CN102304472A - Device and method for detecting sputum smear tubercle bacillus quickly - Google Patents
Device and method for detecting sputum smear tubercle bacillus quickly Download PDFInfo
- Publication number
- CN102304472A CN102304472A CN201110128958A CN201110128958A CN102304472A CN 102304472 A CN102304472 A CN 102304472A CN 201110128958 A CN201110128958 A CN 201110128958A CN 201110128958 A CN201110128958 A CN 201110128958A CN 102304472 A CN102304472 A CN 102304472A
- Authority
- CN
- China
- Prior art keywords
- image
- visual field
- tubercule bacillus
- detection
- micro
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention relates to a device and a method for detecting sputum smear tubercle bacillus quickly. The device comprises a support, a charge coupled device (CCD) image sensor, a microscopic imaging system, a computer system, a three-dimensional accurate control console, an X-axis motor, a Y-axis motor, a Z-axis motor, wherein a high-accuracy grating ruler with a grating sensor is arranged in the Z-axis direction of the support additionally; a light-emitting diode (LED) light source plate of which the light intensity is controllable is positioned on the lower side of the microscopic imaging system and is used as an object stage; and the computer system comprises three software functional modules, namely a control module, an image target fusion module and a neural network classification and counting module. The method comprises a visual sense area interpretation algorithm and an optimum detection path scanning algorithm which are used for image detection and a method for fusing targets twice in image segmentation. By the device and the method, the detection accuracy and efficiency of machine vision can be improved, and the scientific reference is provided for the clinical diagnosis and prevention and treatment of tuberculosis.
Description
Technical field
The invention belongs to biomedical diagnostic techniques field, the apparatus and method of particularly a kind of rapid detection phlegm smear tubercule bacillus of detecting.
Background technology
Tuberculosis is the chronic infectious disease that is caused by mycobacterium tuberculosis infection; Tuberculosis is called tuberculosis and " white pestilence " again; It is a kind of ancient transmissible disease; It is current global range to one of infectious diseases of the most threatening property of the mankind; It is the highest disease of mortality ratio in the infectious diseases due to single cause of disease; Be the No.1 infectivity killer of developing country, become great public health problem and social concern.
The develop rapidly of science and technology brings a lot of novel methods for the tuberculosis laboratory diagnosis, but bacteriology checking is because its objectivity, intuitive and advantage such as easy, inexpensive will be the important means of tuberculosis experiment, diagnosis.Pulmonary tuberculosis, discovery contagium, observe the curative effect are made a definite diagnosis by various countries in the world at present, whether decision cures and the main method of epidemiology survey statistics is exactly that patient's sputum is made smear staining; Carry out microscopy, the bar number through searching tubercule bacillus in the sputum smear micro-image is as the main basis for estimation of the state of an illness.This method employing microscope amplifies about 1000 times to the phlegm smear and carries out manual observation, tubercule bacillus number in the interpretation micro-image.Usually, (size 2.0 * 2.5cm) will be carried out 300 different visuals field of continuous review at most, and it is negative still positive just to make a definite diagnosis acid-fast bacilli to be coated with the phlegm sheet to one.Phlegm smear for microscopic examination method is the most basic bacteriology checking method of all using outside the tuberculosis laboratory of various countries, the present whole world; It also is the important component part of China's tuberculosis control planning (NTP); China generally all adopts this method from the township to provincial tuberculosis medical treatment at different levels and controlling organization at present, and it plays indispensable vital role in China's tuberculosis prophylaxis work.
But there are many deficiencies in the manual detection method.At first, adopt manual observation, observe 300 visuals field at most continuously in the MIcrosope image of amplifying 1000 times, workload greatly, very easily makes the people tired; Secondly, artificial pair cell is discerned, and its detected result is subject to the influence of subjective factors such as people's interpretation technology and mood, is difficult to guarantee the exactness of detected result, is unfavorable for that tuberculosis detects and the quality control that prevents.Have, the detection efficiency of this method is low again, is difficult to adapt to the needs of the modernized prevention of infectious disease and control.In order to make existing bacteriology checking method lungy realize stdn, standardization, intellectuality and automatization; Improve the scientific and technological level of this detection method; Proposed to adopt the tubercule bacillus automatic testing method of machine vision at present; But, limit its application because it exists many gordian techniquies not solve.
Summary of the invention
Technical problem to be solved by this invention provides the apparatus and method of a kind of rapid detection phlegm smear tubercule bacillus; Realize the dynamic and intelligent identification and counting in the continuous visual field of same phlegm smear, with problem such as solve that conventional machines vision detection system focusing speed is slow, detection efficiency is low and the micro-image dynamicrange is little.
The present invention provides the device of a kind of rapid detection phlegm smear tubercule bacillus, comprises support, ccd image sensor, micro imaging system; Computer system, three-dimensional accurately supervisory control desk, X-axis motor; Y-axis motor, the Z spindle motor is equipped with the high precision grating chi of grating belt transmitter on the described support Z-direction; The led light source plate that controlled light intensity is arranged below the described micro imaging system is as Stage microscope; Described computer system comprises: output a control signal to three-dimensional accurately supervisory control desk control module, go forward side by side image object Fusion Module and the output detection information of line data fusion treatment to micro imaging system and discern and count the neural network classification and the counting module of processing from ccd image sensor reading images information.
Described control module is implemented in visual field conversion in different exposure image collections of the same visual field and the automatic testing process, light intensity control and focusing control.
Described image object Fusion Module earlier with the single exposure image to image do color cut apart with gray processing after boundary operator cut apart, the result who obtains is done Data Fusion, realization single-frame images target is cut apart; Segmentation result to each single exposure picture frame of the same visual field carries out target level view data fusion treatment more then, realizes that the same visual field detects accurately cutting apart of target.
Described neural network classification and counting module extract the tubercule bacillus characteristic, and these characteristics are input to neural network handle, and realize identification and the counting of the tubercule bacillus in the micro-image at last.
The present invention provides a kind of method of using the device of rapid detection phlegm smear tubercule bacillus, may further comprise the steps:
(i) start-up system is carried out initialize, and calibrates, input detected person information, and with on the led light source plate of phlegm smear placement as Stage microscope;
(ii) system begins to get into the automatic focusing process, and the accurate grating chi of Controlling System utilization obtains the Z axis information that is placed on Stage microscope, and according to the pinpointed focus of system's micro-imaging, the control detection platform moves to the optimal imaging position;
(iii) the detection and localization visual field utilizes Image Definition to carry out the automatic accurate focusing of digitizing;
(iv) adopt the clear area decision making algorithm, find first to detect the visual field, and confirm that best scanning detects the path;
(v) the scanning according to the best detects the path, the lock-in detection visual field, and regulation and control light source light intensity is gathered four frame micro-images of the different exposures of the same visual field respectively, and is deposited internal memory in;
(vi) the different micro-images that make public of four frames of gathering being carried out target with colouring information and half-tone information cuts apart; Carry out the single-frame images Target Fusion then; The target that micro-image detected to the different exposures of the same visual field four frames; Carry out the fusion of target level, be partitioned into the detection target of the same visual field;
(vii) the image that merges is carried out filtering earlier, extract the tubercule bacillus cell characteristic then;
(viii) the tubercule bacillus cell characteristic that extracts is input to neural network classifier, carries out tubercule bacillus classification and counting, the tubercule bacillus of accomplishing a visual field detects;
(ix) whether the quantity that detects tubercule bacillus if not, is changeed step and (v), carry out the detection in the next visual field, otherwise the complete detected result of demonstration smear is withdrawed from detection system greater than set(ting)value.
Beneficial effect
The present invention provides the apparatus and method of a kind of rapid detection phlegm smear tubercule bacillus; Replace artificial interpretation and counting; Overcome the shortcoming of artificial interpretation; Realize the dynamic and intelligent identification and counting in the continuous visual field of same phlegm smear; Whole testing process fully automated; Can improve conventional machines vision detection system focusing speed and detection efficiency; Improve accuracy in detection; Enlarge problems such as the micro-image dynamicrange is little; Help to improve the level of tubercule bacillus conventional sense method simultaneously, for clinical diagnosis lungy and control provide scientific basis.
Description of drawings
Fig. 1 is the phlegm micro-image tubercule bacillus device for dynamically detecting synoptic diagram of filming.
Fig. 2 is the blank visual area and detection path synoptic diagram of micro-image.
Fig. 3 is that the different exposure images of the same visual field detect the Target Fusion synoptic diagram.
Fig. 4 is a tubercule bacillus characteristic synoptic diagram.
Fig. 5 is the fast automatic detecting schema.
1. three-dimensional accurately supervisory control desk 7.X spindle motor 8.Y spindle motor 9.Z spindle motors of support 2.CCD image sensor 3. grating chis 4. micro imaging system 5.LED light source boards 6..
Embodiment
Below in conjunction with specific embodiment, further set forth the present invention.Should be understood that these embodiment only to be used to the present invention is described and be not used in the restriction scope of the present invention.Should be understood that in addition those skilled in the art can do various changes or modification to the present invention after the content of having read the present invention's instruction, these equivalent form of values fall within the application's appended claims institute restricted portion equally.
Embodiment 1
The present invention adopts following equipment in implementation process:
1.CCD model is HV1303UM, significant parameter: resolving power is the CMOS CCD of 1280*1024; Optical dimensions is 1/1.8 "; The highest 1,300,000 pixels; The highest horizontal resolution is 1280; The digital-to-analog conversion precision is 10bit; The high s/n ratio of 45dB (AGC OFF); Can control (AGC) by open/close automatic gain, the digital gain multiple is: * 2, * 1, * 0.5, * 0.25; For the light source of 550nm, its sensitivity is: 2.1V/Lux-s; Can be from the black-to-white level weighing apparatus correcting mode of motion tracking (ATW)/manually setting; Can accept two kinds of power supply supplies of 24V AC and 12V DC.
2. micro imaging system 4: adopt to meet German standard DIN standard, can directly be connected with DIN eyepiece or C interface CCD camera.The monotubular dimensions length is fit to the DIN standard, and any DIN standard lens all can be received this micro imaging system, and enlargement ratio equals the multiplying power of object lens.The visual field of video system equals the enlargement ratio of the size of CCD divided by object lens.Camera lens adopts the Nikon camera lens, and magnification is 100; Numerical aperture is 1.25; The oil mirror; Lens barrel length/optimal imaging distance is: 160/0.17.
3. grating chi 3:SM12; Measure length 12.5mm resolving power 0.1 μ m accuracy ± 1 μ m/12mm; The withstand voltage 0.4-0.8N of maximum measuring speed 0.5ms-1 spring, operating voltage 5Vss ± 5%, electric current (output of LD linear differential) is 130mA to the maximum; The maximum 50mA of electric current (TTL output); Degree of protection IP40, working temperature 0-50 ℃, the insulation impedance minimum is 20MW; Output signal (output of LD linear differential) RS422 is 20mA, the maximum 0.5V at＜10mA of output signal (TTL output) L.The minimum 3.5V at＞2.5mA of H.
4.LED controlled light intensity back light: YHFL-100-100-W electric parameter 24V/11.5W, eight grades of controlled change light intensity are arranged, when envrionment temperature is 25 ℃, surpass 30,000 hours (decrement is) at 50% o'clock with the continuous reliable operation of 50% white light source brightness.
5. computer system: Lenovo Qitian M7300:CPU, Intel Duo i3 550, double-core/four threads, frequency 3.2GHz L2 cache 2 * 256KB, three grades of buffer memory 4MB; Board chip set Intel H57, the mainboard memory size is 2GB; Type of memory is DDR3 1333; Hard-disk capacity 500GB, rotating speed 7200 changes SATAII; The video card type is solely to show AMD Radeon HD4350, video memory capacity 512MB, and display sizes is 17 inches.
As shown in Figure 1, the present invention relates to the device of a kind of rapid detection phlegm smear tubercule bacillus, comprise support 1; Ccd image sensor 2; Micro imaging system 4, computer system, three-dimensional accurately supervisory control desk 6; X-axis motor 7; Y-axis motor 8, Z spindle motor 9 is equipped with the high precision grating chi 3 of grating belt transmitter on the described support Z-direction; Can accomplish obtaining automatically of smear Z axis information fast, to realize the rapid focus of detection system to smear; The led light source plate 5 that controlled light intensity is arranged below the described micro imaging system 4 is as Stage microscope; Led light source plate 5 is through becoming the light intensity transmission imaging; The multiframe micro-image that obtains the different exposures of the same visual field carries out information fusion and does, to enlarge the dynamicrange of scene image information; Described computer system comprises three software function modules: output a control signal to the control module of three-dimensional accurately supervisory control desk 6, the image object Fusion Module of handling the graphic information that reads from ccd image sensor 2 and processing and output detection information neural network classification and the counting module to micro imaging system 4; Described control module is implemented in visual field conversion in different exposure image collections of the same visual field and the automatic testing process, light intensity control, focusing control; Described image object Fusion Module earlier with the single exposure image to image do color cut apart with gray processing after boundary operator cut apart, the result who obtains is done Data Fusion, realization single-frame images target is cut apart; Segmentation result to each single exposure picture frame of the same visual field carries out target level view data fusion treatment more then, realizes that the same visual field detects accurately cutting apart of target; Described neural network classification and counting module extract the tubercule bacillus characteristic, and these characteristics are input to neural network handle, and realize identification and the counting of the tubercule bacillus in the micro-image at last.
Detection of dynamic to a phlegm smear tubercule bacillus is in 800-1000 optical microphotograph imaging system 4 doubly, to carry out, and thickness, uniformity coefficient and the carrier table motion in the detection of dynamic process that seized phlegm smear is smeared all can cause the out of focus phenomenon.So propose a kind of automatic focusing system of full digital micro-imaging.Consider focusing accuracy and focusing efficient, intend to adopt a kind of thick/smart bonded focus adjustment method, promptly at first the micro-image at general profile and edge is carried out rough focusing based on the marginal sharpness algorithm; Adopt sharpness evaluation function to carry out accurate adjustment Jiao to certain target area that comprises tubercule bacillus on the panoramic picture then, make micro-image demonstrate more rich details based on the color high fdrequency component.
Owing in detecting focusing accuracy and speed are had requirement; Therefore be main execution unit with ceramic motor, adopt closed loop control method to realize optimum control, realize accurate control quick, high stability control process; Making it can reach precision is 1～2 μ m, satisfies the requirement of system.
In dynamic auto testing process, (20 * 25mm) will detect 300 visuals field at most, could judge detected result to same phlegm smear.Adopt visual field conversion sequence as shown in Figure 2 to detect, help to realize the Control and Optimization of testing process, reduce the displacement error of X, Y-axis.
Owing to reasons such as the influence of the factors such as non-linear, diffraction effect of detected object illumination unevenness, illumination intensity variations, opto-electronic conversion and artificial smear processing qualities, can cause the micro-image noise.In order to obtain micro-image clearly; Except adopting more advanced hardware device; Also need adopt the method for software that the view data that collects is carried out pre-treatment; Eliminate systematic error and random noise as much as possible; Make the calculation result of focusing evaluation function can reflect the out of focus information of imaging system accurately and rapidly, to realize accurate focusing.
Carry out pre-treatment, at first will eliminate the The noise that imaging process is introduced.The one, eliminate the influence of illumination.Because ccd image sensor 2 is when light is dark, imaging effect is relatively poor, thereby is necessary its gray-scale value is revised with enhancing contrast ratio; Also to eliminate micro imaging system 4 in addition and in imaging process, introduce other various noises.The composition of noise is different, and the method for the elimination noise of taking is also different.
Consider and in the original micro-image of tubercule bacillus,, mainly show as Gaussian noise and impulse noise though be mixed with various noises.Median filtering algorithm has simply, fast, can filtering impulse noise and salt-pepper noise, and stronger details hold facility, and mean filter has stronger filtering ability to Gaussian noise.The algorithm that the present invention adopts median filtering algorithm and mean filter algorithm to combine carries out pre-treatment to image, can reach good filtering effect.
Because censorship phlegm smear spreadable substance is limited to light intensity sensitivity and CCD dynamicrange, causes the micro-image dynamicrange limited, cause that detecting target information loses, if a visual field is only taken a two field picture and detected, will cause the tubercule bacillus omission.The present invention adopts and becomes the light intensity technology; The same visual field absorbs the micro-image of the different exposures of four frames; Carry out target respectively and extract, utilize again digital image processing techniques with the detection Target Fusion of the micro-images of the different exposures of the same visual field on a two field picture, supply subsequent algorithm to carry out classification processing.
Micro-image is through after the pre-treatment, must utilize the characteristics of microscopic image information and tubercule bacillus after the pre-treatment, effectively the cutting apart of realization tubercule bacillus.The present invention proposes after image is carried out pre-treatment, respectively to image do color cut apart with gray processing after boundary operator cut apart, and the result that will obtain does Data Fusion, obtains the border of cell, realizes cutting apart of single frames tubercule bacillus target image; To the segmentation result of the different four two field picture targets of making public of the same visual field, carry out the data fusion of target level then, obtain the detection target of the same visual field.Fig. 3 has shown the implementation procedure that the same visual field target is cut apart.Cut apart the certain characteristics that the sample cell is extracted in the back in target, mainly comprise characteristics such as area, girth, color and shape, these characteristics are input to neural network train, realize identification and the counting of the tubercule bacillus in the sputum smear micro-image at last.
Integration technology is exactly the data to the various information source that satisfies certain condition, carries out comprehensive treating process according to certain criterion, to obtain description more accurately.According to two kinds of segmentation results of top gained, and utilize the existing priori of tubercule bacillus (as: area etc.), in fusion and the extraction data being carried out eliminate the false and retain the true after the association.Result after fusion treatment can eliminate because the uneven influence that is brought of illumination and dyeing; Can reduce the influence that noise brought in the image dramatically again; And blending algorithm can also be removed the non-tubercule bacillus in the segmentation result twice, handles laying a solid foundation for next step identification.
Feature Extraction is the key of carrying out cell divide, after the tubercule bacillus in the micro-image is cut apart, must extract the CF characteristic that the tuberculosis cell is had.
Color characteristic is the key feature of identification tubercule bacillus; Through the phlegm smear that dyes and handled; Tubercule bacillus shows garnet; And other impurity generally shows blue-greenish colour, and background is yellowish redness, also has the color of some impurity and goal in research very similar; Sometimes; The color of background also can be also approaching with the color that detects target in some zone, as shown in Figure 4, the present invention proposes a definite method that is suitable for the color characteristic collection of native system characteristics.
Confirming characteristic color set in the tuberculosis sputum smear micro-image, adopt two principles, at first is the principle of maximum frequency, secondly is the compatible principle of color.The principle of maximum frequency has guaranteed that the feature of the N in the color set represented the distribution of the main component in the tubercule bacillus image, and each feature of consistency principles and requirements of color has distribution relatively uniformly in the color space.Adopt compatible spheroid spatial color model, i.e. each feature spheroid space that to have a radius be R, in the image color in this space all merger be this feature.Each feature is to each other apart from greater than R.Obviously, compatible scope is big more, and promptly R is big more, and then the quantity of feature is just few more, in detection, through the method for experiment, selectes the value of R.Value by R just can be come concrete definite characteristic color set according to algorithm.After image is represented with characteristic color, a certain micro-image zone described with regard to an available characteristic in right amount.
C
f＝[S(x
1)，S(x
2)，....S(x
k)]
T
Wherein: S (x
i) representation feature concentrates the number of I feature shared pixel in this image area.C
fThe vector of expression just can reflect the difference that exists between them.So can define a texture similarity: promptly:
Its value is big more, and then the colour characteristics in two zones is similar more.Characteristics according to tubercule bacillus; Choose a certain size rectangular window; Color property vector in the calculation window; Sample characteristics as tubercule bacillus; Then to each pixel in the image; Calculate its eigenvector in a certain size field window and with the similarity measure of sample characteristics, obtain the new images of a width of cloth like this about similarity measure.Obviously, estimate, can obtain a more complete target area with the method for thresholding according to the similar of image.
Tubercule bacillus picture shape Feature Extraction.Usually, tubercule bacillus has certain size, the tubercule bacillus in the smear micro-image shows elongated strip shaped more, the one-tenth bending that has, and the shape that is in line that has also has the ground tubercule bacillus to spiral into non-bar shape, also has many tubercule bacilluss to be intertwined into bulk sometimes.The size of tubercule bacillus, shape facility are the important evidence of identification; Based on tubercule bacillus ground characteristics; We adopt the chain representation based on the zone iimit; Extract the information from objective pattern of tuberculosis cell image. the existing calculating that is beneficial to relevant morphological specificity of this chain representation also helps saving storage space.Realize that with chain code image is carried out the edge to be followed the tracks of, and can obtain the geometric characteristic of a series of tuberculosis cells such as girth, area.Area, cell compartment rectangle degree and extensibility etc. like the width of the girth of tubercule bacillus, cell compartment and height, cell compartment.Through image process method, above-mentioned characteristic can extract in treating processes, according to the characteristic of the tuberculosis cell that extracts, for the rule of finding out correct discriminator provides foundation.
Effectively neural network classifier according to the characteristic of input picture, is realized the classification to the tuberculosis cell.
A kind of improvement algorithm that is the basis with domain theory self-adapting resonance neural network algorithm is suitable for the identification and the classification of tuberculosis cell.Through increasing the mapping ability of hidden neuron; Realize the hidden neuron differentiated treatment of inner input category of sample and inner output category; And conceal layer at two and eliminate competition respectively; Make network increase hidden neuron neatly, reduced the complexity of network topology to the defeated people/output characteristic of sample.In addition, when the inside of learning sample output category meets the demands, realize domain of attraction is carried out little moving, protected to greatest extent to be stored in the knowledge in the network.
As shown in Figure 5, the present invention may further comprise the steps in concrete use:
(i) start-up system is carried out initialize, and calibrates, input detected person information, and with on the led light source plate 5 of phlegm smear placement as Stage microscope;
(ii) system begins to get into the automatic focusing process, and the accurate grating chi 3 of Controlling System utilization obtains the Z axis information that is placed on Stage microscope, and according to the pinpointed focus of system's micro-imaging, the control detection platform moves to the optimal imaging position;
(iii) the detection and localization visual field utilizes Image Definition to carry out the automatic accurate focusing of digitizing;
(iv) adopt the clear area decision making algorithm, find first to detect the visual field, and confirm that best scanning detects the path;
(v) the scanning according to the best detects the path, the lock-in detection visual field, and regulation and control light source light intensity is gathered four frame micro-images of the different exposures of the same visual field respectively, and is deposited internal memory in;
(vi) the different micro-images that make public of four frames of gathering being carried out target with colouring information and half-tone information cuts apart; Carry out the single-frame images Target Fusion then; The target that micro-image detected to the different exposures of the same visual field four frames; Carry out the fusion of target level, be partitioned into the detection target of the same visual field;
(vii) the image that merges is carried out filtering earlier, extract the tubercule bacillus cell characteristic then;
(viii) the tubercule bacillus cell characteristic that extracts is input to neural network classifier, carries out tubercule bacillus classification and counting, the tubercule bacillus of accomplishing a visual field detects;
(ix) whether the quantity that detects tubercule bacillus if not, is changeed step and (v), carry out the detection in the next visual field, otherwise the complete detected result of demonstration smear is withdrawed from detection system greater than set(ting)value.
Claims (5)
1. the device of a rapid detection phlegm smear tubercule bacillus; Comprise support (1); Ccd image sensor (2), micro imaging system (4), computer system; Three-dimensional accurately supervisory control desk (6); X-axis motor (7), y-axis motor (8), Z spindle motor (9); It is characterized in that, the high precision grating chi (3) of grating belt transmitter is housed on described support (1) Z-direction; The led light source plate (5) that controlled light intensity is arranged below the described micro imaging system (4) is as Stage microscope; Described computer system comprises: output a control signal to three-dimensional accurately supervisory control desk (6) control module, go forward side by side image object Fusion Module and the output detection information of line data fusion treatment to micro imaging system (4) and discern and count the neural network classification and the counting module of processing from ccd image sensor (2) reading images information.
2. the device of a kind of rapid detection phlegm smear tubercule bacillus according to claim 1 is characterized in that, described control module is implemented in visual field conversion in different exposure image collections of the same visual field and the automatic testing process, light intensity control and focusing control.
3. the device of a kind of rapid detection phlegm smear tubercule bacillus according to claim 1; It is characterized in that; Described image object Fusion Module earlier with the single exposure image to image do color cut apart with gray processing after boundary operator cut apart; The result who obtains is done Data Fusion, and realization single-frame images target is cut apart; Segmentation result to each single exposure picture frame of the same visual field carries out target level view data fusion treatment more then, realizes that the same visual field detects accurately cutting apart of target.
4. the device of a kind of rapid detection phlegm smear tubercule bacillus according to claim 1; It is characterized in that; Described neural network classification and counting module extract the tubercule bacillus characteristic; And these characteristics are input to neural network handle, realize identification and the counting of the tubercule bacillus in the micro-image at last.
5. a method of using the device of the described rapid detection phlegm of claim 1 smear tubercule bacillus is characterized in that, may further comprise the steps:
(i) start-up system is carried out initialize, and calibrates, input detected person information, and with on the led light source plate (5) of phlegm smear placement as Stage microscope;
(ii) system begins to get into the automatic focusing process, and Controlling System utilizes accurate grating chi (3) to obtain the Z axis information that is placed on Stage microscope, and according to the pinpointed focus of system's micro-imaging, the control detection platform moves to the optimal imaging position;
(iii) the detection and localization visual field utilizes Image Definition to carry out the automatic accurate focusing of digitizing;
(iv) adopt the clear area decision making algorithm, find first to detect the visual field, and confirm that best scanning detects the path;
(v) the scanning according to the best detects the path, the lock-in detection visual field, and regulation and control light source light intensity is gathered four frame micro-images of the different exposures of the same visual field respectively, and is deposited internal memory in;
(vi) the different micro-images that make public of four frames of gathering being carried out target with colouring information and half-tone information cuts apart; Carry out the single-frame images Target Fusion then; The target that micro-image detected to the different exposures of the same visual field four frames; Carry out the fusion of target level, be partitioned into the detection target of the same visual field;
(vii) the image that merges is carried out filtering earlier, extract the tubercule bacillus cell characteristic then;
(viii) the tubercule bacillus cell characteristic that extracts is input to neural network classifier, carries out tubercule bacillus classification and counting, the tubercule bacillus of accomplishing a visual field detects;
(ix) whether the quantity that detects tubercule bacillus if not, is changeed step and (v), carry out the detection in the next visual field, otherwise the complete detected result of demonstration smear is withdrawed from detection system greater than set(ting)value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110128958A CN102304472A (en) | 2011-05-18 | 2011-05-18 | Device and method for detecting sputum smear tubercle bacillus quickly |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110128958A CN102304472A (en) | 2011-05-18 | 2011-05-18 | Device and method for detecting sputum smear tubercle bacillus quickly |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102304472A true CN102304472A (en) | 2012-01-04 |
Family
ID=45378390
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110128958A Pending CN102304472A (en) | 2011-05-18 | 2011-05-18 | Device and method for detecting sputum smear tubercle bacillus quickly |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102304472A (en) |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103453889A (en) * | 2013-09-17 | 2013-12-18 | 深圳市创科自动化控制技术有限公司 | Calibrating and aligning method of CCD (Charge-coupled Device) camera |
| CN103728183A (en) * | 2013-10-08 | 2014-04-16 | 广东石油化工学院 | Single fiber fragmentation test device based on machine vision and control method of single fiber fragmentation test device |
| CN104887322A (en) * | 2015-05-06 | 2015-09-09 | 赵甲辰 | Intelligent diagnoses and treatment table for lung abscess |
| CN104897633A (en) * | 2015-06-01 | 2015-09-09 | 宁波江丰生物信息技术有限公司 | Automatic mycobacterium tuberculosis screening system for mycobacterium tuberculosis sputum smear |
| CN104966282A (en) * | 2014-12-24 | 2015-10-07 | 广西师范大学 | Image acquiring method and system for detecting single erythrocyte |
| CN106635771A (en) * | 2015-10-29 | 2017-05-10 | 刘娜 | Integrated microbiological examination device |
| CN106873142A (en) * | 2017-03-15 | 2017-06-20 | 北方工业大学 | High-quality image acquisition device and method of tubercle bacillus detector |
| CN107058083A (en) * | 2017-05-12 | 2017-08-18 | 吴长静 | A kind of medical test quick detection Sputum smears tulase device |
| CN107426464A (en) * | 2017-09-01 | 2017-12-01 | 上海极清慧视科技有限公司 | A kind of pathology print scanning shoot method using ERS sensors |
| CN107505698A (en) * | 2017-07-17 | 2017-12-22 | 中国家用电器研究院 | A kind of biological identification counting device and method |
| CN108447047A (en) * | 2018-02-11 | 2018-08-24 | 深圳市恒扬数据股份有限公司 | Acid-fast bacilli detection method and device |
| CN108764329A (en) * | 2018-05-24 | 2018-11-06 | 复旦大学附属华山医院北院 | A kind of construction method of lung cancer pathology image data set |
| CN108918519A (en) * | 2018-07-05 | 2018-11-30 | 深圳辉煌耀强科技有限公司 | A kind of cell smear image obtains and analysis system |
| CN109919863A (en) * | 2019-02-15 | 2019-06-21 | 佛山市博朋生物科技有限公司 | A kind of full-automatic bacterial colony counting instrument, system and its method for counting colonies |
| CN110389131A (en) * | 2019-08-21 | 2019-10-29 | 汕头市结核病防治所(汕头市呼吸系疾病防治所) | A kind of sputum qualification tests view apparatus |
| CN110736747A (en) * | 2019-09-03 | 2020-01-31 | 深思考人工智能机器人科技（北京）有限公司 | cell liquid based smear under-mirror positioning method and system |
| CN110967302A (en) * | 2019-11-06 | 2020-04-07 | 清华大学 | Microbial panoramic smear detection device and detection method |
| CN112446427A (en) * | 2020-11-27 | 2021-03-05 | 王伟佳 | Method and device for identifying myeloid blood cells, storage medium and electronic equipment |
| CN112924452A (en) * | 2021-01-29 | 2021-06-08 | 西安博锐轶信息科技有限公司 | Blood examination auxiliary system |
| CN113884489A (en) * | 2021-09-29 | 2022-01-04 | 电子科技大学 | Grating ruler assisted positioning thick liquid layer cell automatic microscopic imaging method |
| CN114112576A (en) * | 2021-11-22 | 2022-03-01 | 驻马店市中心医院 | Mycobacterium tuberculosis drug resistance detection system based on acid-fast staining of sputum smear |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004092730A2 (en) * | 2003-04-15 | 2004-10-28 | Biogenon Ltd. | Method and device for detecting the presence of an analyte |
| US20060057707A1 (en) * | 2000-10-30 | 2006-03-16 | Sru Biosystmes, Inc. | Optical detection of label-free biomolecular interactions using microreplicated plastic sensor elements |
| CN101221118A (en) * | 2007-12-07 | 2008-07-16 | 东华大学 | System and method for intelligent recognizing and counting sputum smear micro-image tubercle bacillus |
| CN101225428A (en) * | 2007-01-19 | 2008-07-23 | 华南农业大学 | Dyeing liquid as well as dyeing method and uses thereof |
| AU2007273113B2 (en) * | 2006-07-07 | 2010-11-25 | Sru Biosystems, Inc. | Near ultraviolet-wavelength photonic-crystal biosensor with enhanced surface to bulk sensitivity ratio |
-
2011
- 2011-05-18 CN CN201110128958A patent/CN102304472A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060057707A1 (en) * | 2000-10-30 | 2006-03-16 | Sru Biosystmes, Inc. | Optical detection of label-free biomolecular interactions using microreplicated plastic sensor elements |
| WO2004092730A2 (en) * | 2003-04-15 | 2004-10-28 | Biogenon Ltd. | Method and device for detecting the presence of an analyte |
| AU2007273113B2 (en) * | 2006-07-07 | 2010-11-25 | Sru Biosystems, Inc. | Near ultraviolet-wavelength photonic-crystal biosensor with enhanced surface to bulk sensitivity ratio |
| CN101225428A (en) * | 2007-01-19 | 2008-07-23 | 华南农业大学 | Dyeing liquid as well as dyeing method and uses thereof |
| CN101221118A (en) * | 2007-12-07 | 2008-07-16 | 东华大学 | System and method for intelligent recognizing and counting sputum smear micro-image tubercle bacillus |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103453889B (en) * | 2013-09-17 | 2016-02-17 | 深圳市创科自动化控制技术有限公司 | Ccd video camera calibration alignment method |
| CN103453889A (en) * | 2013-09-17 | 2013-12-18 | 深圳市创科自动化控制技术有限公司 | Calibrating and aligning method of CCD (Charge-coupled Device) camera |
| CN103728183A (en) * | 2013-10-08 | 2014-04-16 | 广东石油化工学院 | Single fiber fragmentation test device based on machine vision and control method of single fiber fragmentation test device |
| CN103728183B (en) * | 2013-10-08 | 2016-02-10 | 广东石油化工学院 | A kind of fragmentation test device based on machine vision and control method thereof |
| CN104966282B (en) * | 2014-12-24 | 2017-12-08 | 广西师范大学 | A kind of image-pickup method and system for single blood erythrocyte by mocro detection |
| CN104966282A (en) * | 2014-12-24 | 2015-10-07 | 广西师范大学 | Image acquiring method and system for detecting single erythrocyte |
| CN104887322A (en) * | 2015-05-06 | 2015-09-09 | 赵甲辰 | Intelligent diagnoses and treatment table for lung abscess |
| CN104897633A (en) * | 2015-06-01 | 2015-09-09 | 宁波江丰生物信息技术有限公司 | Automatic mycobacterium tuberculosis screening system for mycobacterium tuberculosis sputum smear |
| CN106635771A (en) * | 2015-10-29 | 2017-05-10 | 刘娜 | Integrated microbiological examination device |
| CN106635771B (en) * | 2015-10-29 | 2019-04-16 | 陆文燕 | Integral type Micro biological Tests device |
| CN106873142A (en) * | 2017-03-15 | 2017-06-20 | 北方工业大学 | High-quality image acquisition device and method of tubercle bacillus detector |
| CN106873142B (en) * | 2017-03-15 | 2019-04-12 | 北方工业大学 | High-quality image acquisition device and method of tubercle bacillus detector |
| CN107058083A (en) * | 2017-05-12 | 2017-08-18 | 吴长静 | A kind of medical test quick detection Sputum smears tulase device |
| CN107505698A (en) * | 2017-07-17 | 2017-12-22 | 中国家用电器研究院 | A kind of biological identification counting device and method |
| CN107505698B (en) * | 2017-07-17 | 2024-01-12 | 中国家用电器研究院 | Biological identification counting device and method |
| CN107426464A (en) * | 2017-09-01 | 2017-12-01 | 上海极清慧视科技有限公司 | A kind of pathology print scanning shoot method using ERS sensors |
| CN108447047A (en) * | 2018-02-11 | 2018-08-24 | 深圳市恒扬数据股份有限公司 | Acid-fast bacilli detection method and device |
| CN108764329A (en) * | 2018-05-24 | 2018-11-06 | 复旦大学附属华山医院北院 | A kind of construction method of lung cancer pathology image data set |
| CN108918519A (en) * | 2018-07-05 | 2018-11-30 | 深圳辉煌耀强科技有限公司 | A kind of cell smear image obtains and analysis system |
| CN109919863A (en) * | 2019-02-15 | 2019-06-21 | 佛山市博朋生物科技有限公司 | A kind of full-automatic bacterial colony counting instrument, system and its method for counting colonies |
| CN109919863B (en) * | 2019-02-15 | 2023-06-20 | 佛山市博朋生物科技有限公司 | Full-automatic colony counter, system and colony counting method thereof |
| CN110389131A (en) * | 2019-08-21 | 2019-10-29 | 汕头市结核病防治所(汕头市呼吸系疾病防治所) | A kind of sputum qualification tests view apparatus |
| CN110736747A (en) * | 2019-09-03 | 2020-01-31 | 深思考人工智能机器人科技（北京）有限公司 | cell liquid based smear under-mirror positioning method and system |
| CN110967302A (en) * | 2019-11-06 | 2020-04-07 | 清华大学 | Microbial panoramic smear detection device and detection method |
| CN112446427A (en) * | 2020-11-27 | 2021-03-05 | 王伟佳 | Method and device for identifying myeloid blood cells, storage medium and electronic equipment |
| CN112446427B (en) * | 2020-11-27 | 2021-07-27 | 王伟佳 | Method and device for identifying myeloid blood cells, storage medium and electronic equipment |
| CN112924452A (en) * | 2021-01-29 | 2021-06-08 | 西安博锐轶信息科技有限公司 | Blood examination auxiliary system |
| CN113884489A (en) * | 2021-09-29 | 2022-01-04 | 电子科技大学 | Grating ruler assisted positioning thick liquid layer cell automatic microscopic imaging method |
| CN113884489B (en) * | 2021-09-29 | 2023-06-20 | 电子科技大学 | Automatic microscopic imaging method for thick liquid layer cells with auxiliary positioning of grating ruler |
| CN114112576A (en) * | 2021-11-22 | 2022-03-01 | 驻马店市中心医院 | Mycobacterium tuberculosis drug resistance detection system based on acid-fast staining of sputum smear |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102304472A (en) | Device and method for detecting sputum smear tubercle bacillus quickly | |
| CN101221118A (en) | System and method for intelligent recognizing and counting sputum smear micro-image tubercle bacillus | |
| CN101441320B (en) | High dynamic image acquisition device based on microscopic imaging detection and method thereof | |
| CN105574527B (en) | A kind of quick object detecting method based on local feature learning | |
| CN106210520B (en) | A kind of automatic focusing electronic eyepiece and system | |
| CN110146974A (en) | A kind of intelligent biological microscope | |
| CN103258316A (en) | Method and device for picture processing | |
| CN103201769A (en) | Image processing device, image processing method, program, integrated circuit | |
| CN103645573A (en) | Liquid crystal display (LCD) detection method and system based on machine vision | |
| CN101493312B (en) | Micro imaging high precision three-dimensional detection device and method | |
| JP2015123047A (en) | Image processing device and program | |
| CN103984979A (en) | Lens-diffraction-imaging-free automatic algae detection and counting device and method | |
| Zou et al. | Malaria cell counting diagnosis within large field of view | |
| CN109182081A (en) | A kind of unicellular separation system based on image processing model | |
| CN110378946A (en) | Depth map processing method, device and electronic equipment | |
| JP2012042669A (en) | Microscope control device and optical distortion correction method | |
| Shah et al. | Identification of robust focus measure functions for the automated capturing of focused images from Ziehl–Neelsen stained sputum smear microscopy slide | |
| CN101930606A (en) | Field depth extending method for image edge detection | |
| CN103217108A (en) | Method for detecting geometrical parameters of spectacle frame | |
| CN109001902A (en) | Microscope focus method based on image co-registration | |
| CN108830858A (en) | It is a kind of based on infrared and optical image double-mode imaging information living body method for counting colonies | |
| CN105180802A (en) | Identification method and device of object size information | |
| US11953670B2 (en) | System and method for rapid focusing and analysis using a micro-camera array microscope | |
| CN106254855B (en) | A kind of three-dimensional modeling method and system based on zoom ranging | |
| Xiang et al. | An edge detection algorithm based-on Sobel operator for images captured by binocular microscope |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120104 |