiation module;3, image capture module;4, main control module;5, scheme
As identification module;6, image segmentation module;7, data memory module;8, display module.
Fig. 2 is image segmentation module dividing method flow chart provided in an embodiment of the present invention.
Specific embodiment
In order to further understand the content, features and effects of the present invention, the following examples are hereby given, and cooperate attached drawing
Detailed description are as follows.
Structure of the invention is explained in detail with reference to the accompanying drawing.
As shown in Figure 1, the image detecting system in cell differentiation provided in an embodiment of the present invention, comprising: marker gene is led
Enter module 1, induction differentiation module 2, image capture module 3, main control module 4, picture recognition module 5, image segmentation module 6, number
According to memory module 7, display module 8.
Marker gene import modul 1 is connect with main control module 4, for (existing to differentiation state detection marker gene is imported with
Have disclosed in technology or the gene of similar functions) promoter region (disclose in the prior art or the gene of similar functions)
Culture processes are carried out with the cell of the fusion of light generating protein gene (disclose in the prior art or the gene of similar functions);
Induction differentiation module 2, connect with main control module 4, for for the cell after 1 process of marker gene import modul,
It is cultivated under conditions of inducing differentiation;
Image capture module 3 is connect with main control module 4, for in cell as produced by the expression of light generating protein gene
Luminescent image shot;
Main control module 4, with marker gene import modul 1, induction differentiation module 2, image capture module 3, image recognition mould
Block 5, image segmentation module 6, data memory module 7, display module 8 connect, and work normally for controlling modules;
Picture recognition module 5 is connect with main control module 4, for information such as image recognition cell class states to acquisition;
Image segmentation module 6 is connect with main control module 4, filters out nucleus figure for being split to the image of acquisition;
Data memory module 7 is connect with main control module 4, for storing the image data information of acquisition;
Display module 8 is connect with main control module 4, for showing the cell image data information of acquisition.
5 recognition methods of picture recognition module provided by the invention is as follows:
A) cell image is pre-processed, is specifically included to the processing of cell image gray processing, image denoising sonication, figure
As gray-level histogram equalization processing, image dividing processing process;
B) establishing sparse coefficient to the pretreated cell image using compressed sensing technology indicates, this step includes
Following procedure:
1) cell image is extracted first by column and constitutes cell image sample column vector;
2): the perception matrix A ∈ R that the training sample cell image in database is constitutedm×nMeasurement as compressed sensing
Matrix, wherein m is sample characteristics dimension, and n is sample size;
3): according to compressive sensing theory, to real-time collected cell image sample y ∈ Rm, by solving optimal l1Norm
Construct sparse coefficient x, solution mode is
C): cell recognition calculates the difference r of the linear weighted function of cell to be identified and every a kind of training sample celli(y),
Classification of the generic of the smallest a kind of training sample of difference as test sample is selected, specific formula for calculation is;
In formula,Indicate extract cell sparse coefficient x to be identified with
The corresponding coefficient of i-th class cell image.
As indicated with 2, image segmentation module dividing method provided in an embodiment of the present invention, comprising:
S101: one width cell TCT image of input carries out gradation conversion, median filtering denoising and contrast to image and increases
It is strong to wait pretreatment;
S102: carrying out the processing of piecemeal Threshold segmentation to all images, Threshold segmentation is carried out using OTSU method, before being partitioned into
Scape and background;
S103: carrying out image shape test, and contexts image distinctness is directly placed into result output set to be detected, not
Pass through the carry out adaptive threshold fuzziness again of detection;
S104: merging the result divided twice, just obtains final ideal segmented image set.
The starting when present invention detects, by marker gene import modul 1 to differentiation state detection marker gene is imported with
The cell of the fusion of subregion and light generating protein gene carries out culture processes;By induction differentiation module 2 for marking base
Because of the cell after 1 process of import modul, cultivated under conditions of inducing differentiation;By image capture module 3 in cell
The luminescent image as caused by the expression of light generating protein gene is shot;Main control module 4 dispatches 5 pairs of picture recognition module acquisitions
The information such as image recognition cell class state;The image acquired by 6 Duis of image segmentation module, which is split, filters out cell
Core figure;The image data information of acquisition is stored by data memory module 7;Pass through the cell image of the display acquisition of display module 8
Data information.
