CN102446353B - For machine vision interpretation method and the device of blood type analysis - Google Patents
For machine vision interpretation method and the device of blood type analysis Download PDFInfo
- Publication number
- CN102446353B CN102446353B CN201010508384.XA CN201010508384A CN102446353B CN 102446353 B CN102446353 B CN 102446353B CN 201010508384 A CN201010508384 A CN 201010508384A CN 102446353 B CN102446353 B CN 102446353B
- Authority
- CN
- China
- Prior art keywords
- image
- microcosmic
- sample
- pixel
- aggegation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention discloses a kind of machine vision interpretation method for blood type analysis and device, wherein said method comprises macroscopical deterministic process and microcosmic deterministic process, the sample being defined as sentencing for macroscopical deterministic process carries out microcosmic judgement further, described microcosmic deterministic process comprises: microcosmic acquisition step, for gathering the micro-image of the rear sample of reaction; Extraction step, for extracting the edge image of gathered micro-image; And microcosmic determining step, by the number of edge calculation image pixel at least one direction, sample pixel number being less than the minimum value of setting is judged to occur aggegation, and the sample being greater than the maximal value of setting is judged to not occur aggegation.When grand design analysis cannot judge whether sample occurs aggegation, micro-image can be utilized to analyze further and to judge whether sample occurs aggegation according to the method and apparatus of the embodiment of the present invention.
Description
Technical field
The present invention relates to a kind of image processing method and device, particularly relate to a kind of machine vision interpretation method for blood type analysis and device.
Background technology
At present, in blood type analytical instrument, be all the analysis only carrying out grand design to the process of reacted sample image, determine that aggegation or not aggegation appear in sample.But, if when in blood type analysis, the situations such as weak solidifying or false set appear in sample, this method of grand design analysis that utilizes just cannot judge, and need carry out artificial interpretation.
Therefore, need a kind of when grand design analysis cannot judge whether sample occurs aggegation, reacted sample image can be utilized further, carry out analyzing the method and apparatus judged.
Summary of the invention
The object of the invention is the defect in order to overcome existing blood type analysis method, there is provided a kind of can when grand design analysis cannot judge whether sample occurs aggegation, utilize reacted sample micro-image further, carry out analyzing the method and apparatus judged.In order to realize this purpose, the technical solution used in the present invention is as follows.
According to the first aspect of the embodiment of the present invention, a kind of machine vision interpretation method for blood type analysis is provided, comprise macroscopical deterministic process and microcosmic deterministic process, the sample being wherein defined as sentencing for macroscopical deterministic process carries out microcosmic judgement further, described microcosmic deterministic process comprises: microcosmic acquisition step, for gathering the micro-image of the rear sample of reaction; Extraction step, for extracting the edge image of gathered micro-image; And microcosmic determining step, by the number of edge calculation image pixel at least one direction, sample pixel number being less than the minimum value of setting is judged to occur aggegation, and the sample being greater than the maximal value of setting is judged to not occur aggegation.According to an embodiment, the minimum value of wherein said setting and the maximal value of described setting depend on cell dimensions in sample, setting aggegation block yardstick and gather micro-image time visual field yardstick.
According to an embodiment, described macroscopical deterministic process comprises: macroscopical acquisition step, for gathering the grand design of the rear sample of reaction; Separating step, isolates RED sector image from gathered grand design; Calculation procedure, for calculating the difference of each pixel and neighborhood territory pixel in isolated RED sector image, and is added the absolute value of each pixel difference; And macroscopical determining step, if the absolute value sum of each pixel difference is less than predetermined first threshold, then described sample is judged to not occur aggegation, if the absolute value sum of each pixel difference is greater than predetermined Second Threshold, then described sample is judged to occur aggegation; If wherein the absolute value sum of each pixel difference is between first threshold and Second Threshold, then described sample can not be sentenced thus carry out described microcosmic judgement.
