CN109117937A - A kind of Leukocyte Image processing method and system subtracted each other based on connection area - Google Patents
A kind of Leukocyte Image processing method and system subtracted each other based on connection area Download PDFInfo
- Publication number
- CN109117937A CN109117937A CN201810936991.2A CN201810936991A CN109117937A CN 109117937 A CN109117937 A CN 109117937A CN 201810936991 A CN201810936991 A CN 201810936991A CN 109117937 A CN109117937 A CN 109117937A
- Authority
- CN
- China
- Prior art keywords
- image
- area
- leukocyte
- leucocyte
- connected region
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06M—COUNTING MECHANISMS; COUNTING OF OBJECTS NOT OTHERWISE PROVIDED FOR
- G06M11/00—Counting of objects distributed at random, e.g. on a surface
-
- 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
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10056—Microscopic image
- G06T2207/10061—Microscopic image from scanning electron microscope
-
- 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)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (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)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention belongs to basic ecsomatics technical fields, disclose a kind of Leukocyte Image processing method and system subtracted each other based on connection area, and to the label of white pixel in bianry image, each individually connected region forms an identified block；After connected component labeling, the area of each connected region is obtained, area is deleted using MATLAB function bwareaopen and obtains connected region less than certain value；In comprising particle and agranular leukocyte original image, utilize a kind of classical and useful image partition method -- dam algorithm, namely watershed, agranular leukocyte is split and counted, image only comprising granular leucocyte is obtained using the agranular leukocyte part in function imsubtract removal original image, the granular leucocyte gross area is recycled to obtain granular leucocyte quantity divided by average area.The present invention can clearly distinguish granular leucocyte and agranular leukocyte, and image clearly, it is accurate to count.
Description
Technical field
The invention belongs to basic ecsomatics technical field more particularly to a kind of Leukocyte Images subtracted each other based on connection area
Processing method and system.
Background technique
Currently, the prior art commonly used in the trade is such that
Leucocyte is colourless karyocyte, common in normal peripheral blood to have neutrophil leucocyte, eosinophil, thermophilic
Alkaline granulocyte, lymphocyte and monocyte.Leucocyte is very important one kind haemocyte in blood of human body.Leucocyte exists
Many important tasks are undertaken in human body, it have the function of swallow foreign matter and generate antibody, body injury healing ability, resist cause of disease
Body invasion ability, to Immunoresistance of disease etc..Leucocyte can also be subdivided into five seed types in fact, use instrument or people
Work method counts these five types of cells respectively, referred to as Arneth's count.Neutrophil cell accounts in these five types of leucocytes
0.5~0.7, lymphocyte accounts for 0.2~0.4, and monocyte accounts for 0.03~0.08, and eosinophil accounts for 0.01~0.05, thermophilic
Alkaline granulocyte is no more than 0.01.Human body is not in due course, is often showed by the significant changes of quantity of leucocyte.Normally
It is worth term of reference are as follows: adult (3.50~9.50) × 109/ L (3500~9500/mm3)；Children (5.0~12.0) × 109/L
(5000~12000/mm3)；Baby (10~22.0) × 109/ L (10000~22000/mm3).White blood cell count(WBC) refers to counting
Number of white blood cells contained in unit volume blood.It is once called as white blood cell, is the important component of body defending system.It is existing
There are many kinds of white blood cell count(WBC) methods.
Manual count method be by various samples after treatment, rush pond on tally, so that leucocyte is sunk after standing, then
The leucocyte on tally is counted by optical microscopy, is averaged after counting multiple small lattice, then is converted into required
Unit obtains report.Since manual count method detection process is complicated, and subjectivity is strong, is often easy to influence count results.
In conclusion problem of the existing technology is:
(1) be directly split counting using watershed algorithm, count accuracy is not high, can not especially divide be adhered it is non-
Normal serious cell, and it be easy to cause over-segmentation
(2) prior art can only calculate total number of cells, can not carry out differential counting to leucocyte, cannot enough clearly
Distinguish granular leucocyte and agranular leukocyte；
(3) without complete program code, user, which can not directly extract test and obtain final segmentation, counts effect.
Solve the difficulty and meaning of above-mentioned technical problem:
Difficulty:
1: count accuracy is not high, can not especially divide and be adhered cell very serious, and be easy to cause over-segmentation
2: total number of cells can only be calculated, differential counting can not be carried out to leucocyte, such counting is only used for learning
Research, it is impossible to be used in clinical practice.
