CN100483432C - Genital-infection inteeligent identification system and method - Google Patents
Genital-infection inteeligent identification system and method Download PDFInfo
- Publication number
- CN100483432C CN100483432C CNB031402585A CN03140258A CN100483432C CN 100483432 C CN100483432 C CN 100483432C CN B031402585 A CNB031402585 A CN B031402585A CN 03140258 A CN03140258 A CN 03140258A CN 100483432 C CN100483432 C CN 100483432C
- Authority
- CN
- China
- Prior art keywords
- green value
- carry out
- carried out
- reproduction
- filtering
- 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.)
- Expired - Lifetime
Links
Images
Landscapes
- Investigating Or Analysing Biological Materials (AREA)
- Apparatus Associated With Microorganisms And Enzymes (AREA)
Abstract
The present invention relates to a genital infection intelligent identification system and its method. It is used for making intelligent identification and judgement of biological fluorescent microscopic picture inputted into the computer and making auxiliary diagnosis of infection condition of microorganism and virus which can result in genital infection of patient. Said system includes biological fluorescence microscope, induotrial video cmera head, picture collecting card and computer. Said invention provides the application method and concrete steps of said system. Said invented system and method can automatically and accurately identify various microorganisms and viruses of chlamydia, mycoplasma, herpes simplex virus, candida albicans, gonococcus and others which can result in genital infection.
Description
Technical field
The present invention relates to general view data handles, particularly relate to the enhancing or the recovery of image, utilize the filtering of non-space territory and use histogram technology, automatically discern original figure from the digit order number image pattern, especially relate to intelligent identifying system and method the bioluminescence microscope picture that causes microorganism that reproduction is infected and virus.
Background technology
Prior art relates to identification and the judgement that reproduction is infected, usually use fluorescence microscopy inspection technology, promptly utilize the light of certain wavelength that sample is excited, produce the fluorescence of different colours, observe and differentiate some material and character thereof in the sample again by optical microscope, this technology is extensively used at biology and medical domain.Cause pathogenic microorganisms and virus that the human body reproduction is infected, comprise bacterium, pathogen and virus, some intrinsic material in these microorganisms and virus, as protein, DNA (DNA (deoxyribonucleic acid)) etc., perhaps these microorganisms and virus have the outside fluorescent material of introducing, as fluorescent marker, photosensitizer etc., when being subjected to the exciting of outside short-wavelength light, can send the fluorescence that omits another long wavelength than excitation wavelength, the microorganism and the virus of fluorescent material have been absorbed with the biological microscopic examination, just can observe the distribution situation of fluorescent material in these microorganisms and virus, thereby discern and judge the existence and the distribution situation of these microorganisms and virus.Prior art is assisted the microexamination technology by computing machine usually, just microscopical observations is passed through industrial video camera and image pick-up card, perhaps device transmission such as digital camera is to the calculator memory storage, and represent to the operator by display, yet the identification of these microorganisms and virus and judgement must rely on operator's experience: the operator must have observation and the differentiation experience that surpasses 10,000 such bioluminescence microscope pictures, could guarantee the correctness of assay, this just requires that the operator is carried out long-term training and operator itself and has for many years practical work experience.
Summary of the invention
The technical problem to be solved in the present invention is to avoid above-mentioned the deficiencies in the prior art part, and proposes a kind of to causing that microorganism and virus that reproduction is infected need not rely on intelligent identifying system operator's experience, robotization and method.
The technical scheme that the present invention solves the problems of the technologies described above employing is, manufacture and design a kind of reproduction and infect intelligent identifying system, in order to the bioluminescence microscope picture that is input in the computing machine is carried out Intelligent Recognition and judgement, cause microorganism and the viral infection state that prescription on individual diagnosis person's reproduction is infected with auxiliary diagnosis, comprise bioluminescence microscope, industrial video camera or digital camera, image pick-up card and computing machine, especially, also comprise fluorescence light source lamp, stabilized voltage supply, shooting extend neck and computer numeral camera; Described fluorescence light source lamp provides direct supply by stabilized voltage supply; Described shooting extend neck is used for connecting bioluminescence microscope and industrial video camera or digital camera; By described image pick-up card, described industrial video camera is connected with computing machine, perhaps directly is connected with computing machine with described digital camera; Described computer numeral camera is used to take patient's diseased region picture and patient's head portrait; Described computing machine infects each device realization systemic-function in the Intelligent Recognition software control system by reproduction, comprising: the bitmap that obtains picture; Obtain bitmap matrix; Obtain the average of bitmap matrix; Carry out medium filtering; Carry out Laplce's filtering; Carry out image segmentation; Carrying out the image zonule cuts apart; Ferret out; Carry out goal filtering; Draw target.