According to an embodiment, preferably, added up respectively by the difference absolute value sum of each pixel in the image that several occurred to aggegation and do not occur aggegation and neighborhood territory pixel, and then be averaged to determine described first threshold and described Second Threshold.
According to an embodiment, preferably, in described extraction step, the micro-image of sobel operator and collection is adopted to obtain gradient map as planar convolution, by this gradient map binaryzation is obtained edge image.
According to an embodiment, in described microcosmic acquisition step, by taking several micro-images at diverse location, therefrom choose a width the most clearly micro-image for carrying out edge extracting.Preferably, according to an embodiment, by several micro-images described are converted into gray-scale map, and calculate the variance of each pixel and neighborhood territory pixel point gray-scale value in every width image in two perpendicular direction, then the variance of calculating is added up, image maximum for accumulated value is defined as most picture rich in detail.
According to the second aspect of the embodiment of the present invention, a kind of machine vision interpretation device for blood type analysis is provided, comprise macroscopical judging unit and microcosmic judging unit, the sample being wherein defined as sentencing for macroscopical judging unit is further by the process of described microcosmic judging unit, described microcosmic judging unit comprises: microcosmic acquisition module, for gathering the micro-image of the rear sample of reaction; Extraction module, for extracting the edge image of gathered micro-image; And microcosmic judge module, by the number of edge calculation image pixel at least one direction, spectral discrimination pixel number being less than the minimum value set, as there is aggegation, is greater than the spectral discrimination of the maximal value set as there is not aggegation.
According to an embodiment, described macroscopical judging unit comprises: macroscopical acquisition module, for gathering the grand design of the rear sample of reaction; Separation module, isolates RED sector image from gathered grand design; Computing module, for calculating the difference of each pixel and neighborhood territory pixel in isolated RED sector image, and is added the absolute value of each pixel difference; And macroscopical judge module, if the absolute value sum of each pixel difference is less than first threshold, then described sample is judged to occur aggegation, if the absolute value sum of each pixel difference is greater than Second Threshold, then described sample is judged to not occur aggegation; If wherein the absolute value sum of each pixel difference is between first threshold and Second Threshold, then described sample can not be sentenced thus is further processed by described microcosmic judging unit.
According to the third aspect of the embodiment of the present invention, provide a kind of blood type analytical instrument, comprise the machine vision interpretation device according to embodiment of the present invention second aspect.
Be specifically described the present invention by specific embodiment below in conjunction with accompanying drawing, wherein identical or substantially identical parts adopt identical Reference numeral instruction.
Accompanying drawing explanation
Fig. 1 is the schematic diagram that the sample of aggegation does not appear in blood sample cell;
Fig. 2 is the schematic diagram that the sample of aggegation appears in blood sample cell;
Fig. 3 is the process flow diagram of the machine vision interpretation method for blood type analysis according to one embodiment of the invention;
Fig. 4 is the process flow diagram of the macroscopical deterministic process according to one embodiment of the invention;
Fig. 5 is the process flow diagram of the microcosmic deterministic process according to one embodiment of the invention;
Fig. 6 is the schematic diagram of the calculating pixel variance according to one embodiment of the invention;
Fig. 7 shows the schematic diagram be connected of macroscopical deterministic process and microcosmic deterministic process according to one embodiment of the invention and these two processes;
Fig. 8 is the schematic diagram of the machine vision interpretation device for blood type analysis according to an embodiment;
Fig. 9 is the schematic diagram of the machine vision interpretation device for blood type analysis according to another embodiment;
Figure 10 is the partial schematic diagram of the blood type analytical instrument of the machine vision interpretation device combined according to an embodiment.
Embodiment
According to the machine vision interpretation method of the embodiment of the present invention based on the following fact: in blood type analysis, blood preparation mixes with reagent and after fully reacting, there will be aggegation and not aggegation two kinds of situations.When aggegation, blood in most cases all can be aggregated into bulk on the surface of mixed liquor, and take on a red color bulk, carries out grand design identification just can obtain result to this situation.But for the weak reaction of some blood preparations and reagent, would not occur large block agglutinator, but form relatively more uniform very little discrete particles, at this moment just need to carry out Micrograph image processing, whether the cell membrane observing it is destroyed.Time inagglutinable, blood cell is evenly distributed in mixed liquor, macroscopically shows as uniform pale red.And in inagglutinable blood sample, each cell is more clear, distribution is relatively even, and cell can not be overlapped, and edge is more unicellular edge, as shown in Figure 1; And in the blood sample of aggegation, cell can be piled up, do not exist unicellular, edge can more much larger than normal cell times, as shown in Figure 2.This is the theoretical foundation of machine vision interpretation method.
As shown in Figure 3, it is the process flow diagram of the machine vision interpretation method for blood type analysis according to one embodiment of the invention, the method comprises macroscopical deterministic process and microcosmic deterministic process, and the sample being wherein defined as sentencing for macroscopical deterministic process carries out microcosmic judgement further.Described macroscopical deterministic process comprises: macroscopical acquisition step 300, separating step 302, calculation procedure 304, and macroscopical determining step 306.Described microcosmic deterministic process comprises: microcosmic acquisition step 308, extraction step 310, and microcosmic determining step 312.
Wherein macroscopical acquisition step 300 is for gathering the grand design of the rear sample of reaction; In separating step 302, from gathered grand design, isolate RED sector image; The absolute value of each pixel difference for calculating the difference of each pixel and neighborhood territory pixel in isolated RED sector image, and is added by calculation procedure 304; In macroscopical determining step 306, if the absolute value sum of each pixel difference is less than predetermined first threshold, then described sample is judged to not occur aggegation, if the absolute value sum of each pixel difference is greater than predetermined Second Threshold, then described sample is judged to occur aggegation; If wherein the absolute value sum of each pixel difference is between first threshold and Second Threshold, then described sample can not be sentenced thus carry out microcosmic judgement.
Wherein microcosmic acquisition step 308 is for gathering the micro-image of the rear sample of reaction; In extraction step 310, extract the edge image of the micro-image gathered; And in microcosmic determining step 312, by the number of edge calculation image pixel at least one direction, sample pixel number being less than the minimum value of setting is judged to occur aggegation, and the sample being greater than the maximal value of setting is judged to not occur aggegation.Below for positive definite method, above-mentioned steps is specifically described.
Mixed liquor for blood sample respectively with positive definite after anti-A and anti-B reagent reacting, utilize grand design collector (such as common digital camera or Digital Video) to take two pictures, judged the aggegation situation of reacting by the difference value between the color distribution of analysis chart picture and image pixel.Macroscopic view deterministic process as illustrated in the flow diagram of fig. 4.
First, send instruction to sample adding vessel wheelwork by controller (such as CPU, DSP, PLC, MCU, PC or main frame etc.), require that it rotates sample adding vessel device and the reaction utensil that the blood sample reacted is housed is moved on the just right position of grand design collector.Then, start grand design collector and carry out image taking (step 300) to this reaction utensil, then the image photographed is sent to controller by USB interface, controller will carry out data processing to the image photographed.Processing procedure is: the RED sector (step 302) being isolated roughly image by the red component R value of image; Then difference is done to each pixel of the RED sector separated and neighbours' threshold pixel of surrounding, take absolute value addition, and the difference value of each pixel of image is added up (pixel at the most edge of image does not calculate), obtain a total difference value (step 304), this numerical value is the parameter evaluating blood sample aggegation situation.
The redness of this numerical response image is evenly distributed situation, and this numerical value that is evenly distributed is just little, and skewness this numerical value is just large.By calculating and statistical study a large amount of experimental image, two threshold values can be pre-determined; What be less than first threshold is all not aggegation, and what be greater than Second Threshold is all aggegation, between then sentencing for being not suitable for judging i.e. macroscopic view process therebetween, needs to carry out microcosmic process.When getting image size and being 240 × 180 sizes, instead determining in the test of method blood group, by calculating and statistical study a large amount of experimental image, determine after being averaged that first threshold is about 60000, Second Threshold is about 140000.Can not aggegation be judged to when total difference value is less than 60000, when being greater than 140000, can aggegation be judged to.That is, if total difference value is less than 60000 or be greater than 140000 and be and can sentence, can directly judge, and export judged result (step 306), if total difference value is between 60000 to 140000, be weak aggegation, be and can not sentence, just enter Micrograph image processing process.Similarly, to positive definite and the test of RhD blood group, add up by experiment, can determine that first threshold is about 100000, Second Threshold is about 160000.
When macroscopic view can not judge or result is suspicious time, microcosmic judgement can be carried out.Under the control of the controller, rotate sample adding vessel device, make the reaction utensil that the blood sample that above-mentioned grand design can not be sentenced is housed move to the position of microscopic photography.When gathering micro image, automatic focus micro image collection device (such as the microscope of enlargement factor about 100 times) can be utilized to take.Such as, but in one embodiment, to a photography of sample multiple image, ten width images, therefrom select a width the most clearly and process.Microcosmic deterministic process as illustrated in the flow chart of figure 5.
Specifically, controller can be configured to send instruction to micro image collection device mobile controller by USB interface, regulate the distance between the camera lens of micro image collection device and reaction utensil.At ten position shooting micro-images, such as equally spaced ten positions between distance reaction utensil 11-12mm.The mobile micro image collection device camera lens of micro-mobile controller energy, to the distance of setting, carries out image taking, often takes piece image and is just sent to controller by USB interface.Repeat said process, obtain one group of image.Then, the ten width original images collected are converted into gray-scale map, carry out sharpness evaluation to find the image of the most accurate Jiao of a width.Sharpness evaluation function can adopt variance function, because cell is similar to circle, selecting the pixel of two vertical direction, including but not limited to horizontal and vertical direction when calculating variance.Concrete evaluation procedure is as follows.
As shown in Figure 6, to each pixel Ai in image, its variance Mi is calculated as follows:
Mi＝(Ai-P2)^2+(Ai-P4)^2；
Wherein Ai is the gray-scale value of pixel, and P2 is the gray-scale value of first pixel above Ai, and P4 is the gray-scale value of first pixel on the right side of Ai.In order to reduce calculated amount, only choosing the P2 point in vertical direction and the P4 point in horizontal direction, in like manner can get P6 and P8.Then, the Mi of each pixel is added up, cumulative sum is exactly sharpness evaluation of estimate, and this value larger expression image is more clear, thus is used for carrying out edge extracting (step 308) by corresponding image.
In one embodiment, sobel operator is adopted to carry out edge extracting (step 310) to the micro-image the most clearly chosen.In another embodiment, before carrying out edge extracting, also comprise alternatively and enhancing process is carried out to the image chosen.Edge extracting adopts template way to carry out, such as, can adopt following sobel operator template:
| -1 | 0 | 1 |
| -2 | 0 | 0 |
| -1 | 0 | 1 |
| -1 | -2 | -1 |
| 0 | 0 | 0 |
| 1 | 2 | 1 |
This operator comprises the matrix of two group 3 × 3, is respectively laterally and longitudinally, itself and image is made planar convolution, can draw horizontal and longitudinal brightness difference approximate value respectively.If represent original image with A, Gx and Gy represents the image detected through transverse direction and longitudinal edge respectively, then can obtain:
Thus each pixel transverse direction of image and longitudinal gradient approximation can calculate with following formula:
So, obtain a width gradient map G.Choose a threshold T H, such as, can choose 128, judge as follows.If G (i, j) > is TH, then (i, j) is step-like marginal point, and putting this value is 255, otherwise is set to 0.Like this, just gradient map is converted to an outline map S (i, j) by binaryzation.
The edge image extracted is bianry image (black and white), and edge is stain or the white point of single pixel.Be expert at respectively and/or arrange, statistics edge pixel point number, and asking its mean value.If the field size of micro image collection device is approximately 10 cell sizes (can be set as N number of cell for generalized case), if the most low aggegation cell accumulation block that can identify is set as that the yardstick of 3 times of cells (can be set as M cell for generalized case, and M < N), then maximum in microscopic field of view aggegation blocks is 3 × 3 (is ([N/M]) for generalized case
2individual, wherein [] expression rounds).An annulus is having 2 marginal points through on certain round row or certain row), the maximal value so set is 10 (maximal value for generalized case setting is N), and the minimum value of setting is 6 (minimum value for generalized case setting is 2 × [N/M]).So, following judgement can be done, if be all distributed with edge pixel and its mean number is more than or equal to 10 on each row and each row, cell not aggegation (cell membrane is complete) can be thought; On the contrary, marginal distribution some row and column, mean number can think cell agglutination (cell membrane be destroyed) less than 6; Other situations then do not make a decision, and directly export micro-image and carry out artificial judgment.Finally, judge to bleeding type according to the aggegation of two pictures and inagglutinable combination, such as, if be aggegation with anti-A reagent reacting, anti-B reagent reacting is not aggegation, then this blood group is A type (step 312).
The process flow diagram of Fig. 7 compares and fully illustrates being connected of above-mentioned macroscopical deterministic process and microcosmic deterministic process and these two processes.
Fig. 8 is the schematic diagram of the machine vision interpretation device 900 for blood type analysis according to an embodiment, this device comprises macroscopical judging unit and microcosmic judging unit, and the sample being wherein defined as sentencing for macroscopical judging unit is processed by described microcosmic judging unit further.Described macroscopical judging unit comprises: macroscopical acquisition module 902 (including but not limited to the grand design collector of such as ordinary digital camera or Digital Video), separation module 904, computing module 906, and macroscopical judge module 908; Described microcosmic judging unit comprises: microcosmic acquisition module 910 (including but not limited to the microscopical micro image collection device of such as enlargement factor about 100 times), extraction module 912, and microcosmic judge module 914.Wherein:
-macroscopical acquisition module 902 is for performing step 300;
-separation module 904 is for performing step 302;
-computing module 906 is for performing step 304;
-macroscopical judge module 908 is for performing step 306;
-microcosmic acquisition module 910 is for performing step 308;
-extraction module 912 is for performing step 310; And
-microcosmic judge module 914 is for performing step 312.
Fig. 9 is the another kind of embodiment for the machine vision interpretation device 900 of blood type analysis, and this device 900 comprises processing unit 913, such as DSP or CPU etc.Processing unit 913 can be individual unit or multiple unit, to perform described different step.In addition, this device 900 also comprises interactive interface 980 and output unit 990 alternatively.In addition, this device 900 also comprises at least one computer program 910 of nonvolatile memory form, such as EEPROM, flash memory or hard disk drive etc.This computer program 910 comprises computer program 911, and computer program 911 comprises program code, when it is run, this device 900 is performed about the step shown in Fig. 3.
Specifically, the program code in the computer program 911 of device 900 comprises: macroscopical acquisition module 911a, for performing step 300; Separation module 911b, for performing step 302; Computing module 911c, for performing step 304; Macroscopic view judge module 911d, for performing step 306; Microcosmic acquisition module 911e, for performing step 308; Extraction module 911f, for performing step 310; And microcosmic judge module 911g, for performing step 312.In other words, when running different module 911a-911g and/or these modules run under control of the processing unit on processing unit 913, they correspond to the module 902,904,906,908,910,912 and 914 shown in Fig. 8.
According to the machine vision interpretation device 900 for blood type analysis of above-described embodiment, software, hardware, firmware or its combination can be passed through, realize in blood type analytical instrument.As shown in Figure 10, it is the partial schematic diagram of the blood type analytical instrument of the machine vision interpretation device 900 combined according to an embodiment, wherein grand design collector 110 and the micro image collection device 114 comprising mobile controller 112 are disposed in parallel in the top of reaction utensil 116, and relative to the moving direction of reaction utensil 116, grand design collector 110 is positioned at upstream and micro image collection device 114 is positioned at downstream.In addition, micro image collection device mobile controller 112 includes but not limited to by the gear mechanism of driven by motor.This realization is that appearance is facile to those skilled in the art, does not describe in detail at this.
Described the present invention by specific embodiment above, but the present invention is not limited to these specific embodiments.Those skilled in the art should be understood that, various amendment, equivalent replacement, change etc. can also be made to the present invention, such as two or more steps or module is divided into realize the step of in above-described embodiment or module, or contrary, the function of two or more steps in above-described embodiment or module is placed in a step or module and realizes.But, as long as these conversion do not deviate from spirit of the present invention, all should within protection scope of the present invention.In addition, some terms that present specification and claims use, such as " level ", " vertically " etc. are not restriction, are only used to be convenient to describe.In addition, " embodiment ", " another embodiment " etc. described in above many places represent different embodiments, can certainly by its all or part of realization in one embodiment.
Claims (10)
1., for a machine vision interpretation method for blood type analysis, it is characterized in that comprising macroscopical deterministic process and microcosmic deterministic process, the sample being wherein defined as sentencing for macroscopical deterministic process carries out microcosmic judgement further, and described microcosmic deterministic process comprises:
Microcosmic acquisition step, for gathering the micro-image of the rear sample of reaction;
Extraction step, for extracting the edge image of gathered micro-image; And
Microcosmic determining step, by the number of edge calculation image pixel at least one direction, sample pixel number being less than the minimum value of setting is judged to occur aggegation, and the sample being greater than the maximal value of setting is judged to not occur aggegation.
2. machine vision interpretation method as claimed in claim 1, it is characterized in that, described macroscopical deterministic process comprises:
Macroscopic view acquisition step, for gathering the grand design of the rear sample of reaction;
Separating step, isolates RED sector image from gathered grand design;
Calculation procedure, for calculating the difference of each pixel and neighborhood territory pixel in isolated RED sector image, and is added the absolute value of each pixel difference; And
Macroscopic view determining step, if the absolute value sum of each pixel difference is less than predetermined first threshold, then described sample is judged to not occur aggegation, if the absolute value sum of each pixel difference is greater than predetermined Second Threshold, then described sample is judged to occur aggegation;
If wherein the absolute value sum of each pixel difference is between first threshold and Second Threshold, then described sample can not be sentenced thus carry out described microcosmic judgement.
3. machine vision interpretation method as claimed in claim 2, it is characterized in that: added up respectively by the difference absolute value sum of each pixel in the image that several occurred to aggegation and do not occur aggegation and neighborhood territory pixel, and then be averaged to determine described first threshold and described Second Threshold.
4. the machine vision interpretation method as described in any one of claims 1 to 3, it is characterized in that: in described extraction step, the micro-image of sobel operator and collection is adopted to obtain gradient map as planar convolution, by this gradient map binaryzation is obtained edge image.
5. the machine vision interpretation method as described in any one of claims 1 to 3, is characterized in that: in described microcosmic acquisition step, by taking several micro-images at diverse location, therefrom choose a width the most clearly micro-image for carrying out edge extracting.
6. machine vision interpretation method as claimed in claim 5, it is characterized in that: by several micro-images described are converted into gray-scale map, and calculate the variance of each pixel and neighborhood territory pixel point gray-scale value in every width image in two perpendicular direction, then the variance of calculating is added up, image maximum for accumulated value is defined as most picture rich in detail.
7. the machine vision interpretation method as described in any one of claims 1 to 3, is characterized in that: the minimum value of described setting and the maximal value of described setting depend on cell dimensions in sample, setting aggegation block yardstick and gather micro-image time visual field yardstick.
8. the machine vision interpretation device for blood type analysis, it is characterized in that comprising macroscopical judging unit and microcosmic judging unit, the sample being wherein defined as sentencing for macroscopical judging unit is processed by microcosmic judging unit further, and described microcosmic judging unit comprises:
Microcosmic acquisition module, for gathering the micro-image of the rear sample of reaction;
Extraction module, for extracting the edge image of gathered micro-image; And
Microcosmic judge module, by the number of edge calculation image pixel at least one direction, spectral discrimination pixel number being less than the minimum value set, as there is aggegation, is greater than the spectral discrimination of the maximal value set as there is not aggegation.
9. machine vision interpretation device as claimed in claim 8, it is characterized in that, described macroscopical judging unit comprises:
Macroscopic view acquisition module, for gathering the grand design of the rear sample of reaction;
Separation module, isolates RED sector image from gathered grand design;
Computing module, for calculating the difference of each pixel and neighborhood territory pixel in isolated RED sector image, and is added the absolute value of each pixel difference; And
Macroscopic view judge module, if the absolute value sum of each pixel difference is less than predetermined first threshold, then described sample is judged to not occur aggegation, if the absolute value sum of each pixel difference is greater than predetermined Second Threshold, then described sample is judged to occur aggegation;
If wherein the absolute value sum of each pixel difference is between first threshold and Second Threshold, then described sample can not be sentenced thus is further processed by described microcosmic judging unit.
10. a blood type analytical instrument, is characterized in that: comprise the machine vision interpretation device described in claim 8 or 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010508384.XA CN102446353B (en) | 2010-09-30 | 2010-09-30 | For machine vision interpretation method and the device of blood type analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010508384.XA CN102446353B (en) | 2010-09-30 | 2010-09-30 | For machine vision interpretation method and the device of blood type analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102446353A CN102446353A (en) | 2012-05-09 |
| CN102446353B true CN102446353B (en) | 2016-01-06 |
Family
ID=46008831
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010508384.XA Active CN102446353B (en) | 2010-09-30 | 2010-09-30 | For machine vision interpretation method and the device of blood type analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102446353B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103543277B (en) * | 2013-09-13 | 2016-03-16 | 中国科学院苏州生物医学工程技术研究所 | A kind of blood group result recognizer based on gray analysis and category identification |
| CN105550507A (en) * | 2015-12-09 | 2016-05-04 | 上海斐讯数据通信技术有限公司 | Intelligent terminal for measuring blood type and measuring method |
| CN105678752A (en) * | 2015-12-30 | 2016-06-15 | 合肥天一生物技术研究所 | Blood type identification method through segmentation of microcolumn vasculum based on fixed parameters |
| CN106127738B (en) * | 2016-06-16 | 2018-12-11 | 上海荣盛生物药业有限公司 | agglutination test interpretation method |
| CN106199014A (en) * | 2016-06-29 | 2016-12-07 | 中山生物工程有限公司 | A kind of blood group method of discrimination based on blood type card |
| CN106980860A (en) * | 2017-04-07 | 2017-07-25 | 广州誉康医药有限公司 | A kind of result automatic interpretation method for liquid medium cross matching |
| CN107992851B (en) * | 2017-12-20 | 2020-05-01 | 闫鸿远 | Identification method of agglutination test |
| CN109324197A (en) * | 2018-11-16 | 2019-02-12 | 北京诚名扬科技发展有限公司 | Bracket for blood grouping recorder and identification recording method |
| CN110286239A (en) * | 2019-08-12 | 2019-09-27 | 杭州美川合佳生物科技有限公司 | A kind of blood group automatical analysis device |
| CN115808421B (en) * | 2023-02-09 | 2023-04-14 | 广州誉康医药有限公司 | Agglutination state interpretation method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101410049A (en) * | 2006-01-31 | 2009-04-15 | 芝加哥大学 | Method and apparatus for assaying blood clotting |