3: without complete program code, user, which can not directly extract test and obtain final segmentation, counts effect
Meaning:
There is good segmentation effect while the present invention improves count accuracy and can divide to being adhered severe cellular
Class particle count leucocyte and non-particulate leucocyte.The quantity for calculating various types of cells in blood cell thus has practical meaning
Justice.And there is complete code and be encapsulated and be designed to gui interface, user, which can directly use, can also read code and in this base
Program is improved on plinth.
Summary of the invention
In view of the problems of the existing technology, the present invention provides at a kind of Leukocyte Image subtracted each other based on connection area
Manage method and system,
The invention is realized in this way
It is a kind of based on the Leukocyte Image processing method subtracted each other of connection area, it is described subtracted each other based on connection area it is white thin
Born of the same parents' image processing method includes:
Step 1, connected component labeling allow each individual connected region shape to the label of white pixel in bianry image
At an identified block；
Specific implementation method: using bwlabel function in Matlab to white pixel [mesh in blood cell binary image
Mark] be marked, allow each individual connected region to form an identified block, specific code are as follows: [P, num1]=
Bwlabel (I1,8), wherein P is the image after label；Num1 is connected region number；I1 is original image to be marked；8 be ginseng
Number calculates connected region using 8 neighborhoods here
Step 2 calculates the area of every piece of connected region, deletes area using MATLAB function bwareaopen and is less than
The connected region of certain value, the granular leucocyte being adhered and non-particulate Leukocyte Image；
Specific implementation method: every piece of connected region being labeled in previous step is calculated using regionprops function
Area, the following stats1=regionprops of specific code (P, ' Area'), wherein stats1 is a data matrix, is saved
Each connected region size (number of pixels)；P is the image after being labeled obtained in previous step；Area is parameter, is used
Regionprops function calculates connected region area.It obtains each connected region area and then utilizes bwareaopen letter
Number deletes connected region of the area less than 20000, and specific code is LIBO=bwareaopen (P, 20000,8), wherein LIBO
To delete the image after small area；P is image to be deleted；20000 is make parameter by oneself, for deleting area less than 20000
Connected region；8 be parameter, calculates connected region using 8 neighborhoods here.
Step 3 utilizes a kind of classical and useful image partition method -- dam algorithm, that is,
Watershed splits agranular leukocyte；
Specific implementation method: specific function is as follows:
D=-bwdist (~LIBO)；
Mask=imextendedmin (D, 2)；
D2=imimposemin (D, mask)；
Ld=watershed (D2)；
Water_splited=LIBO；
Water_splited (Ld==0)=0；
In order to allow watershed function to obtain best effects when being split, first with bwdist function to image
Range conversion is carried out, the region of some especially small (referring to that region is small) is filtered out using imextendedmin this function；It uses
Imextendedmin will only want to generate dot among the block of segmentation.Finally divide to obtain using watershed function
Non-particulate Leukocyte Image.
Non-particulate leucocyte is partitioned into simultaneously by step 4, image subtraction in comprising particle and non-particulate leucocyte original image
It counts, the agranular leukocyte part in function imsubtract removal original image provided using Matlab is obtained only comprising particle
The image of leucocyte recycles the gross area to obtain granular leucocyte quantity divided by average area, realizes granular leucocyte and non-
Grain Arneth's count.
Specific implementation method: it will be subtracted (comprising granular leucocyte and non-particulate Leukocyte Image) obtained in the previous step
Non-particulate Leukocyte Image obtains granular leucocyte image.Specific code is as follows: Z=imsubtract (X, Y), wherein Z be
The granular leucocyte image arrived, X include the original image of granular leucocyte and non-particulate leucocyte, and Y is obtained in previous step
Non-particulate Leukocyte Image.After obtaining non-particulate Leukocyte Image, code is utilized
Sum is the area summation of all granular leucocytes, and particle can be obtained in the area divided by individual particle leucocyte
White thin white number, to realize the differential counting of granular leucocyte and non-particulate leucocyte.
Another object of the present invention is to provide the Leukocyte Image processing side subtracted each other described in a kind of realize based on connection area
The computer program of method.
Another object of the present invention is to provide the Leukocyte Image processing side subtracted each other described in a kind of realize based on connection area
The information data processing terminal of 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 Leukocyte Image processing method subtracted each other based on connection area.
