CN100465990C - Intelligent locating method face for micro-fluidic chip - Google Patents

Abstract

The invention relates to an intelligent positioning method facing a microfluidic chip, the use of chip planar images in CCD camera acquisition microfluidic chip analysis system, using morphological filters and OPTA (one-pass thinning algorithm) refinement algorithm to denoise and refine the chip ichnography, being a skeleton-map of the chip ichnography; based on chip skeleton map, extracts relevant node on the network of micro-channels of microfluidic chip; According to the joint position of all micro-channels and relations between the two connection, generates adjacent table; According to the adjacent table on the microfluidic chip for intelligent tracking positioning, makes the adjacent table feedback amendments through the relevant feedback algorithm in accordance with the positioning results. It is based on the relevant image processing technology designed with a fully automated positioning, high precision, high speed and can track position, and other characteristics, it can improve the automaticed analysis for all microfluidic chip analysis system.

Description

A kind of intelligent locating method towards micro-fluidic chip

Technical field

The present invention relates to intelligent mode identification and technical field of image processing, relate to a kind of intelligent locating method towards micro-fluidic chip.

Background technology

The micro-fluidic chip analytic system claims micro-total analysis system again, be that Manz and Widmer by Switzerland Ciba Geigy company began one's study at the beginning of the nineties in last century, main at that time research emphasis is the micro-and " entirely " of micro-fluidic chip analytic system, and the MEMS job operation of microchannel network.By 1994, U.S. Oak Ridge National Laboratory Ramsey etc. has delivered a series of papers on the working foundation of Manz, improved the sample injection method of chip capillary cataphoresis, improved its performance and operational, the business development of micro-fluidic chip analytic system is worth and begins to manifest.In recent years, the research about micro-fluidic chip progressively becomes focus in the world, and is quickening to microminiaturization and intelligent development.

Main analytical approach to chip in the present stage micro-total analysis system is a laser inductive fluorescence method, this method need position the microchannel network on the chip, and current micro-total analysis system all adopts basically that manual locator meams, this locator meams have that bearing accuracy is low, time and effort consuming and can't finish shortcoming such as track and localization.

Summary of the invention

The objective of the invention is to overcome the deficiencies in the prior art, a kind of intelligent locating method towards micro-fluidic chip is provided, be applied in all kinds of micro-fluidic chip analytic systems, the micro-fluidic chip plane picture that the Charge Coupled Device (CCD) imageing sensor is collected generates adjacency list according to related algorithm, utilize the data in the adjacency list, micro-fluidic chip is carried out the intelligence location, and, the data in the adjacency list are revised dynamically according to the result who locatees.

To achieve these goals, the present invention has adopted following technical scheme:

A kind of intelligent locating method towards micro-fluidic chip is characterized in that, comprises the following steps:

A. use the Charge Coupled Device (CCD) imageing sensor gather in the micro-fluidic chip analytic system the chip plane picture, utilize morphologic filtering device and single sweep operation thinning algorithm that the chip planimetric map is carried out denoising and refinement, obtain the skeleton diagram of chip plane picture:

1) the chip plane picture that the Charge Coupled Device (CCD) imageing sensor is collected carries out binary conversion treatment;

2) utilize unlatching and closed procedure in the morphology to carry out denoising and edge-smoothing processing to the chip plane picture after the binaryzation;

3) the imagery exploitation single sweep operation thinning algorithm after the denoising is carried out refinement, obtain the skeleton diagram of chip plane picture;

B. according to the chip skeleton diagram, extract the interdependent node on the microchannel network in the micro-fluidic chip:

1) extraction of microchannel network upper extreme point: in the chip skeleton diagram, have and have only a pixel to exist in the 8-neighborhood of certain pixel, then this pixel is exactly an end points;

2) extraction of microchannel network overcrossing point: in the chip skeleton diagram, the pixel more than 3 or 3 is arranged in the 8-neighborhood of certain pixel, then this pixel is exactly the point of crossing;

3) extraction of flex point on the microchannel network:

Here the chip skeleton diagram is carried out thunder and step on conversion (Radon Transform), the central point of getting image is as initial point, and the x axle is parallel with the coboundary of image and pass through initial point, and the y axle is vertical mutually with the x axle; Is being to carry out projection on 180 directions of 0 °～179 ° of degree with it with x coordinate axis angle, transformation results on each angle is as a column vector, all result combinations can form one 700 * 180 transformation matrix together, find out the peak value in the transformation matrix, these peak value correspondences the straight line on the image;