Invention is further described in conjunction with concrete analysis.
Image detecting method in cell differentiation provided in an embodiment of the present invention, comprising:
The luminescent image as caused by the expression of light generating protein gene in cell is shot by image capture module;
The luminescent image generated by picture recognition module acquisition, identifies cell class status information;
In Image Acquisition, the figure of acquisition is improved using gray level image cross hairs region clarity theoretical model algorithm
The image of picture, the shooting for the image detecting system in cell differentiation is made of m × n pixel, grey scale pixel value matrix B
(I, J), wherein 0≤I≤m-1,0≤J≤n-1;
Cross hairs gray level image maximum gradation value Bmax, minimum gradation value Bmin, B is used in the 1/2 of gray scale difference valuedifIt indicates, it is as follows
Formula:
According to cross hairs gray level image sharpness computation theoretical model, if the clarity of gray level image is C, obtained cross
Line gray level image clarity improved model, such as following formula:
It is split by image of the image segmentation module to acquisition and filters out nucleus figure;During image is split, adopt
With FCM image segmentation algorithm, specifically have:
The determination of initialization: according to the requirement of image segmentation, the determination initialized to image is needed, and to needs
Parameter is initialized, and by the cluster centre of histogram;
The determination of the adaptivity of the factor, fitness, according to the fitness function of construction:
F=a/ (b+J);
Wherein, a, b are adjustable parameters;The objective function that J is;
Mutation operation: individual front and back variable quantity be 0.5r (t/T), data r be generated in defined section it is random
Number, T are the maximum algebra calculated;
Iterative calculation: new fuzzy membership matrix will be obtained by new cutting data, generate new cutting parameter, return
Iterative calculation is returned, the segmentation of image is completed in the termination until completing condition;
Pass through the image data information of data memory module storage acquisition;Pass through the cell image of display module display acquisition
Data information;In the cell image data information for showing acquisition, discrete noise image is removed using non-local mean filtering algorithm,
Discrete noise imageTo the estimated value NL [v] (i) of a pixel i, calculate all in image
The weighted average of pixel, w (i, j) be weight, 0≤w (i, j)≤1 and
Gray vector v (Ni) and v (Nj) similitude indicate pixel i and pixel j between similitude,For square of the weighted euclidean distance in the region i, j, a indicates Gaussian kernel standard deviation, and a > 0, h are control
Consider the coefficient of wave-path degree, all areas similarity summation within the scope of Z (i) picture search.
The luminescent image generated by picture recognition module acquisition, identifies in cell class status information, needs to carry out: rebuilding
Biological tissue's attenuation coefficient;Rebuild biological tissue scatters coefficient;Calculate the absorption coefficient of cell;
Before rebuilding biological tissue's attenuation coefficient, the transmission modeling of photon weak scattering biological tissue need to be carried out, comprising:
Indicate Γ-The photon of upper incidence is to positionAnd direction isWhen the distance propagated, then:
WhereinForThe radiancy at place indicates in unit solid angle, the unit time is interior, by perpendicular to unit side
To vectorUnit area on mean power flux density, dimension is W/ (m2.Sr);K0For the photon ballistic propagation of introducing
Operator,Indicate that biological tissue existsTotal attenuation coefficient at place,Indicate the suction of biological tissue
Coefficient is received,Indicate scattering coefficient;
It re-defines:
Wherein K is the photon equilibrium state propagation operator introduced,For normalized scattering phase function, indicate photon from
Direction v' is scattered toThe probability in direction meetsD Ω ' expression unit direction vectorCorresponding solid angle
Infinitesimal;Define m0=K0gin,Then there is nn+1=Kmn(n >=0), thusTotal radiance at place are as follows:
Wherein mnIt indicates to scatter through n times and reachRadiancy component;When photon diffusion zone propagate when, the spectral radius of K