Another object of the present invention is to provide the Leukocyte Image processing side subtracted each other described in a kind of realize based on connection area
The Leukocyte Image processing system of method subtracted each other based on connection area, the Leukocyte Image processing subtracted each other based on connection area
System includes:
Connected component labeling module allows each individual connected region to be formed the label of white pixel in bianry image
One identified block；
The granular leucocyte being adhered and non-particulate Leukocyte Image obtain module, are deleted using MATLAB function bwareaopen
Connected region is obtained except area is less than certain value, the granular leucocyte and non-particulate Leukocyte Image being adhered；
Agranular leukocyte divides module, utilizes image partition method -- dam algorithm, that is, watershed, by nothing
Granular leucocyte is split；
Non-particulate leucocyte is partitioned into and is counted in comprising particle and non-particulate leucocyte original image by image subtraction module
Number, the agranular leukocyte part in function imsubtract removal original image provided using Matlab obtain only white comprising particle
The image of cell recycles the gross area to obtain granular leucocyte quantity divided by average area, realizes granular leucocyte and non-particulate
Arneth's count.
Another object of the present invention is to provide the Leukocyte Image processing system subtracted each other described in a kind of carrying based on connection area
The basic ecsomatics processing platform of system.
In conclusion advantages of the present invention and good effect are as follows:
For the present invention, chooses 10 Blood Corpuscle Images that 10 cell distributions differ greatly and carries out data analysis:
Table counting statistics
Interpretation of result: from the point of view of the counting to ten experiment sample images, to preceding 5 images being adhered substantially without leucocyte
For count accuracy rate and almost reach to 100%, and can accurately distinguish very much agranular leukocyte and granular leucocyte, it is right
Other five are adhered for extremely serious cell image, and counting has certain error, but accuracy rate is also all up to 90% or more,
Because this program, which is used, removes the small connected region of area first with corrosion and watershed algorithm, no particle is first calculated
Quantity of leucocyte, then obtain granular leucocyte quantity divided by average area with the gross area, wherein the parameter used it is more (first is that:
Bwareaopen function twice is utilized, and, come the connected region for deleting area less than certain value, this area threshold is user oneself
It is fixed to take, and should all delete the connected domain of desired deletion, and cannot accidentally delete the connected region for wanting reservation；Second is that: particle
The method of counting of leucocyte and the method for counting of non-particulate leucocyte are different, and the number of non-particulate leucocyte utilizes connected region
Number determines, and granular leucocyte is to obtain granular leucocyte quantity divided by average area with the gross area, this average area
Value or so the accuracy that counts of agranular leukocyte), accuracy can further be improved by continuing adjustment parameter.
The present invention can clearly distinguish granular leucocyte and agranular leukocyte, and image clearly, it is accurate to count.
2: prior art effect and the technology of the present invention Contrast on effect
3: the gui interface for making MATLAB carries out the effect picture 13 of emulation experiment to invention.
Detailed description of the invention
Fig. 1 is the flow chart of the Leukocyte Image processing method provided in an embodiment of the present invention subtracted each other based on connection area.
Fig. 2 is connection algorithm basic schematic diagram provided in an embodiment of the present invention.
Fig. 3 is four field provided in an embodiment of the present invention schematic diagram.
Fig. 4 is eight fields schematic diagram provided in an embodiment of the present invention
Fig. 5 is that A provided in an embodiment of the present invention with C is connected to enlarged diagram.
Fig. 6 is the bianry image schematic diagrames before processing provided in an embodiment of the present invention.
Fig. 7 is the bianry image schematic diagrames after deletion small area provided in an embodiment of the present invention.
Fig. 8 is provided in an embodiment of the present invention comprising particle and agranular leukocyte schematic diagram.
Fig. 9 is provided in an embodiment of the present invention and includes agranular leukocyte schematic diagram.
Figure 10 is provided in an embodiment of the present invention and includes granular leucocyte schematic diagram.
Figure 11 is schematic diagram after watershed segmentation provided in an embodiment of the present invention.
Figure 12 is partial enlargement diagram after watershed segmentation provided in an embodiment of the present invention.
Figure 13 is blood cell processing provided in an embodiment of the present invention and counting effect picture.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to embodiments, to the present invention
It is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to
Limit the present invention.
Further very thin explanation is carried out to the present invention with reference to the accompanying drawing.