| CN101765773A (en) * | 2007-06-29 | 2010-06-30 | 贝克曼考尔特公司 | Agglutination image automatic judging method by MT system, device, program, and recording medium |
| CN101842688A (en) * | 2007-10-29 | 2010-09-22 | 希森美康株式会社 | Cell analysis apparatus and cell analysis method |
-
2010
- 2010-09-30 CN CN201010508384.XA patent/CN102446353B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101410049A (en) * | 2006-01-31 | 2009-04-15 | 芝加哥大学 | Method and apparatus for assaying blood clotting |
| CN101765773A (en) * | 2007-06-29 | 2010-06-30 | 贝克曼考尔特公司 | Agglutination image automatic judging method by MT system, device, program, and recording medium |
| CN101842688A (en) * | 2007-10-29 | 2010-09-22 | 希森美康株式会社 | Cell analysis apparatus and cell analysis method |
Non-Patent Citations (1)
| Title |
|---|
| 数字图象清晰度评价函数研究;李奇 等;《光子学报》;20020630;第31卷(第6期);第1.1节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102446353A (en) | 2012-05-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102446353B (en) | For machine vision interpretation method and the device of blood type analysis | |
| CN109460753B (en) | Method for detecting floating object on water | |
| CN110414334B (en) | Intelligent water quality identification method based on unmanned aerial vehicle inspection | |
| CN102630299B (en) | Image processing device for defect inspection and image processing method for defect inspection | |
| CN103020585A (en) | Method for identifying positive cells and negative cells of immunologic tissue | |
| CN102404602A (en) | Camera definition detection method based on definition test card | |
| CN103344583A (en) | Praseodymium-neodymium (Pr/Nd) component content detection system and method based on machine vision | |
| CN113658174B (en) | Microkernel histology image detection method based on deep learning and image processing algorithm | |
| CN116843999B (en) | Gas cylinder detection method in fire operation based on deep learning | |
| CN106327508B (en) | Ki67 index automatic analysis method | |
| CN111289512A (en) | Rice grain alkali elimination value high-throughput determination method based on deep convolutional neural network | |
| CN108710893A (en) | A kind of digital image cameras source model sorting technique of feature based fusion | |
| CN112580536A (en) | High-order video vehicle and license plate detection method and device | |
| CN104867128A (en) | Image blurring detection method and device | |
| US8185317B2 (en) | Method and system of determining the stain quality of slides using scatter plot distribution | |
| CN112070750A (en) | Leather product defect detection method and device | |
| CN102297833A (en) | Method for determination of content of various cells in human blood or bone marrow | |
| CN115359264A (en) | Intensive distribution adhesion cell deep learning identification method | |
| US8321136B2 (en) | Method and system for classifying slides using scatter plot distribution | |
| CN113554610A (en) | Photovoltaic module operation state detection method and application device thereof | |
| CN111583186A (en) | Pathological ER/PR (ER/PR) cell nucleus counting method and system for clinical application | |
| CN115496716A (en) | Single and double micronucleus cell image detection method based on deep learning and related equipment | |
| CN115240000B (en) | Diabetes detection device and method for forensic identification | |
| CN109919924A (en) | A method of suitable for high-volume HE dyeing picture cell number wordization processing | |
| Saputra et al. | Application Object Detection Using Histogram of Oriented Gradient For Artificial Intelegence System Module of Nao Robot (Control System Laboratory (LSKK) Bandung Institute of Technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address |
Address after: 519031 office 2320, No. 3000, Huandao East Road, Hengqin new area, Zhuhai, Guangdong Patentee after: Yangpu Medical Technology Co.,Ltd. Address before: 510530 No.102, Kaiyuan Avenue, Science City, Guangzhou Economic and Technological Development Zone, Guangdong Province Patentee before: GUANGZHOU IMPROVE MEDICAL TECHNOLOGY Co.,Ltd. |
|
| CP03 | Change of name, title or address |