ρ (K) value is close to 1, when photon is in weak scattering regional spread, ρ (K) > > 1, in this case, as n → ∞,
Fast convergence;
Then, in output boundary Γ+The total amount of data g that upper description detector receivesout, i.e.,
To:
Wherein A is the matrix for describing photon transmission, A0、A1And A2Ballistic transport is described respectively, primary scattering transmits and multiple
Hop is scattered, g is defined0=A0gin, g1=A1ginRespectively indicate the ballistic transport component and primary scattering point in measurement numerical value
Amount, then know:
Incident light direction isIts direction is after primary scatteringThen in above formula aboutIntegral only one
There is value on special angle, takesThe value of coefficient k is by phase functionIt determines, together
Shi Dingyi Respectively indicate light
Attenuation after son scatters and before scattering, then have:
Calculating reconstruction attenuation coefficient includes:
Using the directional light of space uniform distribution to cell g in imaginginIt is irradiated, is integrated by picture recognition module
CCD camera acquires the cell-free irradiation light blocked and measures incident intensity;It is right The right and left is the same as divided by ginAnd take negative logarithm, then:
Collect 360 degree of measurement data G0Afterwards, inverse Radon is realized using accurate efficient filter back-projection reconstruction algorithm
Transformation calculates attenuation coefficient, i.e. μt=FBP (G0);
The reconstruction scattering coefficient is by formulag1Contain OPT
The influence scattered in imaging, as the angle acquisition data g from a certain determination1When,WithScattering angle determine, coefficient k one
The constant of a determination;Both sides are the same as divided by kgin, then have:
Known by above formulaProlong for scattering coefficientThe weighting Radon in direction is converted, institute weighted value ω1(t) and ω2(t)
It is function related with attenuation coefficient, it willDiscretization is simultaneously expressed as follows with a matrix type:
Wμs=G1;
Wherein W indicates the weight matrix after discretization, μsAnd G1Scattering coefficient vector sum different angle is respectively indicated to measure
The AVHRR NDVI vector arrived establishes following objective function using the weighted least-squares criterion with penalty function:
Wherein the first item of expression formula is the approximate expression form of likelihood function, Section 2 R (μs) it is regular terms, usual root
It is constructed according to the prior information of image, β is regularization factors, and Matrix C is covariance matrix;With niIndicate ccd detector inspection
The scattered photon number measured, corresponding covariance matrix indicate are as follows:
Using optimal method to Φ (μs) objective function solve, that is, find out scattering coefficient:
μs=argmin Φ (μs);
It calculates absorption coefficient and utilizes calculated result, calculate the absorption coefficient of cell, utilize relational expression μt=μa+μsIt calculates thin
The absorption coefficient μ of born of the same parentsa。
Image-recognizing method includes:
1) cell image is pre-processed, using to the processing of cell image gray processing, image denoising sonication, image ash
Spend histogram equalization processing, image dividing processing;
2) sparse coefficient is established to pretreated cell image using compressed sensing technology:
Cell image is extracted by column first and constitutes cell image sample column vector;
The perception matrix A ∈ R that training sample cell image in database is constitutedm×nMeasurement square as compressed sensing
Battle array, wherein m is sample characteristics dimension, and n is sample size;
According to compressive sensing theory, to real-time collected cell image sample y ∈ Rm, by solving optimal l1Norm is come
Sparse coefficient x is constructed, solution mode is
3) cell recognition calculates the difference r of the linear weighted function of cell to be identified and every a kind of training sample celli(y), it selects
Classification of the generic of the smallest a kind of training sample of difference as test sample is selected, specific formula for calculation is;
In formula,Indicate extract cell sparse coefficient x to be identified with
The corresponding coefficient of i-th class cell image.