As shown in Figure 1, a kind of Leukocyte Image processing method subtracted each other based on connection area provided by the invention includes:
S101: connected component labeling is allowed using bwlabel function to the label of white pixel in bianry image (target)
Each individually connected region forms an identified block, and is calculated using regionprops function each identified
The number of pixels (area) of block；
S102: connection area subtracts each other, after connected component labeling, using bwlabel function and regionprops function to every
A connected component labeling and after calculating the number of pixels (area) of each identified block, utilizes MATLAB function
Bwareaopen deletes the connected region that area is less than certain value, and the bianry image, Fig. 7 before effect such as Fig. 6 processing delete facet
Bianry image after product；
S103: utilizing a kind of classical and useful image partition method -- dam algorithm, that is, watershed,
Agranular leukocyte is split, provided in an embodiment of the present invention point of schematic diagram, Figure 12 after effect such as Figure 11 watershed segmentation
Partial enlargement diagram after the segmentation of water ridge.
It is separated that this step can will be adhered leucocyte very serious, due to granular leucocyte and agranular leukocyte this
Two class cell maximum differences are in nucleus, and granular leucocyte nucleus is in granular form more, generally by the lesser graininess group of area
At the nucleus that one is connected to can be divided into the connected region of several small areas with watershed segmentation method by such shape
Domain；And the nucleus of agranular leukocyte is formed in usually one biggish rounded or similar round connected region, this
The shape watershed segmentation method of sample will not be divided into fritter, and area is much larger than non-particulate leucocyte so can use connection
Region area extracts granular leucocyte.By image after the segmentation of figure 11 above geomantic omen ridge, then MATLAB function is utilized to it
Bwareaopen deletes the connected region that area is less than certain value, just deletes small area connected region, what is left is exactly to divide
Non-particulate Leukocyte Image after the completion, as shown in figure 12.
S104: image subtraction divides agranular leukocyte in comprising granular leucocyte and agranular leukocyte original image
It out and counts, the agranular leukocyte part in function imsubtract removal original image provided using Matlab is only included
The figure of granular leucocyte and the figure comprising non-particulate leucocyte；
After being partitioned into agranular leukocyte and having counted, need particle count leucocyte, at this moment we just need will be former
Agranular leukocyte part in figure is removed, and present invention employs the method for image subtraction, the function that Matlab is provided is
Imsubtract, Figure 10 provided in an embodiment of the present invention is exactly to utilize this function by Fig. 8 sheet comprising granular leucocyte schematic diagram
What inventive embodiments provided, which is provided in an embodiment of the present invention comprising particle and agranular leukocyte schematic diagram figure and Fig. 9, includes
What agranular leukocyte schematic diagram obtained after subtracting each other.
Then it calculates Figure 10 provided in an embodiment of the present invention and includes granular leucocyte in granular leucocyte schematic diagram
The gross area can be obtained by the number of granular leucocyte divided by the average area of individual particle leucocyte, so that it is white to realize particle
The differential counting of cell and non-particulate leucocyte.
As shown in Fig. 2, connection algorithm basic principle are as follows:
The first row obtains two groups: [2,6] and [10,13], while being labeled as 1 and 2.Second row obtains two groups: [6,
7] and [9,10], but they all have overlapping region with the group of lastrow, so being marked with the group of lastrow, i.e., 1 and 2.Third
Row, two: [2,4] and [7,8].[2,4] this group and the unfolded group of lastrow, so giving its new mark is 3；
And [2,4] this group have with two groups of lastrow it is overlapping, so give the smallest label in its two, i.e., 1, then will
(1,2) write-in is of equal value right.All images traversal terminates, and obtains the origin coordinates of many groups, terminating coordinates, the row where them
And their label, while having obtained the list of an equivalence pair.
As in Figure 3-5, connected component labeling:
In the picture, the smallest unit is pixel, there is 8 adjacent pixels around each pixel, and common syntople has 2
Kind: 4 adjacent and 8 adjoinings.
As shown in figure 3,4 adjacent 4 points altogether, i.e., up and down；
As shown in figure 4,8 adjacent points one share 8, it include the point of diagonal positions；
If pixel A and B are adjacent, we claim A to be connected to B, and what then we were not added proof has following conclusion: if
A is connected to B, and B is connected to C, then A is connected to C.Visually apparently, the point to communicate with each other forms a region, without being connected to
Point form different regions.Such a all points communicate with each other the set constituted, a referred to as connected region.
As shown in figure 5, then having 3 connected regions if it is considered that 4 is adjacent；If it is considered that 8 is adjacent, then there are 2 connected regions
Domain.(note: image is the effect being amplified, and image square is practical to only have 4 pixels).
