CN102564924A - Automatic scanning method of single-frame image of blood cell - Google Patents
Automatic scanning method of single-frame image of blood cell Download PDFInfo
- Publication number
- CN102564924A CN102564924A CN2012100272806A CN201210027280A CN102564924A CN 102564924 A CN102564924 A CN 102564924A CN 2012100272806 A CN2012100272806 A CN 2012100272806A CN 201210027280 A CN201210027280 A CN 201210027280A CN 102564924 A CN102564924 A CN 102564924A
- Authority
- CN
- China
- Prior art keywords
- image
- control signal
- microscope
- haemocyte
- camera
- Prior art date
Links
- 210000000601 Blood Cells Anatomy 0.000 title abstract description 9
- 239000000969 carriers Substances 0.000 claims description 17
- 238000004891 communication Methods 0.000 claims description 6
- 238000000034 methods Methods 0.000 abstract description 15
- 238000004364 calculation methods Methods 0.000 abstract description 2
- 230000003287 optical Effects 0.000 description 5
- 210000004369 Blood Anatomy 0.000 description 4
- 239000008280 blood Substances 0.000 description 4
- 238000010586 diagrams Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 239000002245 particles Substances 0.000 description 4
- 238000004458 analytical methods Methods 0.000 description 3
- 230000000875 corresponding Effects 0.000 description 3
- 239000004744 fabrics Substances 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 210000004027 cells Anatomy 0.000 description 2
- 230000003412 degenerative Effects 0.000 description 2
- 238000005516 engineering processes Methods 0.000 description 2
- 241000287181 Sturnus vulgaris Species 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 230000001276 controlling effects Effects 0.000 description 1
- 230000001808 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reactions Methods 0.000 description 1
- 230000001419 dependent Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000006073 displacement reactions Methods 0.000 description 1
- 239000003814 drugs Substances 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 150000002367 halogens Chemical class 0.000 description 1
- 229910052736 halogens Inorganic materials 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 210000002865 immune cell Anatomy 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000003068 static Effects 0.000 description 1
- 238000005309 stochastic process Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 230000003313 weakening Effects 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Abstract
Description
Technical field
The present invention relates in microscopic system,, and then obtain the method for blood cell image clearly through the movement velocity of control automatic carrier and the shooting time of camera.
Background technology
Traditional medical science micro measurement generally is through optical microscope sample to be carried out visualization by the testing staff, and with hand-written report, the manual count device, or the form of Freehandhand-drawing figure record is accomplished.Conventional microscope is examined inefficiency, intensity is big; Have artificial malobservation, detection person's technical merit is had relatively high expectations, and it can only carry out image recording through cinephotomicrography; Can not carry out necessary processing to image; Can not quick storage and reproduction, more can not pass through network long-distance transmissions image information, thereby adapt to the requirement of modern medicine micro-image development far from.Being applied as of micrometron method solves that problems that manual method brings provide maybe.Micrometron can be discerned the leucocyte in the blood cell sample in a short period of time reliably, accurately measures its color, and shape and texture are classified to them, and can not introduce artificial inevitable subjective factor.
Require the whole surface of blood film that blood cell sample is processed is observed in the blood cell morphology inspection, still, usually, at this moment microscopical field range will just need carry out panoramic scanning work significantly less than whole blood film sample areas.Autoscan is the basis that the haemocyte inspection is analyzed, and it provides the raw data that realizes that image splices and sews up automatically.Carrying out when haemocyte takes pictures simultaneously; Often because the relative motion between camera and the haemocyte; Cause take pictures image blurring, thereby brought certain trouble with using for image studies, for the ease of the haemocyte target is studied; Need carry out sharpening to image, this Technology Need image recovers.For the recovery of blurred picture, generally can adopt two kinds of methods.
A kind of method is applicable to the situation that image is lacked priori, and can set up model this moment to degenerative process (fuzzy and noise), describes, and then seek a kind of process of removing or weakening its influence.Because this method is attempted the situation before the influence of estimated image degenerative process, so be a kind of method of estimation.
A kind of in addition method is applicable to that original image is had enough prioris, can set up a mathematical model and blurred picture is recovered original image.For example; Suppose only to contain in the known image circular object (for example star, particle, cell or the like) of confirming size; Like this, owing to be seldom Several Parameters (number, position, amplitude or the like) the unknown of original image only, so this is a detection problem.
Motion blur image generally can be represented with following linear shift invariant system:
（1）
Here (x y) is original image to f, and (x y) is corresponding blurred picture to g, and (x y) is the point spread function of system to h, and (x y) is random noise to n.
In motion blur image, fuzzy distance is meant the track scope of pixel motion in the original image.(x y) has significance to fuzzy distance for the point spread function h that confirms system.Utilize in the blurred picture frequency domain distance between two zero points to obtain fuzzy distance, and then the point spread function of definite system.But this method has certain limitation, can not be used for asking fuzzy that other motion causes.Though Wiener filtering has suppressed noise to a certain extent; On the lowest mean square meaning, also reach optimum, and improved the quality of image to a certain extent, but owing to point spread function can not accurately be confirmed; And suppose that real system is a stationary stochastic process; This differs bigger with image blurring actual conditions, so recover concrete motion blur image
Effect is not necessarily best.Someone proposes the method for maximum a posteriori probability (MAP) and recovers, and this method supposition system is a two-dimensional linear shift invariant system, and system is set up the AR model carry out image and recover and parameter estimation.But recover image in this way, need carry out a large amount of calculating, can not satisfy the requirement of fast processing.Savakis and Trussell proposition utilize the method for image power spectrum to confirm the parameter of system, but this method is too responsive to noise, and under the situation that noise exists, poor effect need be carried out a large amount of calculating, can not satisfy the requirement of real-time.
Summary of the invention
The present invention provides a kind of automatic scanning method of haemocyte single-frame images, and to solve under the situation that noise exists, poor effect need be carried out a large amount of calculating, problem that can not requirement of real time.
The technical scheme that the present invention takes is: it is made up of computing machine, electric machine controller, stepper motor, microscope automatic carrier, illuminator, microslide, microscopic system, digital camera, image pick-up card; Wherein, Digital camera is contained on the microscopical camera interface; The microscope automatic carrier is provided with the directions X stepper motor; The communication interface of computing machine is connected to the communication module of electric machine controller, and the output of electric machine controller connects stepper motor, and the output signal of digital camera is delivered to the image pick-up card of computing machine;
Said method comprises:
Microslide is put on the microscope automatic carrier;
Computing machine produces motion control signal and IMAQ control signal respectively; And said motion control signal provided to electric machine controller; The translational speed and the direction of motor controller controls microscope automatic carrier, movement speed v=frame per second F * resolution R * pixel dimension P/ enlargement ratio M;
The IMAQ control signal offers image pick-up card, and image pick-up card responds said control signal and catches image, in the said IMAQ control signal:
The haemocyte single-frame images time shutter: Δ T=p * Δ τ, p are the shared number of pixels of haemocyte, and wherein Δ τ is the time shutter of a pixel:
Δ τ=1/ (frame per second F * camera lateral resolution R C).
In the testing process of haemocyte, motor is the straight line uniform motion in the present invention, utilizes the information of blurred picture itself to obtain fuzzy distance; Confirm the point spread function of system; Thereby simplify the computation process of total system, improve detection speed, make system reach the real-time requirement.
The present invention is through the cooperation between microscope, moving stage, camera system and the computer system; Promptly according to the characteristics of linear electric motors uniform motion; Utilize the information of blurred picture itself to obtain fuzzy distance, confirm the point spread function of system, thereby simplify the computation process of total system; Improve detection speed, make system reach the real-time requirement.
Description of drawings
Fig. 1 is the integrally-built synoptic diagram of automatic scanning system that blood cell image is shown;
Fig. 2 illustrates the overlapping region synoptic diagram of taking when blood film at the uniform velocity moves;
Fig. 3 (a) is the piece image that 3 pixels are only arranged, and very strong correlativity is arranged between the pixel in the background;
Fig. 3 (b) is that three pixels among Fig. 3 (a) cause fuzzy picture along horizontal motion, and fuzzy distance is 10 pixels;
Fig. 4 is the proportionate relationship synoptic diagram that the camera exposure time and the cycle of shooting are shown.
Embodiment
It is made up of computing machine 7, electric machine controller 9, stepper motor 3, microscope automatic carrier 1, illuminator 4, microslide 2, microscopic system 5, digital camera 6, image pick-up card 8; Wherein, Digital camera is contained on the microscopical camera interface; The microscope automatic carrier is provided with the directions X stepper motor; The communication interface of computing machine is connected to the communication module of electric machine controller, and the output of electric machine controller connects stepper motor, and the output signal of digital camera is delivered to the image pick-up card of computing machine;
Said method comprises:
Microslide 2 is put on the microscope automatic carrier 1;
Computing machine 7 produces motion control signal and IMAQ control signal respectively; And said motion control signal provided to electric machine controller 9; The translational speed and the direction of electric machine controller 9 control microscope automatic carriers 1, movement speed v=frame per second F * resolution R * pixel dimension P/ enlargement ratio M;
The IMAQ control signal offers image pick-up card, and the said control signal of image pick-up card 8 responses is caught image, in the said IMAQ control signal:
The haemocyte single-frame images time shutter: Δ T=p * Δ τ, p are the shared number of pixels of haemocyte, and wherein Δ τ is the time shutter of a pixel:
Δ τ=1/ (frame per second F * camera lateral resolution R C).
Fig. 1 is the integrally-built synoptic diagram of whole haemocyte automatic scanning system that this embodiment is shown, and is as shown in Figure 1, microscope automatic carrier 1 and stepper motor 3, and electric machine controller 9 connects successively.Moving by the directions X stepper motor of microscope automatic carrier 1 drives, and microscope automatic carrier 1 screw mandrel is advanced along directions X.Digital camera 5 is connected on the microscopical camera interface 6; Be connected to the image pick-up card 8 of computing machine through data line; Computing machine 7 produces motion control signal and IMAQ control signal respectively; And said motion control letter provided to electric machine controller 9 translational speed and the direction of electric machine controller 9 control microscope automatic carriers 1; The IMAQ control signal offers image pick-up card 8, and the said control signal of image pick-up card 8 responses is caught image.The IMAQ control signal is characterized in that controlling the shooting frame per second and the single frames time shutter of image pick-up card.
In the whole haemocyte autoscan process, imaging system cooperates suitable camera lens and light source to form mainly by vision sensor CCD camera.These three parts need be carried out choose reasonable and collocation according to the size of cell to be detected and the working environment at sample surface state and detection scene.Vision sensor is the electron device that optical information is converted into numerical information; The performance principal element of weighing vision sensor comprises: the resolution R of CCD, pixel dimension P, information transmission mode, signal to noise ratio (S/N ratio), to the sensitivity of light etc.; In addition; The working method of pressing the CCD camera is different, is divided into the staggered scanning camera and the camera two big classes of lining by line scan usually, generally; The staggered scanning camera is used under testee is static or speed is the very slow situation, and the camera of lining by line scan then is used to obtain the image information of moving object.For fear of the influence of optical distortion to picture quality, choosing CCD resolution R is 1024 * 1024.Pixel dimension and resolution of microscope coupling collect to such an extent that the haemocyte details is just abundant more, and the pixel dimension P that chooses is 4.65um.The enlargement factor M=10 of optical lens in the low power lens scanning process, just blood cell image will pass through the amplification of 10 times of optical lens.Light source adopts the 100W Halogen lamp LED that normal illumination is provided.
In whole haemocyte autoscan process, because at this moment microscopical field range will just need carry out panoramic scanning work significantly less than whole blood film sample areas.Autoscan is the basis that the haemocyte inspection is analyzed, and it provides the raw data that realizes that image splices and sews up automatically.In order to guarantee that image does not trail, image is not overlapping simultaneously in kinetic control system, and the starting point that then requires the second exposed frame time image is the end point of first frame.Second exposed frame is too early, and the image of shooting will be overlapping, and second exposed frame is late excessively, and image will occur making image can not produce continuous effect at interval, and is as shown in Figure 2.Obtain the computing formula (2) of objective table translational speed according to above requirement:
Movement speed v=frame per second F * resolution R * pixel dimension P/ enlargement ratio M (2)
In order to obtain distinct image, analyze the relation between time shutter and the blurred picture.
Utilize a width of cloth simple image to be example, the reason that blurred picture produces is described, and (b) with reference to figure 3 (a).Fig. 3 (a) is the piece image that 3 pixels are only arranged, and very strong correlativity is arranged between the pixel in the background.Fig. 3 (b) is that three pixels among Fig. 3 (a) cause fuzzy picture along horizontal motion, and fuzzy distance is 10 pixels.
In the haemocyte automatic scanning system, because motor is a linear uniform motion in one direction, elder generation carries out fuzzy analysis to an independent sub-picture wherein.Suppose that (x y) has a plane motion to image f, makes x 0(t) and y 0(t) be respectively the change component of on x and y direction, moving, T representes time of moving.The total exposure amount of recording medium is to open the back to the integration of closing during this period of time at shutter.Image after then fuzzy does
（3）
(x y) is blurred picture to g in the formula.More than be exactly because the image blurring continuous function model that target and video camera relative motion cause.If blurred picture is caused as linear uniform motion on the x direction by scenery, the value of image arbitrfary point, then fuzzy back does
（4）
X in the formula 0(t) be the component motion of scenery on the x direction, if the total displacement of image is a, total time is T, and then the speed of motion is x 0(t)=at/T.Then following formula becomes:
（5）
Discussed above is consecutive image, for discrete picture, following formula is carried out discretize get:
（6）
Wherein L is the integer approximation value of the number of pixels that scenery moves on the photo.△ t is that each pixel is to the fuzzy time factor that exerts an influence.Hence one can see that, and the pixel value of motion blur image is the adding up of product of original image respective pixel values and its time.See that from physical phenomenon in fact motion blur image is exactly that same scene image is through stack again after a series of range delay, the final image that forms.So, if we will simulate horizontal uniform motion blurred picture by a width of cloth picture rich in detail, can be undertaken by following formula:
（7）
We are appreciated that this motion blur and time are irrelevant like this, and only with the distance dependent of motion blur, under this condition, make test obtain simplifying.Because the motion blur image actual to a width of cloth because video camera is different, is difficult to know its time shutter and scenery movement velocity.
Also to simulate the horizontal direction uniform motion fuzzy for the method for available convolution.Its process can be expressed as:
（8）
Wherein
(?9?)
Be called fuzzy operator or point spread function, * representes convolution, and ((x y) representes observed degraded image (blurred picture) to g to f for x, y) expression original (clear) image.
Calculating and analysis that above derivation is just carried out to the process of single pixel; But in the autoscan process of haemocyte; Photograph to such an extent that the haemocyte smallest particles tends to account for a plurality of pixels; So just can derive this system point spread function H (x, y)=h (x, y) * p (10)
P is the shared number of pixels of haemocyte.
Can infer that from following formula (10) this system collects to such an extent that image all is a picture rich in detail p single pixel exposure within the time, calculate an independent pixel exposure time Δ τ so.
Can be with the time shutter of image, the resolution of object of which movement cycle and camera is got in touch the calculation exposure cycle.For example the camera lateral resolution is 1024 pixels, and the time shutter equals image and moves the time that pixel is used, and the moving camera lateral resolution used time of sum of image drift is one-period.Be illustrated in fig. 4 shown below:
T is the shooting cycle, and A is the time shutter, and B is the moving lateral resolution needed time of sum of image drift
The resolution (laterally) of the every frame period time/camera of time shutter Δ τ=camera
Δ τ=1/ (F * R C) (F is a frame per second, R CBe the camera lateral resolution) (11)
Ambiguity function through to image in this system carries out Dimension Reduction Analysis; Derived corresponding relation between point spread function and the single pixel combines corresponding relation between time shutter and the single pixel can obtain that in this system each haemocyte smallest particles accounts for system's time shutter of a plurality of pixels and time shutter relation that each haemocyte smallest particles accounts for single pixel is simultaneously:
ΔT=p×Δτ (12)
Top The whole calculations process is analyzed reason fuzzy in the image motion process, and the dimensionality reduction of the whole fuzzy system of having derived calculates, and has improved the real-time of total system.
Claims (1)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210027280.6A CN102564924B (en) | 2012-02-08 | 2012-02-08 | Automatic scanning method of single-frame image of blood cell |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210027280.6A CN102564924B (en) | 2012-02-08 | 2012-02-08 | Automatic scanning method of single-frame image of blood cell |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102564924A true CN102564924A (en) | 2012-07-11 |
| CN102564924B CN102564924B (en) | 2014-02-26 |
Family
ID=46410922
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210027280.6A CN102564924B (en) | 2012-02-08 | 2012-02-08 | Automatic scanning method of single-frame image of blood cell |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102564924B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109283678A (en) * | 2018-10-25 | 2019-01-29 | 西北农林科技大学 | A kind of fungal spore acquisition glass slide automatic feeding and adsorbent application device and method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050220357A1 (en) * | 2003-07-31 | 2005-10-06 | Toshihiro Rifu | Image enhancement or correction software, method, apparatus and system for substantially minimizing blur in the scanned image |
| JP2009232275A (en) * | 2008-03-24 | 2009-10-08 | Olympus Imaging Corp | Image pickup device |
| CN101852970A (en) * | 2010-05-05 | 2010-10-06 | 浙江大学 | Automatic focusing method for camera under imaging viewing field scanning state |
-
2012
- 2012-02-08 CN CN201210027280.6A patent/CN102564924B/en active IP Right Grant
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050220357A1 (en) * | 2003-07-31 | 2005-10-06 | Toshihiro Rifu | Image enhancement or correction software, method, apparatus and system for substantially minimizing blur in the scanned image |
| JP2009232275A (en) * | 2008-03-24 | 2009-10-08 | Olympus Imaging Corp | Image pickup device |
| CN101852970A (en) * | 2010-05-05 | 2010-10-06 | 浙江大学 | Automatic focusing method for camera under imaging viewing field scanning state |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109283678A (en) * | 2018-10-25 | 2019-01-29 | 西北农林科技大学 | A kind of fungal spore acquisition glass slide automatic feeding and adsorbent application device and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102564924B (en) | 2014-02-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Aod-net: All-in-one dehazing network | |
| US10674138B2 (en) | Autofocus system for a conventional camera that uses depth information from an array camera | |
| US10015469B2 (en) | Image blur based on 3D depth information | |
| CN107087107B (en) | Image processing apparatus and method based on dual camera | |
| TWI510086B (en) | Digital refocusing method | |
| KR101783847B1 (en) | Continuous autofocus based on face detection and tracking | |
| EP2760209B1 (en) | Image processing device, method, program and recording medium, stereoscopic image capture device, portable electronic apparatus, printer, and stereoscopic image player device | |
| Geiger et al. | Are we ready for autonomous driving? the kitti vision benchmark suite | |
| Huang et al. | An advanced single-image visibility restoration algorithm for real-world hazy scenes | |
| JP4938093B2 (en) | System and method for region classification of 2D images for 2D-TO-3D conversion | |
| US8749620B1 (en) | 3D light field cameras, images and files, and methods of using, operating, processing and viewing same | |
| CN102388402B (en) | Image processing apparatus and image processing method | |
| EP2255239B1 (en) | Image sensor and focus detection apparatus | |
| US8432434B2 (en) | Camera and method for focus based depth reconstruction of dynamic scenes | |
| EP3216216B1 (en) | Methods and systems for multi-view high-speed motion capture | |
| JP5138031B2 (en) | Method, apparatus and system for processing depth related information | |
| CN102164293B (en) | Image processing apparatus, image processing method and imaging device | |
| CN101485193B (en) | Image generating device and image generating method | |
| KR101023207B1 (en) | Video object abstraction apparatus and its method | |
| TWI281593B (en) | Projector with automatic focus adjustment | |
| CN1184502C (en) | Process for automatic focusing | |
| US8805112B2 (en) | Image sharpness classification system | |
| US8139823B2 (en) | Method for capturing images comprising a measurement of local motions | |
| CN101605209B (en) | Image pick up device and image reproduction device | |
| US10277809B2 (en) | Imaging device and imaging method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| C06 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| C10 | Entry into substantive examination | ||
| GR01 | Patent grant | ||
| C14 | Grant of patent or utility model | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 130012 Changchun province high tech Zone, the River Street, No. 95, No. Patentee after: Medical Polytron Technologies Inc Address before: 130012 Changchun province high tech Zone, the River Street, No. 95, No. Patentee before: Changchun Dirui Medical Technology Co., Ltd. |
|
| CP01 | Change in the name or title of a patent holder |