Image partition method further comprises:
(1) a width cell TCT image is inputted, carrying out gradation conversion, median filtering denoising and contrast to image enhances
Deng pretreatment;
(2) processing of piecemeal Threshold segmentation is carried out to all images, Threshold segmentation is carried out using OTSU method, is partitioned into prospect
With background;
(3) image shape test is carried out, contexts image distinctness is directly placed into result output set to be detected, does not lead to
Cross the carry out adaptive threshold fuzziness again of detection;
(4) result divided twice is merged, just obtains final ideal segmented image set.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real
It is existing.When using entirely or partly realizing in the form of a computer program product, the computer program product include one or
Multiple computer instructions.When loading on computers or executing the computer program instructions, entirely or partly generate according to
Process described in the embodiment of the present invention or function.The computer can be general purpose computer, special purpose computer, computer network
Network or other programmable devices.The computer instruction may be stored in a computer readable storage medium, or from one
Computer readable storage medium is transmitted to another computer readable storage medium, for example, the computer instruction can be from one
A web-site, computer, server or data center pass through wired (such as coaxial cable, optical fiber, Digital Subscriber Line (DSL)
Or wireless (such as infrared, wireless, microwave etc.) mode is carried out to another web-site, computer, server or data center
Transmission).The computer-readable storage medium can be any usable medium or include one that computer can access
The data storage devices such as a or multiple usable mediums integrated server, data center.The usable medium can be magnetic Jie
Matter, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk Solid
State Disk (SSD)) etc..
The above is only the preferred embodiments of the present invention, and is not intended to limit the present invention in any form,
Any simple modification made to the above embodiment according to the technical essence of the invention, equivalent variations and modification, belong to
In the range of technical solution of the present invention.
Claims (10)
1. the image detecting method in a kind of cell differentiation, which is characterized in that the image detecting method packet in the cell differentiation
It includes:
The luminescent image as caused by the expression of light generating protein gene in cell is shot by image capture module;
The luminescent image generated by picture recognition module acquisition, identifies cell class status information;
In Image Acquisition, the image of acquisition is improved using gray level image cross hairs region clarity theoretical model algorithm, is used
The image of the shooting of image detecting system in cell differentiation is made of m × n pixel, grey scale pixel value matrix B (I, J),
Wherein 0≤I≤m-1,0≤J≤n-1;
Cross hairs gray level image maximum gradation value Bmax, minimum gradation value Bmin, B is used in the 1/2 of gray scale difference valuedifIt indicates, such as following formula:
According to cross hairs gray level image sharpness computation theoretical model, if the clarity of gray level image is C, obtained cross hairs is grey
Image definition improved model is spent, such as following formula:
It is split by image of the image segmentation module to acquisition and filters out nucleus figure;During image is split, using FCM
Image segmentation algorithm specifically has:
The determination of initialization: according to the requirement of image segmentation, the determination initialized to image is needed, and to the parameter of needs
It is initialized, and by the cluster centre of histogram;
The determination of the adaptivity of the factor, fitness, according to the fitness function of construction:
F=a/ (b+J);
Wherein, a, b are adjustable parameters;The objective function that J is;
Mutation operation: the variable quantity of individual front and back is 0.5r (t/T), and data r is the random number generated in defined section, T
For the maximum algebra of calculating;
Iterative calculation: new fuzzy membership matrix will be obtained by new cutting data, generate new cutting parameter, return changes
In generation, calculates, and the segmentation of image is completed in the termination until completing condition;
Pass through the image data information of data memory module storage acquisition;Pass through the cell image data of display module display acquisition
Information;In the cell image data information for showing acquisition, discrete noise image is removed using non-local mean filtering algorithm, it is discrete
Noise image v=v (i) | and i ∈ I } to the estimated value NL [v] (i) of a pixel i, calculate the weighting of all pixels in image
Average value, w (i, j) be weight, 0≤w (i, j)≤1 and
Gray vector v (Ni) and v (Nj) similitude indicate pixel i and pixel j between similitude,
For square of the weighted euclidean distance in the region i, j, a indicates Gaussian kernel standard deviation, and a > 0, h are the coefficient that wave-path degree is considered in control, Z
(i) all areas similarity summation within the scope of picture search.