As shown in Figure 6 and Figure 7, small area connected region is deleted:
After connected component labeling, the area of each connected region is obtained, deletes face using MATLAB function bwareaopen
Product is less than certain value and obtains connected region.
As seen in figs. 8-10, image subtraction
In comprising particle and agranular leukocyte original image, agranular leukocyte is partitioned into and is counted, Matlab is utilized
Agranular leukocyte part in the function ims ubtract removal original image of offer obtains figure only comprising granular leucocyte.
As is illustrated by figs. 11 and 12, a kind of classical and useful image partition method is utilized -- dam algorithm, also
It is watershed, non-particulate leucocyte is split.
Figure 13 is blood cell processing provided in an embodiment of the present invention and counting effect picture.
Working principle part:
Granular leucocyte and non-particulate leucocyte are carried out differential counting by the present invention, by leucocyte feature difference can obtain this two
In nucleus, granular leucocyte nucleus is in granular form class cell maximum difference more, after binary conversion treatment generally by area compared with
The nucleus that one is connected to can be divided into several facets with watershed segmentation method by small graininess composition, such shape
Long-pending connected region；Rather than the nucleus of granular leucocyte is usually a biggish rounded or class after binary conversion treatment
Circular connected region composition, such shape watershed segmentation method will not be divided into fritter, it is possible to utilize two-value
The connected region size of image judges which kind of leucocyte belongs to, so that the classification of leucocyte is realized, for the counting of lower step
It is ready.
1, there are many kinds of connected component labeling algorithms, some algorithms can once traverse image and complete label, and some then needs
Want 2 times or more traversal images.This has also resulted in the difference of different algorithm time efficiencies, introduces 2 kinds of algorithms herein.
The first algorithm is the algorithm made in connected component labeling function bwlabel in present matlab, its primary traversal
Image, and write down in every a line (or column) that continuously the equivalence of group (run) and label is right, then by of equal value to original figure
As being re-flagged, this algorithm is several highest one of middle efficiency that I attempts at present.
Second algorithm is labeling algorithm used in present open source library cvBlob, it is inside and outside by positioning connected region
Profile marks whole image, and the core of this algorithm is the searching algorithm of profile.This algorithm is compared imitates with first method
Want lower in rate, but in connected region number when within 100, the two almost indifference, when connected region is counted to
When 103 order of magnitude, algorithm above can be 10 times faster than the algorithm or more.
The function that Matlab is provided are as follows:
[L, num]=bwlabel (Image_BW_medfilt2,8)；
2, the two methods of connected region area are calculated
Method one: the regionprops of the connected region area of bianry image after the calculating label carried using MATLAB
Function is being sequentially output number of pixels shared by each connected region, and specific MATLAB function is as follows:
Method two: user can customize function, calculate the area summation of all connected regions of bianry image, specific code
It is as follows:
The invention will be further described combined with specific embodiments below.
It is provided in an embodiment of the present invention to include: based on the Leukocyte Image processing system subtracted each other of connection area
Connected component labeling allows each individual connected region to form one the label of white pixel in bianry image
Identified block；
Area is deleted using MATLAB function bwareaopen and obtains connected region less than certain value, and the particle being adhered is white
Cell and non-particulate Leukocyte Image；
Utilize a kind of classical and useful image partition method -- dam algorithm, that is, watershed, will without
Grain leucocyte is split；
Non-particulate leucocyte is partitioned into and is counted in comprising particle and non-particulate leucocyte original image by image subtraction, benefit
The agranular leukocyte part in function imsubtract removal original image provided with Matlab obtains only comprising granular leucocyte
Image, recycle the gross area to obtain granular leucocyte quantity divided by average area, realize granular leucocyte and non-particulate white thin
Born of the same parents' differential counting.
The Leukocyte Image processing system provided in an embodiment of the present invention subtracted each other based on connection area, comprising:
Connected component labeling module allows each individual connected region to be formed the label of white pixel in bianry image
One identified block；
The granular leucocyte being adhered and non-particulate Leukocyte Image obtain module, are deleted using MATLAB function bwareaopen
Connected region is obtained except area is less than certain value, the granular leucocyte and non-particulate Leukocyte Image being adhered；
Agranular leukocyte divides module, utilizes image partition method -- dam algorithm, that is, watershed, by nothing
Granular leucocyte is split；
Non-particulate leucocyte is partitioned into and is counted in comprising particle and non-particulate leucocyte original image by image subtraction module
Number, the agranular leukocyte part in function imsubtract removal original image provided using Matlab obtain only white comprising particle
The image of cell recycles the gross area to obtain granular leucocyte quantity divided by average area, realizes granular leucocyte and non-particulate
Arneth's count.