Then, determine which bar straight line end points and the point of crossing obtained belong to respectively, if an end points and a point of crossing of being interconnected, the straight line under the end points does not pass through the point of crossing, the point of crossing belongs to many straight lines simultaneously, promptly has a flex point between this end points of decidable and the point of crossing; Then the intersection point of straight line under the end points and all straight lines under the point of crossing is all obtained, the intersection point that drops on end points and the point of crossing connecting line is flex point;

C. the generation of micro-fluidic chip planimetric map adjacency list:

All microchannel node locations and connected relation each other thereof according to obtaining above can generate a corresponding adjacency list; Each unit Xiang Youyi in the adjacency list tlv triple formed, and wherein, the central point of image is an initial point, the coordinate of preceding two expression nodes, and the 3rd expression node types, wherein, 0 expression end points, 1 expression point of crossing, 2 expression flex points; According to this adjacency list, system just can be intelligently to micro-fluidic chip track and localization continuously.

Above-mentioned intelligent locating method towards micro-fluidic chip, wherein, described micro-fluidic chip intelligence location is as follows with the feedback modifiers method of adjacency list:

1) when with the method for laser-induced fluorescence (LIF) micro-fluidic chip being analyzed, high-field electrode necessarily is added on two end points, so when following the tracks of detection, generating laser and photomultiplier also must be to move to another end points from an end points; Can be easy to obtain the position of the end points that need detect and end points according to adjacency list to the path between end points; Any two adjacent nodes comprise between end points, point of crossing or the flex point it all being straight-line segment in the path, so the path between any two ends point can be that a broken line is represented with one group of straight-line segment; Carry out track and localization to the path between two end points, will carry out track and localization to each the bar straight-line segment in the path exactly, and the coordinate of all pixels can be obtained on the straight-line segment by the node coordinate at line segment two ends: the coordinate of establishing two nodes is respectively (x ₁, y ₁) and (x ₂, y ₂), to determine on the straight-line segment that then the pixel number of coordinate is:

Wherein, the coordinate of i pixel is:

$([x_1+\frac{i}{\mathrm{num}}(x_2-x_1)],[y_1+\frac{i}{\mathrm{num}}(y_2-y_1)])$

Use this method, just can realize the track and localization of intelligence whole micro-fluidic chip;

2) according to the feedback result of track and localization adjacency list is carried out feedback modifiers: when the enterprising line trace of straight line section location, if do not receive corresponding fluorescence signal at certain pixel place photomultiplier, just in the 16-neighborhood that does not comprise the point on the straight-line segment of this pixel, search for:, then keep former pixel coordinate constant if i) do not have pixel can receive fluorescence signal; If ii) have only a pixel can receive fluorescence signal, then the coordinate with this pixel substitutes former pixel coordinate; If iii) there are two above pixels can receive fluorescence signal, then substitute former pixel coordinate with the coordinate along that nearest pixel of the former pixel of straight-line segment direction distance in these pixels.After intact to this straight-line segment track and localization, the while has also carried out once proofreading and correct to the coordinate of all pixels on the line segment, can utilize following two formula to revise the coordinate of node in the adjacency list then:

(x ₁', y ₁') and (x ₂', y ₂') will be as (x ₁, y ₁) and (x ₂, y ₂) the new coordinate of two nodes removes to revise adjacency list;

In the formula, (x ₀, y ₀) middle point coordinate after the expression straight-line segment is proofreaied and correct:

$(x_0=\left[\frac{\underset{i=1}{\overset{\mathrm{num}}{\mathrm{\Σ}}}{x\′}_i}{\mathrm{num}}\right],y_0=\left[\frac{\underset{i=1}{\overset{\mathrm{num}}{\mathrm{\Σ}}}{y\′}_i}{\mathrm{num}}\right])$

Wherein, (x ' _i, y ' _i) (i=1 ..., the coordinate of pixel on the straight-line segment after num) expression is proofreaied and correct.

Owing to adopted above-mentioned technical scheme, the present invention compared with prior art has following advantage and good effect:

Because the present invention is towards the intelligent locating method of micro-fluidic chip, be based on associated picture treatment technology design, have the full-automation, precision height, speed of location fast and can track and localization etc. characteristics, can carry out the improvement of automated analysis to all kinds of micro-fluidic chip analytic systems.