2. the image detecting method in cell differentiation as described in claim 1, which is characterized in that
The luminescent image generated by picture recognition module acquisition, identifies in cell class status information, needs to carry out: rebuilding biology
Tissue attenuation coefficient;Rebuild biological tissue scatters coefficient;Calculate the absorption coefficient of cell;
Before rebuilding biological tissue's attenuation coefficient, the transmission modeling of photon weak scattering biological tissue need to be carried out, comprising:
Indicate Γ-The photon of upper incidence is to positionAnd direction isWhen the distance propagated, then:
WhereinForThe radiancy at place indicates in unit solid angle, the unit time is interior, by perpendicular to unit direction vectorUnit area on mean power flux density, dimension is W/ (m2.Sr);K0For the photon ballistic propagation operator of introducing,Indicate that biological tissue existsTotal attenuation coefficient at place,Indicate the absorption system of biological tissue
Number,Indicate scattering coefficient;
It re-defines:
Wherein K is the photon equilibrium state propagation operator introduced,For normalized scattering phase function, indicate photon from direction
V' is scattered toThe probability in direction meetsD Ω ' expression unit direction vectorCorresponding solid angle is micro-
Member;Define m0=K0gin,Then there is nn+1=Kmn(n >=0), thusTotal radiance at place are as follows:
Wherein mnIt indicates to scatter through n times and reachRadiancy component;When photon diffusion zone propagate when, the spectral radius ρ (K) of K
Value is close to 1, when photon is in weak scattering regional spread, ρ (K) > > 1, in this case, as n → ∞,Quickly
Convergence;
Then, in output boundary Γ+The total amount of data g that upper description detector receivesout, i.e.,To:
Wherein A is the matrix for describing photon transmission, A0、A1And A2Ballistic transport, primary scattering transmission and Multiple Scattering are described respectively
Hop defines g0=A0gin, g1=A1ginThe ballistic transport component and primary scattering component in measurement numerical value are respectively indicated,
Then know:
Incident light direction isIts direction is after primary scatteringThen in above formula aboutIntegral it is only specific one
There is value in angle, takesThe value of coefficient k is by phase functionIt determines, it is fixed simultaneously
Justice Photon is respectively indicated to occur to dissipate
Attenuation before penetrating rear and scattering, then have:
3. the image detecting method in cell differentiation as claimed in claim 2, which is characterized in that
Calculating reconstruction attenuation coefficient includes:
Using the directional light of space uniform distribution to cell g in imaginginIt is irradiated, the CCD phase integrated by picture recognition module
Machine acquires the cell-free irradiation light blocked and measures incident intensity;It is right
The right and left is the same as divided by ginAnd take negative logarithm, then:
Collect 360 degree of measurement data G0Afterwards, inverse Radon transform is realized using accurate efficient filter back-projection reconstruction algorithm
Calculate attenuation coefficient, i.e. μt=FBP (G0);
The reconstruction scattering coefficient is by formulag1Contain OPT imaging
The influence of middle scattering, as the angle acquisition data g from a certain determination1When,WithScattering angle determine, coefficient k be one really
Fixed constant;Both sides are the same as divided by kgin, then have:
Known by above formulaProlong for scattering coefficientThe weighting Radon in direction is converted, institute weighted value ω1(t) and ω2(t) it is
Function related with attenuation coefficient, willDiscretization is simultaneously expressed as follows with a matrix type:
Wμs=G1;
Wherein W indicates the weight matrix after discretization, μsAnd G1Respectively indicate what scattering coefficient vector sum different angle measurement obtained
AVHRR NDVI vector establishes following objective function using the weighted least-squares criterion with penalty function:
Wherein the first item of expression formula is the approximate expression form of likelihood function, Section 2 R (μs) it is regular terms, generally according to figure
The prior information of picture constructs, and β is regularization factors, and Matrix C is covariance matrix;With niIndicate that ccd detector detects
Scattered photon number, corresponding covariance matrix indicates are as follows:
Using optimal method to Φ (μs) objective function solve, that is, find out scattering coefficient:
μs=argmin Φ (μs);
It calculates absorption coefficient and utilizes calculated result, calculate the absorption coefficient of cell, utilize relational expression μt=μa+μsCalculate cell
Absorption coefficient μa。
4. the image detecting method in cell differentiation as described in claim 1, which is characterized in that image-recognizing method includes:
1) cell image is pre-processed, using straight to the processing of cell image gray processing, image denoising sonication, image grayscale
Square figure equalization processing, image dividing processing;
2) sparse coefficient is established to pretreated cell image using compressed sensing technology:
Cell image is extracted by column first and constitutes cell image sample column vector;
The perception matrix A ∈ R that training sample cell image in database is constitutedm×nAs the calculation matrix of compressed sensing,
Middle m is sample characteristics dimension, and n is sample size;
According to compressive sensing theory, to real-time collected cell image sample y ∈ Rm, by solving optimal l1Norm constructs
Sparse coefficient x, solution mode are
3) cell recognition calculates the difference r of the linear weighted function of cell to be identified and every a kind of training sample celli(y), it is poor to select
It is worth classification of the generic of the smallest a kind of training sample as test sample, specific formula for calculation is;
In formula,Indicate the cell sparse coefficient x and i-th to be identified extracted
The corresponding coefficient of class cell image.
5. the image detecting method in cell differentiation as described in claim 1, which is characterized in that image segmentation dividing method is into one
Step includes:
(1) a width cell TCT image is inputted, it is pre- to carry out gradation conversion, median filtering denoising and contrast enhancing etc. to image
Processing;
(2) processing of piecemeal Threshold segmentation is carried out to all images, Threshold segmentation is carried out using OTSU method, is partitioned into prospect and back
Scape;
(3) image shape test is carried out, contexts image distinctness is directly placed into result output set to be detected, does not pass through inspection
The carry out adaptive threshold fuzziness again surveyed;
(4) result divided twice is merged, just obtains final ideal segmented image set.
6. a kind of computer program, which is characterized in that described in the computer program operation Claims 1 to 5 any one
Image detecting method in cell differentiation.
7. a kind of terminal, which is characterized in that the terminal, which is at least carried, realizes cell described in Claims 1 to 5 any one point
The controller of image detecting method in change.
8. a kind of computer readable storage medium, including instruction, when run on a computer, so that computer is executed as weighed
Benefit requires the image detecting method in cell differentiation described in 1-5 any one.
9. the image detection system in a kind of cell differentiation for realizing the image detecting method in cell differentiation described in claim 1
System, which is characterized in that the image detecting system in the cell differentiation includes:
Marker gene import modul, connect with main control module, for the promoter for being imported with differentiation state detection marker gene
The cell of the fusion of region and light generating protein gene carries out culture processes;
Induction differentiation module, connect, for inducing for the cell after 1 process of marker gene import modul with main control module
It is cultivated under conditions of differentiation;
Image capture module is connect with main control module, for in cell as caused by the expression of light generating protein gene shine
Image is shot;
Main control module, with marker gene import modul, induction differentiation module, image capture module, picture recognition module, image point
Module, data memory module, display module connection are cut, is worked normally for controlling modules;
Picture recognition module is connect with main control module, for information such as image recognition cell class states to acquisition;
Image segmentation module is connect with main control module, filters out nucleus figure for being split to the image of acquisition;
Data memory module is connect with main control module, for storing the image data information of acquisition;
Display module is connect with main control module, for showing the cell image data information of acquisition.