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 foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
1. a kind of Leukocyte Image processing method subtracted each other based on connection area, which is characterized in that described based on connection area
The Leukocyte Image processing method subtracted each other includes:
Step 1, connected component labeling allow each individual connected region to form one the label of white pixel in bianry image
A identified block；
Step 2 calculates the area of every piece of connected region, deletes area using MATLAB function bwareaopen and is less than centainly
It is worth connected region, the granular leucocyte being adhered and non-particulate Leukocyte Image；
Step 3 utilizes image partition method -- and dam algorithm splits agranular leukocyte；
Non-particulate leucocyte is partitioned into and is counted in comprising particle and non-particulate leucocyte original image by step 4, image subtraction
Number, the agranular leukocyte part in function imsubtract removal original image provided using Matlab obtain only white comprising particle
The image of cell recycles the gross area to obtain granular leucocyte quantity divided by average area.
2. the Leukocyte Image processing method subtracted each other as described in claim 1 based on connection area, which is characterized in that step 1
It specifically includes:
White pixel in blood cell binary image is marked using bwlabel function in Matlab, it is each individual
Connected region forms an identified block；
Bwlabel function are as follows: [P, num1]=bwlabel (I1,8), wherein P is the image after label；Num1 is connected region
Number；I1 is original image to be marked；8 be parameter, calculates connected region using 8 neighborhoods.
3. the Leukocyte Image processing method subtracted each other as described in claim 1 based on connection area, which is characterized in that step 2
It specifically includes:
The area of every piece of labeled connected region, regionprops function are calculated using regionprops function are as follows:
Stats1=regionprops (P, ' Area'), wherein stats1 is a data matrix, and it is big to save each connected region area
It is small；P be it is labeled after image；Area is parameter, and connected region area is calculated with regionprops function；
It obtains each connected region area and then deletes connected region of the area less than 20000 using bwareaopen function,
Bwareaopen function is LIBO=bwareaopen (P, 20000,8), and wherein LIBO is the image deleted after small area；P is
Image to be deleted；20000 is make parameter by oneself, for deleting connected region of the area less than 20000；8 be parameter, using 8 neighbours
Domain calculates connected region.
4. the Leukocyte Image processing method subtracted each other as described in claim 1 based on connection area, which is characterized in that step 3
Image partition method -- dam algorithm includes:
D=-bwdist (~LIBO)；
Mask=imextendedmin (D, 2)；
D2=imimposemin (D, mask)；
Ld=watershed (D2)；
Water_splited=LIBO；
Water_splited (Ld==0)=0；
Range conversion is carried out to image first with bwdist function, is filtered out using imextendedmin function especially small
Region；Watershed function is finally reused to divide to obtain non-particulate Leukocyte Image.
5. the Leukocyte Image processing method subtracted each other as described in claim 1 based on connection area, which is characterized in that step 4
It specifically includes:
It will subtract comprising granular leucocyte and non-particulate Leukocyte Image in obtained non-particulate Leukocyte Image, it is white to obtain particle
Cell image；
Function imsubtract are as follows: Z=imsubtract (X, Y), the granular leucocyte image that wherein Z is, X include particle
The original image of leucocyte and non-particulate leucocyte, the non-particulate Leukocyte Image that Y is；Obtain non-particulate Leukocyte Image
Later, code is utilized
Sum=0；It is gross area variable that %, which defines sum,；
For i=1:num%num is granular leucocyte image connectivity areal；
[J1, J2]=find (M==i)；% finds out the pixel coordinate that J intermediate value is i and is stored in J1 respectively, in J2；
Area (i)=length (J1)；%area (i) is the area of i-th of red blood cell；
Sum=sum+area (i)；% area accumulation；
end；
Sum is the area summation of all granular leucocytes, then divided by the area of individual particle leucocyte, it is white white thin to obtain particle
Number.
6. a kind of realize described in Claims 1 to 5 any one based on the Leukocyte Image processing method subtracted each other of connection area
Computer program.
7. a kind of realize described in Claims 1 to 5 any one based on the Leukocyte Image processing method subtracted each other of connection area
Information data processing terminal.
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 Leukocyte Image processing method subtracted each other described in 1~5 any one based on connection area.