The technical scheme that the present invention solves the problems of the technologies described above employing also has, propose a kind of reproduction and infect intelligent identification Method, in order to the bioluminescence microscope picture that is input in the computing machine is carried out Intelligent Recognition and judgement, cause microorganism and the viral infection state that prescription on individual diagnosis person's reproduction is infected with auxiliary diagnosis, comprise step: the bitmap that reads picture; Obtain bitmap matrix; Obtain the average of bitmap matrix; Carry out medium filtering; Carry out Laplce's filtering; Carry out image segmentation; Carrying out the image zonule cuts apart; Ferret out; Carry out goal filtering; Draw number of targets.
Compare with prior art, adopt reproduction of the present invention to infect intelligent identifying system and method, need not rely on the operator discerns and judges, just can discern various microorganism and the viruses such as Chlamydia, mycoplasma, herpes simplex virus, Candida albicans, gonococcus and Gartner bacterium that cause that reproduction is infected automatically, exactly, and judge that assay is negative or positive.
Description of drawings
Fig. 1 infects intelligent identification Method implementation step synoptic diagram for reproduction of the present invention.
Fig. 2 infects intelligent identifying system for reproduction of the present invention and constitutes synoptic diagram.
Fig. 3 is the image Segmentation flow chart of steps of described embodiment one.
Fig. 4 is first, second time ferret out flow chart of steps of described embodiment one.
Fig. 5 is the flow chart of steps of ferret out for the third time of described embodiment one.
Fig. 6 is labeled as target substep process flow diagram in the image Segmentation step of described embodiment one.
Fig. 7 is the precise search target step process flow diagram of described embodiment two.
Fig. 8 is the image Segmentation flow chart of steps of described embodiment three.
Fig. 9 is the image Segmentation flow chart of steps of described embodiment four.
Figure 10 is the acquisition extreme point flow chart of steps of described embodiment five.
Figure 11 calculates weights substep process flow diagram in the acquisition extreme point step of described embodiment five.
Figure 12 is described system software main program flow chart.
Figure 13 A, 13B are described system software image capture module process flow diagram.
Figure 14 A, 14B are described system software figure database management module process flow diagram.
Figure 15 A, 15B are described system software intelligent identification module process flow diagram.
Figure 16 is described system software electronic health record module process flow diagram.
Figure 17 is the detailed module process flow diagram of described system software case history.
Figure 18 is described system software survey report single module process flow diagram.
Figure 19 is described system software health account card module process flow diagram.
Figure 20 A, 20B are described system software system maintaining module process flow diagram.
Figure 21 A, 21B are the Intelligent Recognition surface chart of described system software.
Figure 22 A, 22B are that the patient's sample picture amplifies surface chart in the Intelligent Recognition of described system software.
Figure 23 A, 23B are the electronic health record information interface figure of described system software.
Figure 24 is the inspection report surface chart of described system software.
Figure 25 is the health account card interface figure of described system software.
Embodiment
Be described in further detail below in conjunction with each most preferred embodiment shown in the accompanying drawing.
Intelligent identifying system is infected in reproduction of the present invention, in order to the bioluminescence microscope picture that is input in the computing machine is carried out Intelligent Recognition and judgement, cause microorganism and the viral infection state that prescription on individual diagnosis person's reproduction is infected with auxiliary diagnosis, as shown in Figure 2, comprising: bioluminescence microscope 210, fluorescence light source lamp 290, stabilized voltage supply 291, shooting extend neck 211, industrial video camera 220, keyboard and mouse 230, computing machine 240, printer 250, display 260, image pick-up card 270 and computer numeral camera 280.When adopting the alternative industrial video camera 220 wherein of digital camera, can omit image pick-up card 270.
Wherein, bioluminescence microscope 210 is used for observing patient specimen and for system provides light signal, object lens are good with fluorescence, and the enlargement factor maximal value should reach 100 times, the power supply of fluorescence light source lamp adopts stabilized voltage supply 291 that direct supply is provided, in order to avoid camera and signal wire are caused interference.Shooting extend neck 211 is used for connecting microscope 210 and industrial video camera or digital camera 220, helps imaging, generally has about 1 times and 0.5 times two kinds, is provided by the manufacturer of biological microscope.Camera 220 is converted to electric signal to the light signal from biological microscope, and image pick-up card 270 is the analog signal conversion from industrial video camera 220 digital signal.Computer numeral camera 280 is used to take patient's diseased region picture and patient's head portrait.
The course of work that intelligent identifying system is infected in reproduction of the present invention comprises: the patient specimen fluorescent reagent is dyeed; With the patient specimen after the 210 observation dyeing of biological microscope; Take displaing micro picture by filming apparatus, convert light signal to electric signal and be transferred to image pick-up card 270; Is to send after the digital signal computing machine 240 to handle by image pick-up card 270 with analog signal conversion; Computing machine 240 is converted to digital picture with digital signal on the one hand and displays by display 260, by Intelligent Recognition software the digital picture that obtains handled and fuzzy diagnosis on the other hand, and the output recognition result; By system software Intelligent Recognition result and reference picture are outputed to inspection report automatically again, electronic health record and women of child-bearing age's healthy reproduction registration form; At last, system software is made, preserves, is also printed inspection report, patient's electronic health record and registration forms by printer 250.Patient's diseased region and patient's head portrait can also be taken by system of the present invention, and the input patient and/or the women of child-bearing age's various information.
The software of system of the present invention as shown in figure 12, includes: image acquisition, figure library management, Intelligent Recognition, electronic health record, case history in detail, functional modules such as inspection report, health account card and system maintenance; In addition, also comprise the medical knowledge functional module.
Image capture module is used for the optical imagery of being examined patient specimen on the biological microscope is gathered, and import and be stored in the computing machine, the inner structure of this module, as shown in FIG. 13A, comprising: the user selects grab type, judges flow processs such as administrator right.
The figure database management module, be used for the picture of calculator memory storage is managed, realize checking survey report, add picture, select picture, browsing pictures, deletion picture and check function such as case history, all will cause operation picture, the inner structure of this module is shown in Figure 14 A and 14B.
Intelligent identification module, be used for the picture of calculator memory storage is handled, analyze, discern and judge, and finally produce assay, the inner structure of this module, shown in Figure 15 A, comprising: select picture, begin sub-function module such as identification, result screen, criterion of identification, identification interface, and begin after the identification, and for example shown in Figure 15 B, comprise and call recognition function, selected picture is discerned that the picture that will return after recognition result quantity also will be discerned is preserved.And performed step in the recognition function as shown in Figure 1, comprising: the bitmap 110 that obtains picture; Obtain bitmap matrix 120; Obtain the average 130 of bitmap matrix; Carry out medium filtering 140; Carry out Laplce's filtering 150; Carry out image segmentation 160; Carry out the image zonule and cut apart 170; Ferret out 180; Carry out goal filtering 190; Draw number of targets 191.
The blank list at Intelligent Recognition interface is shown in Figure 21 A; And the interface of demonstration recognition result, shown in Figure 21 B, the top of display page has: reference culture figure, typical positive map, the negative figure of typical case and patient specimen figure, and the target number that demonstrates kind of inspection and recognize in the testing result particulars window of right lower quadrant.Figure 22 A is the patient specimen figure before the identification of having amplified, and whole picture is very fuzzy, and Figure 22 B is the patient specimen figure after discerning, and the part of the shinny encirclement in border is exactly identified target among the figure.
The electronic health record module is in order to show the case history essential information, shown in Figure 23 A, and these information are managed, the inner structure of this module, as shown in figure 16, comprise: consult " medical knowledge " to obtain help, add sub-function module such as case history and retrieval case history.Consult " medical knowledge " and can cause, and the inner structure of this module shown in Figure 13 B, comprising: the medical knowledge flow process the calling of medical knowledge functional module.Figure 23 B is depicted as the situation that system provides relevant medical knowledge.
The detailed module of case history, in order to the record case that provides patient repeatedly to go to a doctor, the inner structure of this module as shown in figure 17, comprising: each sub-function module such as particulars management of going to a doctor.
The survey report single module, in order to fill in, the single content of demonstration, printed report and report managed, the inner structure of this module as shown in figure 18, comprising: to the printing of report, cancel, preserve, add, sub-function module such as retrieval.The display interface of inspection report, as shown in figure 24.
The health account card module manages in order to health status, information to the surveyee, and the inner structure of this module as shown in figure 19, comprising: sub-function module such as interpolation, cancellation, retrieval.The display interface of women of child-bearing age's health account card as shown in figure 25.
System maintaining module shown in Figure 20 A, comprising: system's setting, user management, data backup, data recover and
Sub-function module such as data importing derivation, and the structure of sub-function module is derived in data importing, shown in Figure 20 B.
Intelligent identifying system and method are infected in reproduction of the present invention, need not rely on the operator discerns and judges, just can discern automatically, exactly and variously cause that reproduction is infected, as microorganism and viruses such as Chlamydia, mycoplasma, herpes simplex virus, Candida albicans, gonococcus and Gartner bacterium, and judge that assay is negative or positive.
Below further described by several specific embodiments:
Embodiment one:
System and method for of the present invention is applied to the Intelligent Recognition that the Chlamydia reproduction is infected, and the step of the bioluminescence microscope picture of check sample being carried out Intelligent Recognition comprises: read bitmap; Obtain green value, red value, blue value and saturation degree matrix; Obtain green value, red value, blue value and saturation degree average; Green value is carried out the medium filtering first time, and filtering parameter is: 5,5,2,2; Green value is carried out Laplce's filtering, and filtering parameter is: 5*5; Green value is carried out the medium filtering second time, and filtering parameter is: 3,3,1,1; Carry out image segmentation; Carry out goal filtering; Draw preliminary target number; Regain green value; Obtain the threshold value of green value; Green value is carried out medium filtering, and filtering parameter is: 5,5,2,2; Green value is carried out Laplce's filtering, and filtering parameter is: 5*5; Partitioning parameters is set; Carry out image segmentation; Current green value matrix is backed up; Carry out primary ferret out; Call in the green value matrix of backup; Carry out secondary ferret out, carry out ferret out for the third time again; Obtain the target number discerned at last.
Wherein carry out image segmentation step, as shown in Figure 3, comprise again according to the bitmap that reads and the partitioning parameters of acquisition, carry out the binaryzation of image and be labeled as substep such as target, and this is labeled as the target substep, as shown in Figure 6, itself also comprise a series of deterministic processes of carrying out according to the binary image that reads and entrance coordinate.
Carrying out primary ferret out and carrying out secondary ferret out step wherein as shown in Figure 4, comprises again according to the objective matrix of calling in, and carries out substeps such as target processing singly.
And carrying out wherein ferret out step for the third time as shown in Figure 5, then comprises obtaining substeps such as red value, green value average, and finally exporting the recognition objective number according to the green value, red value and the gray matrix that regain.
Adopt reproduction of the present invention to infect intelligent identifying system and method, chlamydia carries out Intelligent Recognition, and accuracy surpasses 90%, can satisfy the needs of practical application, but need not rely on professional's knowledge and experience.
Embodiment two:
System and method for of the present invention is applied to the Intelligent Recognition that the mycoplasma reproduction is infected, and the step of the bioluminescence microscope picture of check sample being carried out Intelligent Recognition comprises: read bitmap; Obtain green value matrix; Obtain green value average; Obtain the threshold value of green value; Green value is carried out medium filtering, and filtering parameter is: 5,5,2,2; Green value is carried out Laplce's filtering, and filtering parameter is: 5*5; Partitioning parameters is set; Carry out image segmentation; Ferret out singly when all targets were all searched for, finishes search then; Carry out precise search again; Carry out goal filtering; Obtain the target number discerned at last.
Ferret out step wherein comprises: obtain the green value average of target; Obtain the threshold value of the green value of target; Boundary rectangle extends out 7 pixels and obtains new matrix; New matrix is carried out substeps such as image segmentation.
Wherein carry out image segmentation step, as shown in Figure 3, comprise again according to the bitmap that reads and the partitioning parameters of acquisition, carry out the binaryzation of image and be labeled as substep such as target, and this is labeled as the target substep, as shown in Figure 6, itself also comprise a series of deterministic processes of carrying out according to the binary image that reads and entrance coordinate.
And wherein carry out precise search, as shown in Figure 7, then comprise obtaining substeps such as red value, green value average, and output recognition objective number according to the green value, red value and the gray matrix that regain.
Adopt reproduction of the present invention to infect intelligent identifying system and method, mycoplasma is carried out Intelligent Recognition, accuracy surpasses 90%, can satisfy the needs of practical application, but need not rely on professional's knowledge and experience.
Embodiment three:
System and method for of the present invention is applied to the Intelligent Recognition that the herpes simplex virus reproduction is infected, and the step of the bioluminescence microscope picture of check sample being carried out Intelligent Recognition comprises: read bitmap; Obtain gray scale, green value and luminance matrix; Obtain gray scale, green value and brightness average; Green value is carried out medium filtering twice, and filtering parameter is: 19,19,9,9; Gray scale is carried out the medium filtering first time, and filtering parameter is: 19,19,9,9; Gray scale is carried out Laplce's filtering, and filtering parameter is: 5*5; Gray scale is carried out the medium filtering second time, and filtering parameter is: 7,7,3,3; Carry out image segmentation; Target is added up; Obtain the target number discerned at last.
Wherein carry out image segmentation step, as shown in Figure 8, comprise again according to the bitmap that reads, obtain the saturation degree threshold value and be labeled as substep such as target, and this is labeled as the target substep, as shown in Figure 6, itself also comprise a series of deterministic processes of carrying out according to the binary image that reads and entrance coordinate.
Adopt reproduction of the present invention to infect intelligent identifying system and method, herpes simplex virus is carried out Intelligent Recognition, accuracy surpasses 90%, can satisfy the needs of practical application, but need not rely on professional's knowledge and experience.
Embodiment four:
System and method for of the present invention is applied to the Intelligent Recognition that the Candida albicans reproduction is infected, and the step of the bioluminescence microscope picture of check sample being carried out Intelligent Recognition comprises: read bitmap; Obtain gray scale, green value and saturation degree matrix; Obtain grey level histogram; Green value is carried out medium filtering twice, and filtering parameter is: 11,11,5,5; Gray scale is carried out the medium filtering first time, and filtering parameter is: 11,11,5,5; Gray scale is carried out Laplce's filtering, and filtering parameter is: 5*5; Gray scale is carried out the medium filtering second time, and filtering parameter is: 3,3,1,1; Obtain green value, brightness and saturation degree histogram; Obtain analytical parameters; Carry out image segmentation; Carrying out target adds up; Obtain the target number discerned at last.
Wherein carry out image segmentation step, as shown in Figure 9, comprise again according to the bitmap that reads, carry out image classification, and handle accordingly according to 0 type, 1 type and 2 type targets respectively, and comprise the green value figure of binaryzation in these processing procedures and be labeled as substeps such as target, and this is labeled as the target substep, as shown in Figure 6, itself also comprise a series of deterministic processes of carrying out according to the binary image that reads and entrance coordinate.
Adopt reproduction of the present invention to infect intelligent identifying system and method, Candida albicans is carried out Intelligent Recognition, accuracy surpasses 90%, can satisfy the needs of practical application, but need not rely on professional's knowledge and experience.
Embodiment five:
System and method for of the present invention is applied to the Intelligent Recognition that the gonococcus reproduction is infected, and the step of the bioluminescence microscope picture of check sample being carried out Intelligent Recognition comprises: read bitmap; Obtain red value, green value, blue value and saturation degree matrix; Obtain red value, green value, blue value and saturation degree average; Green value is carried out the medium filtering first time, and filtering parameter is: 5,5,2,2; Green value is carried out Laplce's filtering, and filtering parameter is: 5*5; Green value is carried out the medium filtering second time, and filtering parameter is: 3,3,1,1; The image segmentation parameter is set; Carry out image segmentation; Regain the green value matrix of full figure; Green value is carried out medium filtering for the third time, and filtering parameter is: 3,3,1,1; The zonule is set once to be cut apart and filtration parameter; Carrying out the zonule once cuts apart; Carrying out target filters for the first time; Regain the green value matrix of full figure; Green value is carried out medium filtering the 4th time, and filtering parameter is: 5,5,2,2; Image partitioning parameters again is set; Carrying out full figure cuts apart again; Delete overlapping target; Carry out successively target second and third, filter for four and the 5th times; Obtain the gray scale extreme point; Judge the number of gray scale extreme point,, then after obtaining green value extreme point, carry out the 6th, seven, eight time of target and filter if be zero; Otherwise, directly carry out the 6th, seven, eight time of target and filter; Increase object count; Obtain the target number discerned at last.
Carrying out image segmentation step and carrying out the segmentation procedure in zonule wherein, as shown in Figure 3, comprise again according to the bitmap that reads and the partitioning parameters of acquisition, carry out the binaryzation of image and be labeled as substep such as target, and this is labeled as the target substep, as shown in Figure 6, itself also comprise a series of deterministic processes of carrying out according to the binary image that reads and entrance coordinate.
Obtaining gray scale extreme point step and obtaining green value extreme point step wherein, as shown in figure 10, comprise again according to the target that reads, obtain each point in the target boundary rectangle, and a series of substeps that carry out corresponding judgement, and include in these substeps: the weights to current point calculate, this computation process, as shown in figure 11, comprise again on every side search and relevant judgement are waited substep.
Adopt reproduction of the present invention to infect intelligent identifying system and method, gonococcus is carried out Intelligent Recognition, accuracy surpasses 90%, can satisfy the needs of practical application, but need not rely on professional's knowledge and experience.
Embodiment six:
System and method for of the present invention is applied to the Intelligent Recognition that Gartner bacterium reproduction is infected, and the step of the bioluminescence microscope picture of check sample being carried out Intelligent Recognition comprises: read bitmap; Obtain red value, green value and saturation degree matrix; Obtain red value, green value and saturation degree average; Green value is carried out medium filtering, and filtering parameter is: 9,9,4,4; Green value is carried out Laplce's filtering, and filtering parameter is: 5*5; The image segmentation parameter is set; Carry out image segmentation; Delete overlapping target; Judging whether to satisfy the zonule and cut apart condition, is then to carry out: regain full figure green value matrix, green value is carried out medium filtering, filtering parameter is: 3,3,1,1; Green value segmentation threshold is set; Carrying out the zonule once cuts apart; Carry out target and filter for the first time, regain the green value matrix of full figure; Otherwise directly carry out: regain the green value matrix of full figure; Green value is carried out medium filtering, and filtering parameter is: 5,5,2,2; The green value matrix of negative-appearing image full figure; The secondary splitting parameter is set; Carry out the zonule secondary splitting; Carry out second and third time filtration of target; Obtain extreme point; Carrying out the 4th time of target filters; Increase object count; Obtain the target number discerned at last.
Carrying out image segmentation step, carry out the segmentation procedure in zonule and carrying out zonule secondary splitting step wherein, as shown in Figure 3, comprise again according to the bitmap that reads and the partitioning parameters of acquisition, carry out the binaryzation of image and be labeled as substep such as target, and this is labeled as the target substep, as shown in Figure 6, itself also comprise a series of deterministic processes of carrying out according to the binary image that reads and entrance coordinate.
Wherein obtain the extreme point step, as shown in figure 10, comprise again according to the target that reads, obtain each point in the target boundary rectangle, and a series of substeps that carry out corresponding judgement, and include in these substeps: the weights to current point calculate, this computation process, as shown in figure 11, comprise again on every side search and relevant judgement are waited substep.
Adopt reproduction of the present invention to infect intelligent identifying system and method, the Gartner bacterium is carried out Intelligent Recognition, accuracy surpasses 90%, can satisfy the needs of practical application, but need not rely on professional's knowledge and experience.
Claims (10)
1, intelligent identifying system is infected in a kind of reproduction, in order to the bioluminescence microscope picture that is input in the computing machine is carried out Intelligent Recognition and judgement, cause microorganism and the viral infection state that prescription on individual diagnosis person's reproduction is infected with auxiliary diagnosis, comprise bioluminescence microscope, industrial video camera or digital camera, image pick-up card and computing machine, it is characterized in that: also comprise fluorescence light source lamp, stabilized voltage supply, shooting extend neck and computer numeral camera; Described fluorescence light source lamp provides direct supply by stabilized voltage supply; Described shooting extend neck is used for connecting bioluminescence microscope and industrial video camera or digital camera; By described image pick-up card, described industrial video camera is connected with computing machine, perhaps directly is connected with computing machine with described digital camera; Described computer numeral camera is used to take patient's diseased region picture and patient's head portrait; Described computing machine infects in the Intelligent Recognition software control system each device by reproduction and realizes systemic-function, comprise the bitmap that obtains picture, obtain bitmap matrix, obtain bitmap matrix average, carry out medium filtering, carry out Laplce's filtering, carry out image segmentation, carry out that the image zonule is cut apart, ferret out, carry out goal filtering, draw target.
2, intelligent identifying system is infected in reproduction as claimed in claim 1, it is characterized in that the functional module that the described reproduction infection Intelligent Recognition software that moves comprises has: image acquisition, figure library management, Intelligent Recognition, electronic health record, case history, inspection report, healthy reproduction file card and system maintenance in detail in computing machine.
3, intelligent identifying system is infected in reproduction as claimed in claim 1 or 2, it is characterized in that: can Intelligent Recognition cause that microorganism and virus that reproduction is infected comprise: Chlamydia, mycoplasma, herpes simplex virus, Candida albicans, gonococcus and Gartner bacterium.
4, intelligent identification Method is infected in a kind of reproduction, in order to the bioluminescence microscope picture that is input in the computing machine is carried out Intelligent Recognition and judgement, cause microorganism and the viral infection state that prescription on individual diagnosis person's reproduction is infected with auxiliary diagnosis, it is characterized in that, comprise step: the bitmap that reads picture; Obtain bitmap matrix; Obtain the average of bitmap matrix; Carry out medium filtering; Carry out Laplce's filtering; Carry out image segmentation; Carrying out the image zonule cuts apart; Ferret out; Carry out goal filtering; Draw number of targets.
5, intelligent identification Method is infected in reproduction as claimed in claim 4, it is characterized in that: can Intelligent Recognition cause the Chlamydia that reproduction is infected, comprise step: the bitmap that reads picture; Obtain green value, red value, blue value and saturation degree matrix; Obtain green value, red value, blue value and saturation degree average; Green value is carried out the medium filtering first time; Green value is carried out Laplce's filtering; Green value is carried out the medium filtering second time; Carry out image segmentation; Carry out goal filtering; Draw preliminary target number; Regain green value; Obtain the threshold value of green value; Green value is carried out medium filtering; Green value is carried out Laplce's filtering; Partitioning parameters is set; Carry out image segmentation; Current green value matrix is backed up; Carry out primary ferret out; Call in the green value matrix of backup; Carry out secondary ferret out, carry out ferret out for the third time again; Draw number of targets.
6, intelligent identification Method is infected in reproduction as claimed in claim 4, it is characterized in that: can Intelligent Recognition cause the mycoplasma that reproduction is infected, comprise step: the bitmap that reads picture; Obtain green value matrix; Obtain green value average; Obtain the threshold value of green value; Green value is carried out medium filtering; Green value is carried out Laplce's filtering; Partitioning parameters is set; Carry out image segmentation; Ferret out was singly all searched for up to all targets then, finish search, and the search of each target comprised substep: obtain the green value average of target; Obtain the threshold value of the green value of target; Boundary rectangle extends out 7 pixels and obtains new matrix; New matrix is carried out image segmentation; Carry out precise search again; Carry out goal filtering; Draw number of targets.
7, intelligent identification Method is infected in reproduction as claimed in claim 4, it is characterized in that: can Intelligent Recognition cause the herpes simplex virus that reproduction is infected, comprise step: the bitmap that reads picture; Obtain gray scale, green value and luminance matrix; Obtain gray scale, green value and brightness average; Green value is carried out medium filtering twice; Gray scale is carried out the medium filtering first time; Gray scale is carried out Laplce's filtering; Gray scale is carried out the medium filtering second time; Carry out image segmentation; Target is added up; Draw number of targets.
8, intelligent identification Method is infected in reproduction as claimed in claim 4, it is characterized in that: can Intelligent Recognition cause the Candida albicans that reproduction is infected, comprise step: the bitmap that reads picture; Obtain gray scale, green value and saturation degree matrix; Obtain grey level histogram; Green value is carried out medium filtering twice; Gray scale is carried out the medium filtering first time; Gray scale is carried out Laplce's filtering; Gray scale is carried out the medium filtering second time; Obtain green value, brightness and saturation degree histogram; Obtain analytical parameters; Carry out image segmentation; Target is added up; Draw number of targets.
9, intelligent identification Method is infected in reproduction as claimed in claim 4, it is characterized in that: can Intelligent Recognition cause the gonococcus that reproduction is infected, comprise step: the bitmap that reads picture; Obtain red value, green value, blue value and saturation degree matrix; Obtain red value, green value, blue value and saturation degree average; Green value is carried out the medium filtering first time; Green value is carried out Laplce's filtering; Green value is carried out the medium filtering second time; The image segmentation parameter is set; Carry out image segmentation; Regain green value matrix; Green value is carried out medium filtering for the third time; Be provided with that the zonule is cut apart and filtration parameter; Carrying out the zonule cuts apart; Carrying out target filters for the first time; Regain the green value matrix of full figure; Green value is carried out medium filtering the 4th time; Image partitioning parameters again is set; Again carrying out full figure cuts apart; Delete overlapping target; Carry out successively target second and third, filter for four and the 5th times; Obtain the gray scale extreme point; Judge the number of gray scale extreme point, if be zero, then after obtaining green value extreme point, carry out the 6th, seven, eight time of target and filter, otherwise directly carry out the 6th, seven, eight filtration of target; Increase object count; Draw number of targets.
10, intelligent identification Method is infected in reproduction as claimed in claim 4, it is characterized in that: can Intelligent Recognition cause the Gartner bacterium that reproduction is infected, comprise step:
Read the bitmap of picture;
Obtain red value, green value and saturation degree matrix;
Obtain red value, green value and saturation degree average;
Green value is carried out medium filtering;
Green value is carried out Laplce's filtering;
The image segmentation parameter is set;
Carry out image segmentation;
Delete overlapping target;
Judge whether to satisfy the zonule and cut apart condition;
If not satisfying the zonule cuts apart condition then carries out:
Regain the green value matrix of full figure;
Green value is carried out medium filtering;
Green value segmentation threshold is set;
Carrying out the zonule once cuts apart;
Carrying out target filters for the first time;
Carry out steps A-J as described below;
If not satisfying the zonule cuts apart condition then directly carries out following steps A-J:
A. regain the green value matrix of full figure;
B. green value is carried out medium filtering;
C. the green value matrix of negative-appearing image full figure;
D., the secondary splitting parameter is set;
E. carry out the zonule secondary splitting;
F. carry out second and third time filtration of target;
G. obtain extreme point;
H. carrying out the 4th time of target filters;
I. increase object count;
J. draw number of targets.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB031402585A CN100483432C (en) | 2003-08-26 | 2003-08-26 | Genital-infection inteeligent identification system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB031402585A CN100483432C (en) | 2003-08-26 | 2003-08-26 | Genital-infection inteeligent identification system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1581197A CN1581197A (en) | 2005-02-16 |
| CN100483432C true CN100483432C (en) | 2009-04-29 |
Family
ID=34579281
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB031402585A Expired - Lifetime CN100483432C (en) | 2003-08-26 | 2003-08-26 | Genital-infection inteeligent identification system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100483432C (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105296592A (en) * | 2015-11-13 | 2016-02-03 | 浙江大学 | Automatic detection method for mycoplasma culture, identification, counting and drug sensitivity results |
-
2003
- 2003-08-26 CN CNB031402585A patent/CN100483432C/en not_active Expired - Lifetime
Also Published As
| Publication number | Publication date |
|---|---|
| CN1581197A (en) | 2005-02-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| DK2973397T3 (en) | Tissue-object-based machine learning system for automated assessment of digital whole-slide glass | |
| JP2022180419A (en) | Image analysis method, device, program, and method for manufacturing trained deep learning algorithm | |
| JP6453298B2 (en) | System and method for observing and analyzing cytological specimens | |
| US11226280B2 (en) | Automated slide assessments and tracking in digital microscopy | |
| US7526116B2 (en) | Automated microscopic sperm identification | |
| US8582924B2 (en) | Data structure of an image storage and retrieval system | |
| US20210264595A1 (en) | System and Method for Detection and Classification of Objects of Interest in Microscope Images by Supervised Machine Learning | |
| JP4391527B2 (en) | A system that organizes multiple objects of interest within a field of interest | |
| US20220390735A1 (en) | Method and device for capturing microscopy objects in image data | |
| JP4864709B2 (en) | A system for determining the staining quality of slides using a scatter plot distribution | |
| CN101228283A (en) | Method for detecting and quantitating multiple subcellular components | |
| CN114049330A (en) | Method and system for fusing fluorescence characteristics in fluorescence in-situ hybridization image | |
| Hu et al. | Automatic detection of tuberculosis bacilli in sputum smear scans based on subgraph classification | |
| EP0693679A1 (en) | Particle image sorting system | |
| CN100483432C (en) | Genital-infection inteeligent identification system and method | |
| JP4897488B2 (en) | A system for classifying slides using a scatter plot distribution | |
| CN116883323A (en) | Multiple immunofluorescence image analysis system based on computer vision | |
| CN116757998A (en) | Screening method and device for CTC cells and CTC-like cells based on AI | |
| US20220383629A1 (en) | Label-free cell classification and screening system based on hybrid transfer learning | |
| WO2022158843A1 (en) | Method for refining tissue specimen image, and computing system performing same | |
| JP7391285B2 (en) | Program, information processing device, information processing method, and model generation method | |
| JP2007510893A (en) | Automated analysis of cell samples | |
| CN112750118A (en) | Novel method and system for identifying cell number in single cell pore plate sequencing based on automatic visual detection | |
| CN116113980A (en) | Urine analyzer, method for detecting bacteria in urine and storage medium | |
| CN116824579B (en) | Method and device for detecting yarrowia pneumocystis based on direct immunofluorescence staining |
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 | ||
| CX01 | Expiry of patent term | ||
| CX01 | Expiry of patent term |
Granted publication date: 20090429 |