10. a kind of cell differentiation image detection platform, which is characterized in that the cell differentiation image detection platform at least carries power
Benefit require 9 described in image detecting system in cell differentiation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810913993.XA CN109191434A (en) | 2018-08-13 | 2018-08-13 | Image detecting system and detection method in a kind of cell differentiation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810913993.XA CN109191434A (en) | 2018-08-13 | 2018-08-13 | Image detecting system and detection method in a kind of cell differentiation |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109191434A true CN109191434A (en) | 2019-01-11 |
Family
ID=64921558
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810913993.XA Pending CN109191434A (en) | 2018-08-13 | 2018-08-13 | Image detecting system and detection method in a kind of cell differentiation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109191434A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109568047A (en) * | 2018-11-26 | 2019-04-05 | 焦建洪 | A kind of Cardiological intelligence bed special, control system and control method |
CN109948429A (en) * | 2019-01-28 | 2019-06-28 | 上海依智医疗技术有限公司 | Image analysis method, device, electronic equipment and computer-readable medium |
CN110211117A (en) * | 2019-05-31 | 2019-09-06 | 广东世纪晟科技有限公司 | The processing system of identification line tube and the method for Optimized Segmentation in medical image |
CN111855509A (en) * | 2020-07-27 | 2020-10-30 | 淄博市中心医院 | Cell monitoring system based on bone marrow cell differentiation |
CN113263149A (en) * | 2021-05-12 | 2021-08-17 | 燕山大学 | Device and method for detecting and controlling liquid level of molten pool in double-roller thin strip vibration casting and rolling |
CN113580160A (en) * | 2021-08-05 | 2021-11-02 | 中南大学 | Domestic intelligent nursing robot based on big data |
CN116703909A (en) * | 2023-08-07 | 2023-09-05 | 威海海泰电子有限公司 | Intelligent detection method for production quality of power adapter |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101719277A (en) * | 2009-12-31 | 2010-06-02 | 华中科技大学 | Method for partitioning genetic fuzzy clustering image |
CN102822333A (en) * | 2010-03-23 | 2012-12-12 | 奥林巴斯株式会社 | Method for monitoring state of differentiation in stem cells |
CN104850860A (en) * | 2015-05-25 | 2015-08-19 | 广西师范大学 | Cell image recognition method and cell image recognition device |
CN105894562A (en) * | 2016-04-01 | 2016-08-24 | 西安电子科技大学 | Absorption and scattering coefficient reconstruction method in optical projection tomography |
CN107644210A (en) * | 2017-09-22 | 2018-01-30 | 哈尔滨工业大学(威海) | Micro organism quantity evaluation method based on image procossing |
CN107808381A (en) * | 2017-09-25 | 2018-03-16 | 哈尔滨理工大学 | A kind of unicellular image partition method |
WO2018122932A1 (en) * | 2016-12-26 | 2018-07-05 | オリンパス株式会社 | Cell sorting method, method for manufacturing purified cell population, and luminescence imaging system |
-
2018
- 2018-08-13 CN CN201810913993.XA patent/CN109191434A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101719277A (en) * | 2009-12-31 | 2010-06-02 | 华中科技大学 | Method for partitioning genetic fuzzy clustering image |
CN102822333A (en) * | 2010-03-23 | 2012-12-12 | 奥林巴斯株式会社 | Method for monitoring state of differentiation in stem cells |
US20130017570A1 (en) * | 2010-03-23 | 2013-01-17 | Olympus Corporation | Method for monitoring state of differentiation in stem cell |
CN104850860A (en) * | 2015-05-25 | 2015-08-19 | 广西师范大学 | Cell image recognition method and cell image recognition device |
CN105894562A (en) * | 2016-04-01 | 2016-08-24 | 西安电子科技大学 | Absorption and scattering coefficient reconstruction method in optical projection tomography |
WO2018122932A1 (en) * | 2016-12-26 | 2018-07-05 | オリンパス株式会社 | Cell sorting method, method for manufacturing purified cell population, and luminescence imaging system |
CN107644210A (en) * | 2017-09-22 | 2018-01-30 | 哈尔滨工业大学(威海) | Micro organism quantity evaluation method based on image procossing |