9. it is a kind of realize described in claim 1 based on the Leukocyte Image processing method subtracted each other of connection area based on connection area
The Leukocyte Image processing system subtracted each other, which is characterized in that the Leukocyte Image processing system subtracted each other based on connection area
Include:
Connected component labeling module allows each individual connected region to form one the label of white pixel in bianry image
Identified block；
The granular leucocyte being adhered and non-particulate Leukocyte Image obtain module, delete face using MATLAB function bwareaopen
Product is less than certain value and obtains connected region, the granular leucocyte and non-particulate Leukocyte Image being adhered；
Agranular leukocyte divides module, utilizes image partition method -- dam algorithm, that is, watershed, it will be without particle
Leucocyte is split；
Non-particulate leucocyte is partitioned into and is counted in comprising particle and non-particulate leucocyte original image by image subtraction module, benefit
The agranular leukocyte part in function imsubtract removal original image provided with Matlab obtains only comprising granular leucocyte
Image, recycle the gross area to obtain granular leucocyte quantity divided by average area, realize granular leucocyte and non-particulate white thin
Born of the same parents' differential counting.
10. a kind of carried at the Leukocyte Image processing system basis ecsomatics subtracted each other described in claim 9 based on connection area
Platform.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810936991.2A CN109117937B (en) | 2018-08-16 | 2018-08-16 | Leukocyte image processing method and system based on connected area subtraction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810936991.2A CN109117937B (en) | 2018-08-16 | 2018-08-16 | Leukocyte image processing method and system based on connected area subtraction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109117937A true CN109117937A (en) | 2019-01-01 |
| CN109117937B CN109117937B (en) | 2022-03-22 |
Family
ID=64853395
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810936991.2A Active CN109117937B (en) | 2018-08-16 | 2018-08-16 | Leukocyte image processing method and system based on connected area subtraction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109117937B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886928A (en) * | 2019-01-24 | 2019-06-14 | 平安科技（深圳）有限公司 | A kind of target cell labeling method, device, storage medium and terminal device |
| CN110097510A (en) * | 2019-04-11 | 2019-08-06 | 平安科技（深圳）有限公司 | A kind of pure color flower recognition methods, device and storage medium |
| CN110689057A (en) * | 2019-09-11 | 2020-01-14 | 哈尔滨工程大学 | Method for reducing neural network training sample size based on image segmentation |
| CN111178354A (en) * | 2019-12-23 | 2020-05-19 | 深圳市铁汉生态环境股份有限公司 | Mangrove pest monitoring method and system |
| CN111429440A (en) * | 2020-03-31 | 2020-07-17 | 上海杏脉信息科技有限公司 | Microscopic pathological image cell sufficiency detection method, system, equipment, device and medium |
| CN112560844A (en) * | 2020-12-09 | 2021-03-26 | 南京理工大学智能计算成像研究院有限公司 | Automatic speckle pattern identification method |
| CN112545566A (en) * | 2020-12-07 | 2021-03-26 | 武汉中旗生物医疗电子有限公司 | Fetal neck diaphragmatic layer measuring method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1837819A (en) * | 2006-04-18 | 2006-09-27 | 李奕 | Blood cell analysis method based on machine vision |
| US20070014460A1 (en) * | 2003-11-18 | 2007-01-18 | Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. | Method and apparatus for detecting various cell types of cells in a biological sample |
| CN107492088A (en) * | 2016-06-11 | 2017-12-19 | 青岛华晶生物技术有限公司 | Leucocyte automatic identification and statistical method in a kind of gynaecology's micro-image |
-
2018
- 2018-08-16 CN CN201810936991.2A patent/CN109117937B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070014460A1 (en) * | 2003-11-18 | 2007-01-18 | Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. | Method and apparatus for detecting various cell types of cells in a biological sample |
| CN1837819A (en) * | 2006-04-18 | 2006-09-27 | 李奕 | Blood cell analysis method based on machine vision |
| CN107492088A (en) * | 2016-06-11 | 2017-12-19 | 青岛华晶生物技术有限公司 | Leucocyte automatic identification and statistical method in a kind of gynaecology's micro-image |
Non-Patent Citations (1)
| Title |
|---|