The node that this method utilizes morphology algorithm, thinning algorithm and thunder to step on microchannel on the micro-fluidic chip planimetric map that correlated image processing methods such as conversion collect the Charge Coupled Device (CCD) imageing sensor extracts, generate adjacency list, according to adjacency list micro-fluidic chip is carried out the track and localization of intelligence, and can carry out feedback modifiers according to positioning result to adjacency list by relevant feedback algorithm.Thereby the existing bearing accuracy of micro-total analysis system of locator meams that has overcome present employing craft effectively is low, time and effort consuming and can't finish defectives such as track and localization.

Description of drawings

By following examples and in conjunction with the description of its accompanying drawing, can further understand purpose, specific structural features and the advantage of its invention.In the accompanying drawing,

Fig. 1 is the system chart of micro-fluidic chip intelligent positioning system of the present invention;

Fig. 2 is a micro-fluidic chip plane picture pre-service synoptic diagram of the present invention;

Fig. 3 is the characteristic pattern of end points of the present invention, point of crossing, flex point;

Fig. 4 is an overall intelligence localization method process flow diagram of the present invention;

Table 1 is a micro-fluidic chip adjacency list of the present invention.

Among Fig. 1,

1. Charge Coupled Device (CCD) imageing sensor; 2. image pick-up card; 3. computing machine; 4.RS232-RS485 converter; 5. stepping motor control platform; 6. chip placement platform; 7. generating laser; 8. high pressure produces equipment; 9. photomultiplier; 10.PCI bus; 11.RS232; 12.RS485.

Among Fig. 2,

A. the micro-fluidic chip planimetric map after the binaryzation;

B. micro-fluidic chip planimetric map after the effect of morphology noise filtering device;

C. pass through the image thinning result after the effect of morphology noise filtering device;

D. without the image thinning result after the effect of morphology noise filtering device.

Among Fig. 3, the A. end points; B. point of crossing; C. flex point.

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 and limit the scope of the invention.Should be understood that in addition those skilled in the art can make various changes or modifications 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.

A kind of intelligent locating method towards micro-fluidic chip of the present invention comprises the following steps:

1. current micro-fluidic chip plane picture is carried out pre-service:

A. use the chip plane picture in the Charge Coupled Device (CCD) imageing sensor collection micro-fluidic chip analytic system, utilize morphologic filtering device and single sweep operation thinning algorithm that the chip planimetric map is carried out denoising and refinement, obtain the skeleton diagram of chip plane picture:

The chip plane picture that the Charge Coupled Device (CCD) imageing sensor is collected carries out binary conversion treatment;

Utilize unlatching and closed procedure in the morphology to carry out denoising and edge-smoothing processing to the chip plane picture after the binaryzation;

Image after the denoising is carried out refinement with the single sweep operation thinning algorithm, obtain the skeleton diagram of chip plane picture.

B. according to the chip skeleton diagram, extract the interdependent node on the microchannel network in the micro-fluidic chip:

The extraction of microchannel network upper extreme point: in the chip skeleton diagram, have and have only a pixel to exist in the 8-neighborhood of certain picture element, then this pixel is exactly an end points.

The extraction of microchannel network overcrossing point: in the chip skeleton diagram, the pixel more than 3 or 3 is arranged in the 8-neighborhood of certain pixel, then this pixel is exactly the point of crossing.

The extraction of flex point on the microchannel network:

Here the chip skeleton diagram is carried out thunder and step on conversion, the central point of getting image is as initial point, and the x axle is parallel with the coboundary of image and pass through initial point, and the y axle is vertical mutually with the x axle.Is being to carry out projection on 180 directions of 0 °～179 ° with it with x coordinate axis angle, transformation results on each angle is as a column vector, all result combinations can form one 700 * 180 transformation matrix together, find out the peak value in the transformation matrix, these peak value correspondences the straight line on the image.Draw linear positions all on the skeleton line by this method.

Can determine then which bar straight line end points and the point of crossing obtained above belong to respectively, if an end points and a point of crossing of being interconnected, straight line under the end points can be affirmed to have a flex point between this end points and the point of crossing not by point of crossing (point of crossing belongs to many straight lines simultaneously) so.Then the intersection point of straight line under the end points and all straight lines under the point of crossing is all obtained, that intersection point that drops on end points and the point of crossing connecting line is exactly a flex point.By this method, all flex points on the skeleton line can be obtained.

C. the generation of micro-fluidic chip planimetric map adjacency list:

All microchannel node locations and connected relation each other thereof according to obtaining above can generate a corresponding adjacency list.Each unit Xiang Youyi in the adjacency list tlv triple formed, and wherein, the central point of image is an initial point, the coordinate of preceding two expression nodes, and the 3rd expression node types, wherein, 0 expression end points, 1 expression point of crossing, 2 expression flex points.According to this adjacency list, system just can be intelligently to the continuous track and localization of micro-fluidic chip.

2. the feedback modifiers method of micro-fluidic chip intelligence location and adjacency list is as follows:

1) when with the method for laser-induced fluorescence (LIF) micro-fluidic chip being analyzed, high-field electrode necessarily is added on two end points, so when following the tracks of detection, generating laser and photomultiplier also must be to move to another end points from an end points.Can be easy to obtain the position of the end points that need detect and end points according to adjacency list to the path between end points.Referring to Fig. 3, any two adjacent nodes comprise between end points, point of crossing or the flex point it all being straight-line segment in the path, so the path between any two ends point can be that a broken line is represented with one group of straight-line segment.To carry out track and localization to the path between two end points, will carry out track and localization to each the bar straight-line segment in the path exactly in fact, and the coordinate of all pixels can be obtained on the straight-line segment: the coordinate branch of establishing two nodes by the node coordinate at line segment two ends

Wei (x ₁, y ₁) and (x ₂, y ₂), to determine on the straight-line segment that then the pixel number of coordinate is:

Wherein the coordinate of i pixel is:

$([x_1+\frac{i}{\mathrm{num}}(x_2-x_1)],[{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="C200610118447E00151.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+\frac{i}{\mathrm{num}}(y_2-y_1)])$

Use this method, just can realize the track and localization of intelligence whole micro-fluidic chip.

2) consider with the Charge Coupled Device (CCD) imageing sensor and gather the micro-fluidic chip planimetric map and the subsequent image processing process may produce some errors, make final adjacent map not necessarily very accurate, can adopt following method adjacency list to be carried out feedback modifiers: when the enterprising line trace of straight line section location according to the feedback result of track and localization, if do not receive corresponding fluorescence signal at certain pixel place photomultiplier, just in the 16-neighborhood that does not comprise the point on the straight-line segment of this pixel, search for:, then keep former pixel coordinate constant if i) do not have pixel can receive fluorescence signal; If ii) have only a pixel can receive fluorescence signal, then the coordinate with this pixel substitutes former pixel coordinate; If iii) there are two above pixels can receive fluorescence signal, then substitute former pixel coordinate with the coordinate along that nearest pixel of the former pixel of straight-line segment direction distance in these pixels.After intact to this straight-line segment track and localization, the while has also carried out once proofreading and correct to the coordinate of all pixels on the line segment, can utilize following two formula to revise the coordinate of node in the adjacency list then:

$(x_1\&prime;=[x_0-\frac{x_2-x_1}{2}],{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="C200610118447E00151.png"\; state="\{\{state\}\}">y</figure-callout>}}_1\&prime;=[y_0-\frac{y_2-{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="C200610118447E00151.png"\; state="\{\{state\}\}">y</figure-callout>}}_1}{2}])$

$(x_2\&prime;=[x_0+\frac{x_2-x_1}{2}],{\mathrm{<figure-callout\; id="2"\; label="y"\; filenames="C200610118447E00151.png"\; state="\{\{state\}\}">y</figure-callout>}}_2\&prime;=[y_0+\frac{y_2-{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="C200610118447E00151.png"\; state="\{\{state\}\}">y</figure-callout>}}_1}{2}])$

(x ₁', y ₁') and (x ₂', y ₂') will be as (x ₁, y ₁) and (x ₂, y ₂) the new coordinate of two nodes removes to revise adjacency list.

In the formula, (x ₀, y ₀) middle point coordinate after the expression straight-line segment is proofreaied and correct:

$(x_0=\left[\frac{\underset{i=1}{\overset{\mathrm{num}}{\mathrm{\&Sigma;}}}{x\&prime;}_i}{\mathrm{num}}\right],y_0=\left[\frac{\underset{i=1}{\overset{\mathrm{num}}{\mathrm{\&Sigma;}}}{y\&prime;}_i}{\mathrm{num}}\right])$

Wherein, (x ' _i, y ' _i) (i=1 ..., the coordinate of pixel on the straight-line segment after num) expression is proofreaied and correct.

Generating relative theory about adjacency list of the present invention is summarized as follows:

The single sweep operation thinning algorithm can become image thinning single pixel wide well, does not destroy the connectedness of original image, keeps the topological structure of image preferably.To show apparent in view feature in the skeleton diagram of node in the image (end points, point of crossing and flex point) after refinement, can easily they be come out according to certain Rule Extraction.

It is exactly that original image is transformed to its projective representation in all angles that thunder is stepped on conversion.The projection of image is meant the line integral of image on a certain direction, the summation that adds up on this direction just for digital picture.The mathematical notation that thunder is stepped on conversion is: image f (x, y) projection of R is defined as on unspecified angle θ:

$R_{\mathrm{\&theta;}}(x\&prime;)=\underset{-\&infin;}{\overset{\&infin;}{\&Integral;}}f(x\&prime;\cos \mathrm{\&theta;}-y\&prime;\sin \mathrm{\&theta;},x\&prime;\sin \mathrm{\&theta;}-y\&prime;\cos \mathrm{\&theta;})\mathrm{dy}\&prime;$

Wherein,

$\left(\begin{array}{c}x\&prime;\\ y\&prime;\end{array}\right)=\left(\begin{array}{cc}\cos \mathrm{\&theta;}& \sin \mathrm{\&theta;}\\ -\sin \mathrm{\&theta;}& \cos \mathrm{\&theta;}\end{array}\right)\left(\begin{array}{c}x\\ y\end{array}\right)$

Thunder is stepped on the straight line that is easy to find out after the conversion in the skeleton image, thereby makes things convenient for the extraction of flex point.

Principle of work about the intelligent locating method towards micro-fluidic chip of the present invention

The composition of micro-fluidic chip intelligent positioning system mainly is made up of stepping motor control platform, Charge Coupled Device (CCD) imageing sensor, image pick-up card, generating laser, photomultiplier, high pressure generation equipment and computing machine as shown in Figure 1.On the basis of existing micro-fluidic chip analytic system, increased Charge Coupled Device (CCD) imageing sensor image acquisition mechanism, be used for the planimetric map of camera system micro-fluidic chip, and be sent to computing machine by image pick-up card.Image pick-up card links to each other by pci bus with computing machine.Generating laser, photomultiplier, high pressure produce equipment and stepping motor control platform all is to be connected with computing machine through the RS232/RS485 converter by serial port RS485.

Simultaneously, the present invention is based on following understanding:

The topological structure of the microchannel on the micro-fluidic chip in the microchannel network all is straight lines, and not having topological structure is the microchannel of curve or other non-rectilinear classification line style.No matter micro-fluidic chip with which kind of angle is placed in the system, all thinks the positive center of the micro-fluidic chip planimetric map that its initial point collects at the Charge Coupled Device (CCD) imageing sensor.

A kind of intelligent locating method towards micro-fluidic chip of the present invention comprises the following steps:

1. current micro-fluidic chip plane picture is carried out pre-service, referring to Fig. 2:

A. use the Charge Coupled Device (CCD) imageing sensor gather in the micro-fluidic chip analytic system the chip plane picture, utilize morphologic filtering device and single sweep operation thinning algorithm that the chip planimetric map is carried out denoising and refinement, obtain the skeleton diagram of chip plane picture:

The chip plane picture that the Charge Coupled Device (CCD) imageing sensor is collected carries out binary conversion treatment;

Utilize unlatching and closed procedure in the morphology to carry out denoising and edge-smoothing processing to the chip plane picture after the binaryzation;

Image after the denoising is carried out refinement with the single sweep operation thinning algorithm, obtain the skeleton diagram of chip plane picture.

B. according to the chip skeleton diagram, extract the interdependent node on the microchannel network in the micro-fluidic chip, referring to Fig. 3:

The extraction of microchannel network upper extreme point: in the chip skeleton diagram, have and have only a pixel to exist in the 8-neighborhood of certain pixel, then this pixel is exactly an end points.

The extraction of microchannel network overcrossing point: in the chip skeleton diagram, the pixel more than 3 or 3 is arranged in the 8-neighborhood of certain pixel, then this pixel is exactly the point of crossing.

The extraction of flex point on the microchannel network:

Here the chip skeleton diagram is carried out thunder and step on conversion, the central point of getting image is as initial point, and the x axle is parallel with the coboundary of image and pass through initial point, and the y axle is vertical mutually with the x axle.Is being to carry out projection on 180 directions of 0 °～179 ° with it with x coordinate axis angle, transformation results on each angle is as a column vector, all result combinations can form one 700 * 180 transformation matrix together, find out the peak value in the transformation matrix, these peak value correspondences the straight line on the image.Draw linear positions all on the skeleton line by this method.

Can determine then which bar straight line end points and the point of crossing obtained above belong to respectively, if an end points and a point of crossing of being interconnected, straight line under the end points can be affirmed to have a flex point between this end points and the point of crossing not by point of crossing (point of crossing belongs to many straight lines simultaneously) so.Then the intersection point of straight line under the end points and all straight lines under the point of crossing is all obtained, that intersection point that drops on end points and the point of crossing connecting line is exactly a flex point.By this method, all flex points on the skeleton line can be obtained.

C. the generation of micro-fluidic chip planimetric map adjacency list, referring to table 1:

All microchannel node locations and connected relation each other thereof according to obtaining above can generate a corresponding adjacency list.Each unit Xiang Youyi in the adjacency list tlv triple formed, and wherein, the central point of image is an initial point, the coordinate of preceding two expression nodes, and the 3rd expression node types, wherein, 0 expression end points, 1 expression point of crossing, 2 expression flex points.According to this adjacency list, system just can be intelligently to the continuous track and localization of micro-fluidic chip.

2. the feedback modifiers method of micro-fluidic chip intelligence location and adjacency list is as follows:

1) when with the method for laser-induced fluorescence (LIF) micro-fluidic chip being analyzed, high-field electrode necessarily is added on two end points, so when following the tracks of detection, generating laser and photomultiplier also must be to move to another end points from an end points.Can be easy to obtain the position of the end points that need detect and end points according to adjacency list to the path between end points.All be straight-line segment between any two adjacent nodes (end points, point of crossing or flex point) in the path, so the path between any two ends point can be represented with one group of straight-line segment (i.e. broken line).To carry out track and localization to the path between two end points, will carry out track and localization to each the bar straight-line segment in the path exactly in fact, and the coordinate of all pixels can be obtained on the straight-line segment by the node coordinate at line segment two ends: the coordinate of establishing two nodes is respectively (x ₁, y ₁) and (x ₂, y ₂), to determine on the straight-line segment that then the pixel number of coordinate is:

Wherein the coordinate of i pixel is:

$([x_1+\frac{i}{\mathrm{num}}(x_2-x_1)],[{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="C200610118447E00151.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+\frac{i}{\mathrm{num}}(y_2-y_1)])$

Use this method, just can realize the track and localization of intelligence whole micro-fluidic chip.

2) consider with the Charge Coupled Device (CCD) imageing sensor and gather the micro-fluidic chip planimetric map and the subsequent image processing process may produce some errors, make final adjacent map not necessarily very accurate, can adopt following method adjacency list to be carried out feedback modifiers: when the enterprising line trace of straight line section location according to the feedback result of track and localization, if do not receive corresponding fluorescence signal at certain pixel place photomultiplier, just search for:, then keep former pixel coordinate constant if i) do not have pixel can receive fluorescence signal in not comprising in the some 16-neighborhood on the straight-line segment of this pixel; If ii) have only a pixel can receive fluorescence signal, then the coordinate with this pixel substitutes former pixel coordinate; If iii) there are two above pixels can receive fluorescence signal, then substitute former pixel coordinate with the coordinate along that nearest pixel of the former pixel of straight-line segment direction distance in these pixels.After intact to this straight-line segment track and localization, the while has also carried out once proofreading and correct to the coordinate of all pixels on the line segment, can utilize following two formula to revise the coordinate of node in the adjacency list then:

$(x_1\&prime;=[x_0-\frac{x_2-x_1}{2}],{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="C200610118447E00151.png"\; state="\{\{state\}\}">y</figure-callout>}}_1\&prime;=[y_0-\frac{y_2-{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="C200610118447E00151.png"\; state="\{\{state\}\}">y</figure-callout>}}_1}{2}])$

$(x_2\&prime;=[x_0+\frac{x_2-x_1}{2}],{\mathrm{<figure-callout\; id="2"\; label="y"\; filenames="C200610118447E00151.png"\; state="\{\{state\}\}">y</figure-callout>}}_2\&prime;=[y_0+\frac{y_2-{\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="C200610118447E00151.png"\; state="\{\{state\}\}">y</figure-callout>}}_1}{2}])$

(x ₁', y ₁') and (x ₂', y ₂') will be as (x ₁, y ₁) and (x ₂, y ₂) the new coordinate of two nodes removes to revise adjacency list.

In the formula, (x ₀, y ₀) middle point coordinate after the expression straight-line segment is proofreaied and correct:

$(x_0=\left[\frac{\underset{i=1}{\overset{\mathrm{num}}{\mathrm{\&Sigma;}}}{x\&prime;}_i}{\mathrm{num}}\right],y_0=\left[\frac{\underset{i=1}{\overset{\mathrm{num}}{\mathrm{\&Sigma;}}}{y\&prime;}_i}{\mathrm{num}}\right])$

Wherein, (x ' _i, y ' _i) (i=1 ..., the coordinate of pixel on the straight-line segment after num) expression is proofreaied and correct.

Comprehensive above-mentioned steps, overall flow figure of the present invention as shown in Figure 4.

Table one:

Table 1

Node 1	Node 2	Node 3	Node 4	Node 5	Node 6	Node 7
							(-145，138， 0)	(62，128， 0)	(-264， 111，0)	(162， 100，0)	(269，12， 0)	(-83， -94，0)	(37，-124， 0)
(-88，64， 2)	(7，49， 2)	(-210，53， 2)	(165，6， 1)	(165，6， 1)	(-85，-2， 1)	(77，-42， 2)

Node 8	Node 9	Node 10	Node 11	Node 12	Node 13	Node 14
							(-261， -110，0)	(-304， -5，0)	(-206，-6， 1)	(-85，-2， 1)	(10，3，1)	(165，6，1)	(74，2， 1)
(-201， -55，2)	(-206， -6，1)	(-85，-2， 1)	(10，3，1)	(74，2，1)	(162，100， 0)	(165，6， 1)
									(-210，53， 2)	(-83，-94， 0)	(-85，-2， 1)	(269，12， 0)	(10，3， 1)
		(-201， -55，2)	(-88，64， 2)	(7，49，2)	(166，-84， 0)	(77， -42，2)
									(-304，-5， 0)	(-206，-6， 1)		(74，2，1)

Node 15	Node 16	Node 17	Node 18	Node 19	Node 20
						(166， -84，0)	(-210，53， 2)	(-88，64， 2)	(7，49， 2)	(-201，-55， 2)	(77，-42， 2)
(165，6， 1)	(-264， 111，0)	(-145， 138，0)	(62，128， 0)	(-261， -110，0)	(37， -124，0)
							(-206，-6， 1)	(-85，-2， 1)	(10，3， 1)	(-206，-6， 1)	(74，2，1)

Claims (2)

1. the intelligent locating method towards micro-fluidic chip is characterized in that, comprises the following steps:

A. use the Charge Coupled Device (CCD) imageing sensor gather in the micro-fluidic chip analytic system the chip plane picture, utilize morphologic filtering device and single sweep operation thinning algorithm that the chip planimetric map is carried out denoising and refinement, obtain the skeleton diagram of chip plane picture:

1) the chip plane picture that the Charge Coupled Device (CCD) imageing sensor is collected carries out binary conversion treatment;

2) utilize unlatching and closed procedure in the morphology to carry out denoising and edge-smoothing processing to the chip plane picture after the binaryzation;

3) the imagery exploitation single sweep operation thinning algorithm after the denoising is carried out refinement, obtain the skeleton diagram of chip plane picture;

B. according to the chip skeleton diagram, extract the interdependent node on the microchannel network in the micro-fluidic chip:

1) extraction of microchannel network upper extreme point: in the chip skeleton diagram, have and have only a pixel to exist in the 8-neighborhood of certain pixel, then this pixel is exactly an end points;

2) extraction of microchannel network overcrossing point: in the chip skeleton diagram, the pixel more than 3 or 3 is arranged in the 8-neighborhood of certain pixel, then this pixel is exactly the point of crossing;

3) extraction of flex point on the microchannel network:

Here the chip skeleton diagram is carried out thunder and step on conversion, the central point of getting image is as initial point, and the x axle is parallel with the coboundary of image and pass through initial point, and the y axle is vertical mutually with the x axle; Is being to carry out projection on 180 directions of 0 °～179 ° of degree with it with x coordinate axis angle, transformation results on each angle is as a column vector, all result combinations can form one 700 * 180 transformation matrix together, find out the peak value in the transformation matrix, these peak value correspondences the straight line on the image;

Then, determine which bar straight line end points and the point of crossing obtained belong to respectively, if an end points and a point of crossing of being interconnected, the straight line under the end points does not pass through the point of crossing, the point of crossing belongs to many straight lines simultaneously, promptly has a flex point between this end points of decidable and the point of crossing; Then the intersection point of straight line under the end points and all straight lines under the point of crossing is all obtained, the intersection point that drops on end points and the point of crossing connecting line is flex point;

C. the generation of micro-fluidic chip planimetric map adjacency list:

All microchannel node locations and connected relation each other thereof according to obtaining above can generate a corresponding adjacency list; Each unit Xiang Youyi in the adjacency list tlv triple formed, and wherein, the central point of image is an initial point, the coordinate of preceding two expression nodes, and the 3rd expression node types, wherein, 0 expression end points, 1 expression point of crossing, 2 expression flex points; According to this adjacency list, system just can be intelligently to micro-fluidic chip track and localization continuously.

2, the intelligent locating method towards micro-fluidic chip according to claim 1 is characterized in that, described micro-fluidic chip intelligence location is as follows with the feedback modifiers method of adjacency list:

1) when with the method for laser-induced fluorescence (LIF) micro-fluidic chip being analyzed, high-field electrode necessarily is added on two end points, so when following the tracks of detection, generating laser and photomultiplier also must be to move to another end points from an end points; Can be easy to obtain the position of the end points that need detect and end points according to adjacency list to the path between end points; Any two adjacent nodes comprise between end points, point of crossing or the flex point it all being straight-line segment in the path, so the path between any two ends point can be that a broken line is represented with one group of straight-line segment; Carry out track and localization to the path between two end points, will carry out track and localization to each the bar straight-line segment in the path exactly, and the coordinate of all pixels can be obtained on the straight-line segment by the node coordinate at line segment two ends: the coordinate of establishing two nodes is respectively (x ₁, y ₁) and (x ₂, y ₂), to determine on the straight-line segment that then the pixel number of coordinate is:

Wherein, the coordinate of i pixel is:

$([x_1+\frac{i}{\mathrm{num}}(x_2-x_1)],[y_1+\frac{i}{\mathrm{num}}(y_2-y_1)])$

Use this method, just can realize the track and localization of intelligence whole micro-fluidic chip;

2) according to the feedback result of track and localization adjacency list is carried out feedback modifiers: when the enterprising line trace of straight line section location, if do not receive corresponding fluorescence signal at certain pixel place photomultiplier, just in the 16-neighborhood that does not comprise the point on the straight-line segment of this pixel, search for:, then keep former pixel coordinate constant if i) do not have pixel can receive fluorescence signal; If ii) have only a pixel can receive fluorescence signal, then the coordinate with this pixel substitutes former pixel coordinate; If iii) there are two above pixels can receive fluorescence signal, then substitute former pixel coordinate with the coordinate along that nearest pixel of the former pixel of straight-line segment direction distance in these pixels, after intact to this straight-line segment track and localization, while has also carried out once proofreading and correct to the coordinate of all pixels on the line segment, can utilize following two formula to revise the coordinate of node in the adjacency list then:

$(x_1\&prime;=[x_0-\frac{x_2-x_1}{2}],y_1\&prime;=[y_0-\frac{y_2-y_1}{2}])$

$(x_2\&prime;=[x_0+\frac{x_2-x_1}{2}],y_2\&prime;=[y_0+\frac{y_2-y_1}{2}])$

(x ₁', y ₁') and (x ₂', y ₂') will be as (x ₁, y ₁) and (x ₂, y ₂) the new coordinate of two nodes removes to revise adjacency list;

In the formula, (x ₀, y ₀) middle point coordinate after the expression straight-line segment is proofreaied and correct:

$(x_0=\left[\frac{\underset{i=1}{\overset{\mathrm{num}}{\mathrm{\&Sigma;}}}{x\&prime;}_i}{\mathrm{num}}\right],y_0=\left[\frac{\underset{i=1}{\overset{\mathrm{num}}{\mathrm{\&Sigma;}}}{y\&prime;}_i}{\mathrm{num}}\right])$

Wherein, (x ' _i, y ' _i) (i=1 ..., the coordinate of pixel on the straight-line segment after num) expression is proofreaied and correct.

Images