CN107808381A (en) * | 2017-09-25 | 2018-03-16 | 哈尔滨理工大学 | A kind of unicellular image partition method |
Non-Patent Citations (3)
Title |
---|
周颖等: "基于X射线的小麦三维精准模型的构建", 《中国粮油学报》 * |
杜飞明等: "基于十字线灰度图像清晰度模型的构建与应用", 《计算技术与自动化》 * |
范瑜: "论优化遗传算法的模糊聚类在图像分割算法应用", 《电子测试》 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109568047A (en) * | 2018-11-26 | 2019-04-05 | 焦建洪 | A kind of Cardiological intelligence bed special, control system and control method |
CN109948429A (en) * | 2019-01-28 | 2019-06-28 | 上海依智医疗技术有限公司 | Image analysis method, device, electronic equipment and computer-readable medium |
CN110211117A (en) * | 2019-05-31 | 2019-09-06 | 广东世纪晟科技有限公司 | The processing system of identification line tube and the method for Optimized Segmentation in medical image |
CN111855509A (en) * | 2020-07-27 | 2020-10-30 | 淄博市中心医院 | Cell monitoring system based on bone marrow cell differentiation |
CN113263149A (en) * | 2021-05-12 | 2021-08-17 | 燕山大学 | Device and method for detecting and controlling liquid level of molten pool in double-roller thin strip vibration casting and rolling |
CN113580160A (en) * | 2021-08-05 | 2021-11-02 | 中南大学 | Domestic intelligent nursing robot based on big data |
CN116703909A (en) * | 2023-08-07 | 2023-09-05 | 威海海泰电子有限公司 | Intelligent detection method for production quality of power adapter |
CN116703909B (en) * | 2023-08-07 | 2023-10-27 | 威海海泰电子有限公司 | Intelligent detection method for production quality of power adapter |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109191434A (en) | Image detecting system and detection method in a kind of cell differentiation | |
CN110287932B (en) | Road blocking information extraction method based on deep learning image semantic segmentation | |
Li et al. | Automatic organ-level point cloud segmentation of maize shoots by integrating high-throughput data acquisition and deep learning | |
CN113344849B (en) | Microemulsion head detection system based on YOLOv5 | |
Rahaman et al. | An efficient multilevel thresholding based satellite image segmentation approach using a new adaptive cuckoo search algorithm | |
Versari et al. | Long-term tracking of budding yeast cells in brightfield microscopy: CellStar and the Evaluation Platform | |
CN111028327B (en) | Processing method, device and equipment for three-dimensional point cloud | |
CN109919241B (en) | Hyperspectral unknown class target detection method based on probability model and deep learning | |
US11080830B2 (en) | Systems and methods for segmentation and analysis of 3D images | |
CN108426994A (en) | Digital holographic microscopy data are analyzed for hematology application | |
CN109344917A (en) | A kind of the species discrimination method and identification system of Euproctis insect | |
Chen et al. | Agricultural remote sensing image cultivated land extraction technology based on deep learning | |
US11804029B2 (en) | Hierarchical constraint (HC)-based method and system for classifying fine-grained graptolite images | |
CN112819821A (en) | Cell nucleus image detection method | |
JP6882329B2 (en) | Spatial index creation for IHC image analysis | |
KR102624956B1 (en) | Method for detecting cells with at least one malformation in a cell sample | |
CN115295154A (en) | Tumor immunotherapy curative effect prediction method and device, electronic equipment and storage medium | |
CN103279944A (en) | Image division method based on biogeography optimization | |
CN108154513A (en) | Cell based on two photon imaging data detects automatically and dividing method | |
CN113096080B (en) | Image analysis method and system | |
CN110490159A (en) | Identify method, apparatus, equipment and the storage medium of the cell in micro-image | |
Wang et al. | Classification and extent determination of rock slope using deep learning | |
Cao et al. | 3D convolutional neural networks fusion model for lung nodule detection onclinical CT scans | |
CN112071423B (en) | Immunochromatography concentration detection method and system based on machine learning | |
Mayerich et al. | Fast cell detection in high-throughput imagery using GPU-accelerated machine learning |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190111 |