| 陈竞洲: "血液细胞识别分类研究与实现", 《中国优秀博硕士学位论文全文数据库(硕士) 医药卫生科技辑》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886928A (en) * | 2019-01-24 | 2019-06-14 | 平安科技（深圳）有限公司 | A kind of target cell labeling method, device, storage medium and terminal device |
| CN109886928B (en) * | 2019-01-24 | 2023-07-14 | 平安科技（深圳）有限公司 | Target cell marking method, device, storage medium and terminal equipment |
| CN110097510A (en) * | 2019-04-11 | 2019-08-06 | 平安科技（深圳）有限公司 | A kind of pure color flower recognition methods, device and storage medium |
| CN110097510B (en) * | 2019-04-11 | 2023-10-03 | 平安科技（深圳）有限公司 | Pure-color flower identification method, device and storage medium |
| CN110689057A (en) * | 2019-09-11 | 2020-01-14 | 哈尔滨工程大学 | Method for reducing neural network training sample size based on image segmentation |
| CN110689057B (en) * | 2019-09-11 | 2022-07-15 | 哈尔滨工程大学 | Method for reducing neural network training sample size based on image segmentation |
| CN111178354A (en) * | 2019-12-23 | 2020-05-19 | 深圳市铁汉生态环境股份有限公司 | Mangrove pest monitoring method and system |
| CN111178354B (en) * | 2019-12-23 | 2024-02-27 | 中节能铁汉生态环境股份有限公司 | Mangrove pest monitoring method and system |
| CN111429440A (en) * | 2020-03-31 | 2020-07-17 | 上海杏脉信息科技有限公司 | Microscopic pathological image cell sufficiency detection method, system, equipment, device and medium |
| CN111429440B (en) * | 2020-03-31 | 2023-04-28 | 上海杏脉信息科技有限公司 | Method, system, equipment, device and medium for detecting sufficiency of microscopic pathology image cells |
| CN112545566A (en) * | 2020-12-07 | 2021-03-26 | 武汉中旗生物医疗电子有限公司 | Fetal neck diaphragmatic layer measuring method and device |
| CN112560844A (en) * | 2020-12-09 | 2021-03-26 | 南京理工大学智能计算成像研究院有限公司 | Automatic speckle pattern identification method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109117937B (en) | 2022-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109117937A (en) | A kind of Leukocyte Image processing method and system subtracted each other based on connection area | |
| Dzyubachyk et al. | Advanced level-set-based cell tracking in time-lapse fluorescence microscopy | |
| Shan et al. | Automatic skin lesion segmentation based on FC-DPN | |
| Rodriguez et al. | Automated three-dimensional detection and shape classification of dendritic spines from fluorescence microscopy images | |
| Ghosh et al. | Blood smear analyzer for white blood cell counting: a hybrid microscopic image analyzing technique | |
| CN104766324B (en) | Based on the colored blood cell image leucocyte dividing method of super-pixel and abnormality detection | |
| CN111091573B (en) | CT image pulmonary vessel segmentation method and system based on deep learning | |
| CN106340016A (en) | DNA quantitative analysis method based on cell microscope image | |
| EP3475683A1 (en) | Image atlas systems and methods | |
| CN112991263B (en) | Method and equipment for improving TPS (tissue specific differentiation) calculation accuracy of PD-L1 immunohistochemical pathological section | |
| Tantikitti et al. | Image processing for detection of dengue virus based on WBC classification and decision tree | |
| US20230273188A1 (en) | Method and system for detecting cell-killing efficacy and/or immune activity, and application thereof | |
| Zhao et al. | Emds-6: Environmental microorganism image dataset sixth version for image denoising, segmentation, feature extraction, classification, and detection method evaluation | |
| Su et al. | A benchmark bone marrow aspirate smear dataset and a multi-scale cell detection model for the diagnosis of hematological disorders | |
| Kausar et al. | Multi-scale deep neural network for mitosis detection in histological images | |
| Chiang et al. | Analysis of in vivo single cell behavior by high throughput, human-in-the-loop segmentation of three-dimensional images | |
| Lam et al. | WBC-based segmentation and classification on microscopic images: a minor improvement | |
| Li et al. | Exponential distance transform maps for cell localization | |
| Funke et al. | A benchmark for epithelial cell tracking | |
| Hagwood et al. | Evaluation of segmentation algorithms on cell populations using CDF curves | |
| Bergler et al. | Automatic detection of tumor buds in pan-cytokeratin stained colorectal cancer sections by a hybrid image analysis approach | |
| Niederberger et al. | Factor graph analysis of live cell–imaging data reveals mechanisms of cell fate decisions | |
| JPH0584544B2 (en) | ||
| Martin et al. | A graph based neural network approach to immune profiling of multiplexed tissue samples | |
| CN116246100A (en) | Chromosome image interpretable evaluation selection method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |