CN111311631A - Fluid velocity detection method, device and equipment in microfluidic chip - Google Patents

Abstract

The invention belongs to the technical field of visual detection, and discloses a method, device and equipment for detecting fluid velocity in a microfluidic chip. The method includes: acquiring a flow velocity video corresponding to the fluid to be detected, performing image frame extraction on the flow velocity video to obtain a set of image frames; acquiring the fluid motion time corresponding to each frame of the image in the set of image frames; vector data corresponding to the fluid to be detected in the frame image, and the velocity of the fluid to be detected is calculated according to the fluid movement time and the vector data. Through the above method, the flowing liquid can be effectively tracked, so as to accurately obtain the speed and direction of the liquid movement.

Description

Fluid velocity detection method, device and equipment in microfluidic chip

technical field

The invention relates to the technical field of visual inspection, and in particular to a method, device and equipment for detecting fluid velocity in a microfluidic chip.

Background technique

In recent years, the advantages of microfluidic technology, such as low sample consumption, short analysis time, high throughput, and compatibility with most experimental samples, have enabled the miniaturization and high efficiency of current research tools, thus greatly promoting the The development of biology, chemical industry, agricultural food and other fields.

Therefore, in order to accelerate the development of these fields, it is becoming more and more urgent to detect the movement of fluids in microchannels in real time and accurately. By detecting the flow rate state in the microchannel, the viscosity of the fluid and the shear force of the microfluid flowing in the chip can be easily obtained. This information can provide great help to the majority of researchers. However, in the prior art, the microfluidic chip pipeline size is very small, and the channel width is often between tens of micrometers to hundreds of micrometers, so that the conventional measurement methods cannot be used normally, and the quantification of the fluid motion information in the microchannel is greatly affected. degree of restriction.

The above content is only used to assist the understanding of the technical solutions of the present invention, and does not mean that the above content is the prior art.

SUMMARY OF THE INVENTION

The main purpose of the present invention is to provide a fluid velocity detection method, device and equipment in a microfluidic chip, which aims to solve the technical problem of how to accurately obtain the fluid velocity in the microfluidic chip.

To achieve the above object, the invention provides a fluid velocity detection method in a microfluidic chip, the method comprising the following steps:

The microfluidic chip contains the fluid to be detected, and the method includes:

obtaining the flow velocity video corresponding to the fluid to be detected, and extracting image frames from the flow velocity video to obtain an image frame set;

Obtain the fluid motion time corresponding to each frame of image in the image frame set;

The vector data corresponding to the fluid to be detected in each frame of image is acquired, and the velocity of the fluid to be detected is calculated according to the movement time of the fluid and the vector data.

Preferably, the step of acquiring the flow velocity video corresponding to the fluid to be detected, extracting image frames from the flow velocity video, and obtaining a set of image frames, includes:

obtaining the flow velocity video corresponding to the fluid to be detected, and performing image frame extraction on the flow velocity video to obtain an initial image frame;

An image frame to be processed is extracted from the initial image frame according to a preset ROI area, and an image frame set is constructed according to the image frame to be processed.

Preferably, the step of acquiring vector data corresponding to the fluid to be detected in each frame of image includes:

Select a target image frame from the set of image frames;

Obtain a corresponding target image according to the target image frame;

Converting the target image into a grayscale image, and performing Gaussian filtering on the grayscale image to obtain a smooth image corresponding to the fluid to be detected;

Calculate the optical flow field corresponding to the fluid to be detected according to the smoothed image;

Determine the optical flow field detection area corresponding to the fluid to be detected according to the optical flow field;

Dividing the optical flow field detection area to obtain a vector field image;

According to the vector field image, vector data corresponding to the fluid to be detected is calculated by a preset optical flow algorithm.

Preferably, the step of performing Gaussian filtering on the target image to obtain a vector field image includes:

Preferably, the step of calculating the vector data corresponding to the fluid to be detected by using a preset optical flow algorithm according to the vector field image includes:

Determine the coordinate value corresponding to the fluid to be detected according to the smooth image;

Determine the pixel value corresponding to the fluid to be detected according to the vector field image;

According to the coordinate value, the effective weight corresponding to the fluid to be detected is obtained through a normal distribution algorithm;

Derivating the pixel value to obtain a directional gradient value corresponding to the fluid to be detected;

The vector data corresponding to the fluid to be detected is calculated according to the effective weight and the directional gradient value.

Preferably, the step of obtaining the effective weight corresponding to the fluid to be detected through a normal distribution algorithm according to the coordinate value includes:

According to the coordinate value, calculate the initial weight corresponding to the fluid to be detected by a normal distribution algorithm;

The initial weights are screened to obtain effective weights corresponding to the fluid to be detected.

Preferably, the step of screening the initial weight to obtain an effective weight corresponding to the fluid to be detected includes:

judging whether the initial weight falls within a preset weight threshold range;

If the initial weight falls within the preset weight threshold range, the initial weight is used as an effective weight corresponding to the fluid to be detected.

In addition, in order to achieve the above purpose, the present invention also provides a fluid velocity detection device in a microfluidic chip, the microfluidic chip contains the fluid to be detected, and the device includes: an extraction module for obtaining the to-be-detected fluid. The flow velocity video corresponding to the fluid, extracting image frames from the flow velocity video to obtain an image frame set;

an acquisition module, used for acquiring the fluid motion time corresponding to each frame of image in the image frame set;

A calculation module, configured to acquire vector data corresponding to the fluid to be detected in each frame of the image, and calculate the velocity of the fluid to be detected according to the fluid movement time and the vector data.

Preferably, the extraction module is further configured to obtain a flow velocity video corresponding to the fluid to be detected, and perform image frame extraction on the flow velocity video to obtain an initial image frame;

The extraction module is further configured to extract image frames to be processed from the initial image frames according to a preset ROI area, and construct an image frame set according to the image frames to be processed.

Preferably, the computing module is further configured to select a target image frame from the image frame set;

The computing module is further configured to obtain a corresponding target image according to the target image frame;

The computing module is further configured to perform Gaussian filtering processing on the target image to obtain a vector field image;

The calculation module is further configured to calculate vector data corresponding to the fluid to be detected through a preset optical flow algorithm according to the vector field image.

In addition, in order to achieve the above purpose, the present invention also proposes a fluid velocity detection device in a microfluidic chip, the device includes: a CCD camera, a microscope, a backlight board, a chip fixing fixture, a vibration isolation platform, a display, and a servo injection controller , a syringe pump, a memory, a processor, and a fluid velocity detection program in a microfluidic chip stored on the memory and operable on the processor, the fluid velocity detection program in the microfluidic chip being configured to achieve the above The steps of the fluid velocity detection method in the microfluidic chip described in any one of the texts.

The present invention first obtains the flow velocity video corresponding to the fluid to be detected, extracts image frames from the flow velocity video, combines the image frames to obtain an image frame set, and then obtains the corresponding image frame of each frame in the image frame set. The fluid movement time is obtained, and then the vector data corresponding to the fluid to be detected in each frame of image is obtained, and the velocity of the fluid to be detected is calculated according to the fluid movement time and the vector data. Because the present invention extracts the image frame containing fluid motion according to the flow velocity video, so that the user can effectively track the fluid motion state, then obtain the fluid motion time corresponding to the image frame, and convert, process and calculate the image frame. The vector data corresponding to each frame of image, and finally the fluid velocity is determined according to the fluid movement time and the vector data, which can ensure that the calculation result of the fluid velocity has a high accuracy, and at the same time provides an effective reference for predicting the direction and velocity of the fluid movement to be detected. .

Description of drawings

1 is a schematic structural diagram of a fluid velocity detection device in a microfluidic chip of a hardware operating environment involved in an embodiment of the present invention;

2 is a schematic flowchart of the first embodiment of the fluid velocity detection method in the microfluidic chip of the present invention;

3 is a schematic flowchart of a second embodiment of a fluid velocity detection method in a microfluidic chip of the present invention;

FIG. 4 is a structural block diagram of the first embodiment of the fluid velocity detection device in the microfluidic chip of the present invention.

The realization, functional characteristics and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments.

Detailed ways

It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.

Referring to FIG. 1 , FIG. 1 is a schematic structural diagram of a fluid velocity detection device in a microfluidic chip of a hardware operating environment involved in an embodiment of the present invention.

As shown in FIG. 1, the fluid velocity detection equipment in the microfluidic chip may include: CCD camera 1001, microscope 1002, backlight plate 1003, chip fixing fixture 1004, vibration isolation platform 1005, display 1006, servo injection controller 1007, injection Pump 1008 , start switch 1009 , power switch 1010 , calculator 1011 and heat sink 1012 . Among them, the CCD camera 1001 is a part of the image acquisition device. The microscope 1002 magnifies the fluid movement in the microchannel to improve image quality. The backlight plate 1003 provides stable lighting conditions, which is beneficial to the realization of the optical flow method. The chip fixing jig 1004 and the vibration isolation platform 1005 play a role of vibration isolation to improve the stability of detection. The display 1006 displays a moving video of the fluid to be detected, and displays an image converted from the moving video. Servo injector controller 1007 and injector pump 1008 are the pumping devices of the present invention to power the liquid.

The calculator 1011 includes a high-speed random access memory (Random Access Memory, RAM) memory, and may also be a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. The calculator 1011 may optionally also include a processor-independent storage device.

Those skilled in the art can understand that the structure shown in FIG. 1 does not constitute a limitation on the fluid velocity detection device in the microfluidic chip, and may include more or less components than the one shown, or combine some components, or Different component arrangements.

As shown in FIG. 1 , the calculator 1011 as a storage medium may include an operating system, a processor, a storage device, and a fluid velocity detection program in a microfluidic chip.

In the fluid velocity detection device in the microfluidic chip shown in FIG. 1 , the CCD camera 1001 , the microscope 1002 , the backlight plate 1003 , the chip fixing fixture 1004 , and the vibration isolation platform 1005 in the fluid velocity detection device in the microfluidic chip of the present invention , display 1006, servo injection controller 1007, injection pump 1008, start switch 1009, power switch 1010, calculator 1011 and heat sink 1012 can be set in the fluid velocity detection device in the microfluidic chip, in the microfluidic chip The fluid velocity detection device passes the fluid velocity detection program in the microfluidic chip stored in the calculator 1011, and executes the fluid velocity detection method in the microfluidic chip provided by the embodiment of the present invention.

An embodiment of the present invention provides a method for detecting fluid velocity in a microfluidic chip. Referring to FIG. 2 , FIG. 2 is a schematic flowchart of a first embodiment of a method for detecting fluid velocity in a microfluidic chip of the present invention.

In this embodiment, the fluid velocity detection method in the microfluidic chip includes the following steps:

Step S10: Obtain a flow velocity video corresponding to the fluid to be detected, and perform image frame extraction on the flow velocity video to obtain an image frame set.

It should be noted that the execution body of the method in this embodiment is a calculator, that is, a calculation method that can receive the flow velocity video corresponding to the fluid to be detected, process the flow velocity video to obtain vector data, and then calculate the vector data. device.

In this solution, the fluid to be detected is a flowable liquid, specifically ink, or a liquid with tracer particles added, or micro-droplets, etc. The microfluidic chip is a kind of microfluidic chip selected by the user The chip used in the test, the microfluidic chip adopts a PIE-14-007 hybrid chip, and another liquid or chip can also be used in the test, which is not limited in this embodiment.

The step of acquiring and acquiring the flow velocity video corresponding to the fluid to be detected, performing image frame extraction on the flow velocity video, and obtaining a set of image frames, includes acquiring the flow velocity video corresponding to the fluid to be detected, and performing image processing on the flow velocity video. Frame extraction, obtaining an initial image frame, extracting a to-be-processed image frame from the initial image frame according to a preset ROI area, and constructing an image frame set according to the to-be-processed image frame.

Wherein, the flow velocity video is the video of the fluid motion detection in the microfluidic chip under a given pump speed provided by the user, and then the flow velocity video is extracted, and the initial image frame of the flow velocity video is obtained, and then according to The user-defined area is to extract the image frame to be processed from the initial image frame. There may be blank frames corresponding to the fluid to be detected and motion frames corresponding to the fluid to be detected in the initial image frame. The processed image frames are motion frames corresponding to the fluid to be detected, and finally the motion frames corresponding to the fluid to be detected are put together to construct an image frame set.

Step S20: Acquire the fluid motion time corresponding to each frame of the image in the set of image frames.

It should be noted that, each image frame in the image frame set is in an adjacent relationship, and each frame in the image frame also has a corresponding fluid motion time.

Step S30: Obtain vector data corresponding to the fluid to be detected in each frame of the image, and calculate the velocity of the fluid to be detected according to the fluid movement time and the vector data.

The obtaining of the vector data corresponding to the fluid to be detected in each frame of image includes selecting a target image frame from the image frame set, then obtaining a corresponding target image according to the target image frame, and analyzing the target image. Gaussian filtering is performed on the image to obtain a vector field image, and finally vector data corresponding to the fluid to be detected is calculated by a preset optical flow algorithm according to the vector field image.

The above-mentioned target image is that the user uses a calculator to select a plurality of adjacent target image frames from the image frame set, then obtains the corresponding target image according to the target image frame, and converts the target image into grayscale. After that, Gaussian filtering is performed on the grayscale image to obtain a smooth image corresponding to the fluid to be detected, and the optical flow field corresponding to the fluid to be detected is calculated according to the smooth image, and the optical flow field is determined according to the optical flow field. The optical flow field detection area corresponding to the fluid to be detected is divided into regions to obtain a vector field image.

It should be understood that the smooth image is divided into grids to calculate the optical flow field, and then the isolated optical flow vector caused by the noise of the image is filtered out through threshold segmentation, and finally the selected optical flow vector is converted into two real lines and columns. Value image and solve. These values are averaged as the motion distance between the two frames.

The calculating, according to the vector field image, the vector data corresponding to the fluid to be detected by using a preset optical flow algorithm includes determining the coordinate value corresponding to the fluid to be detected according to the smooth image, and determining the coordinate value corresponding to the fluid to be detected according to the vector field image. Determine the pixel value corresponding to the fluid to be detected, obtain the effective weight corresponding to the fluid to be detected through a normal distribution algorithm according to the coordinate value, and derive the pixel value to obtain the corresponding value of the fluid to be detected. The directional gradient value is calculated according to the effective weight and the directional gradient value, and the corresponding vector data of the fluid to be detected is calculated.

The above step of obtaining the effective weights corresponding to the fluid to be detected according to the coordinate values through a normal distribution algorithm includes, according to the coordinate values, calculating the initial weights corresponding to the fluid to be detected through a normal distribution algorithm, The initial weights are screened to obtain effective weights corresponding to the fluid to be detected, which further includes: judging whether the initial weights belong to a preset weight threshold range, if the initial weights belong to the preset weights. If the weight threshold range is set, the initial weight is used as the effective weight corresponding to the fluid to be detected.

In addition, the preset optical flow algorithm mentioned above is to calculate the motion information of the corresponding pixels between two frames, and calculate the motion information of the small target pixels, so as to obtain the optical flow vector.

也称之为光流矢量。光流场是所有像素点的瞬时运动矢量。如果物体运动时，物体的像素亮度保持不变，则亮度的瞬时速度为光流，图像中所有像素的光流构成光流场。Among them, assuming that the motion interval is extremely small, we can regard it as a two-dimensional vector describing the instantaneous velocity of the point

Also known as the optical flow vector. The optical flow field is the instantaneous motion vector of all pixels. If the pixel brightness of the object remains unchanged when the object moves, the instantaneous speed of the brightness is the optical flow, and the optical flow of all pixels in the image constitutes the optical flow field.

The optical flow method is realized by the following three assumptions: (1) brightness consistency; (2) small motion between two frames; (3) spatial consistency.

When carrying out the optical flow constraint equation, the optical flow is also the fluid to be detected in this scheme. Let I(x,y,t) be the brightness value at the point (x,y) at time t, and I(x+dx , y+dy, t+dt) is the brightness value of the corresponding pixel at (x, y) when t+dt, assuming the brightness consistency, there are:

I(x,y,t)=I(x+dx,y+dy,t+dt) (1)

Expand the above right side with Taylor series, since the optical flow defines (u,v)=(dx/dt,dy/dt), (u,v)=(dx/dt,dy/dt) can be used as (u, v) Substitute to get the basic equation of optical flow:

That is, the optical flow constraint equation:

I _x u+I _y v+I _t =0 (3)

However, due to the existence of the aperture problem, only one optical flow constraint equation cannot solve the two unknowns (u, v), so other constraints need to be introduced. Different optical flow methods create different constraints, establish a new constraint equation, and make the solution of the equation unique.

Assuming that the motion vector is constant in a small spatial neighborhood Ω, then assign different weights to each point in the region. Assuming that there are n pixels in the neighborhood Ω, then each pixel should satisfy:

I _xi u+I _yi v+I _ti =0 i=1,2...,n (4)

The basic constraint equation of optical flow becomes:

E _c (u,v)=∫∫[I _x u+I _y v+I _t ] ² dxdy (5)

In the neighborhood Ω, the error of Lucas-Kanade optical flow is formulated as:

E _LK (u,v)＝∫∫W ² (x,y).(I _x u+I _y v+I _t ) ² dxdy (6)

Among them, W(x,y)={ _wi |i=1,2,...,n} is the weight of each point in the neighborhood, and its distribution characteristic is that the further away from the center, the greater the corresponding weight Small.

Discretizing formula (6), we get:

Among them, I _xi , I _yi , and I _ti are the gradient values of each point pixel in the field in the x, y, and t directions, and _wi is the weight of each point. make:

W=diag(w _x1 ,w _x2 ,...,w _xn ) (9)

b=[I _t1 ,I _t2 ,...,I _tn ] ^T (10)

Then formula (7) can be expressed as:

Then the solution of formula (11) can be expressed as:

Among them, A, W and b in the formula are the influencing factors in the formula, and the solution of formula (12) is the obtained optical flow vector data.

It should be noted that in recent years, moving target detection algorithms have emerged one after another, but the more classic algorithms can be divided into three types: frame difference method, background subtraction and optical flow method. Among them, the frame difference method has a simple algorithm and strong adaptability, but it is usually difficult to obtain the complete outline of the moving object, which is prone to the phenomenon of "hollow". Get the foreground object. Although background subtraction is relatively simple and has high real-time performance, it has high requirements for background quality and is very sensitive. Compared with these two algorithms, optical flow method has higher complexity and higher requirements for brightness stability. But the optical flow method can not only know the position of the moving object, but also get the speed and direction of the moving object clearly. The optical flow method does not require background modeling and background updating, and the resulting moving objects will not appear "hollow". In this experiment, the brightness of the background light source is high, the stability is high, and the moving distance between different frames is short, which is suitable for the use of the optical flow method.

In this embodiment, the flow velocity video corresponding to the fluid to be detected is acquired, the image frame is extracted from the flow velocity video, the initial image frame is obtained, and then the to-be-processed image frame is extracted from the initial image frame according to the preset ROI area, and Build an image frame set according to the to-be-processed image frames, obtain the fluid motion time corresponding to each frame of the image in the image frame set, select a target image frame from the image frame set, and obtain the corresponding image frame according to the target image frame target image, and then Gaussian filtering is performed on the target image to obtain a vector field image. According to the vector field image, the vector data corresponding to the fluid to be detected is calculated by a preset optical flow algorithm, and according to the fluid motion time and the vector data to calculate the velocity of the fluid to be detected. Through the above method, image processing algorithms such as preset optical flow algorithm and Gaussian filter algorithm can be used, which can effectively track the flowing liquid, and can accurately obtain the speed and direction of the fluid to be detected, with good stability and high efficiency. Lay the foundation for real-time control of all-position flow velocity in the tube.

Referring to FIG. 3 , FIG. 3 is a schematic flowchart of a second embodiment of a fluid velocity detection method in a microfluidic chip of the present invention.

Based on the above-mentioned first embodiment, the method for detecting fluid velocity in the microfluidic chip of this embodiment, in the step S30, further includes:

Step S301: Select a target image frame from the image frame set.

Step S302: Acquire a corresponding target image according to the target image frame.

Step S303: Convert the target image into a grayscale image, and perform Gaussian filtering on the grayscale image to obtain a smooth image corresponding to the fluid to be detected.

Step S304: Calculate the optical flow field corresponding to the fluid to be detected according to the smoothed image.

Step S305: Determine an optical flow field detection area corresponding to the fluid to be detected according to the optical flow field.

Step S306: Divide the optical flow field detection area to obtain a vector field image.

Step S307: Calculate vector data corresponding to the fluid to be detected by using a preset optical flow algorithm according to the vector field image.

Step S308: Calculate the velocity of the fluid to be detected according to the fluid movement time and the vector data.

The obtaining of the vector data corresponding to the fluid to be detected in each frame of image includes selecting a target image frame from the image frame set, then obtaining a corresponding target image according to the target image frame, and analyzing the target image. Gaussian filtering is performed on the image to obtain a vector field image, and finally vector data corresponding to the fluid to be detected is calculated by a preset optical flow algorithm according to the vector field image.

The above-mentioned target image is that the user uses a calculator to select a plurality of adjacent target image frames from the image frame set, then obtains the corresponding target image according to the target image frame, and converts the target image into grayscale. After that, Gaussian filtering is performed on the grayscale image to obtain a smooth image corresponding to the fluid to be detected, and the optical flow field corresponding to the fluid to be detected is calculated according to the smooth image, and the optical flow field is determined according to the optical flow field. The optical flow field detection area corresponding to the fluid to be detected is divided into regions to obtain a vector field image.

It should be understood that the smooth image is divided into grids to calculate the optical flow field, and then the isolated optical flow vector caused by the noise of the image is filtered out through threshold segmentation, and finally the selected optical flow vector is converted into two real lines and columns. Value image and solve. These values are averaged as the motion distance between the two frames.

The calculating, according to the vector field image, the vector data corresponding to the fluid to be detected by using a preset optical flow algorithm includes determining the coordinate value corresponding to the fluid to be detected according to the smooth image, and determining the coordinate value corresponding to the fluid to be detected according to the vector field image. Determine the pixel value corresponding to the fluid to be detected, obtain the effective weight corresponding to the fluid to be detected through a normal distribution algorithm according to the coordinate value, and derive the pixel value to obtain the corresponding value of the fluid to be detected. The directional gradient value is calculated according to the effective weight and the directional gradient value, and the corresponding vector data of the fluid to be detected is calculated.

The above step of obtaining the effective weights corresponding to the fluid to be detected according to the coordinate values through a normal distribution algorithm includes, according to the coordinate values, calculating the initial weights corresponding to the fluid to be detected through a normal distribution algorithm, The initial weights are screened to obtain effective weights corresponding to the fluid to be detected, which further includes: judging whether the initial weights belong to a preset weight threshold range, if the initial weights belong to the preset weights. If the weight threshold range is set, the initial weight is used as the effective weight corresponding to the fluid to be detected.

In addition, the preset optical flow algorithm mentioned above is to calculate the motion information of the corresponding pixels between two frames, and calculate the motion information of the small target pixels, so as to obtain the optical flow vector.

也称之为光流矢量。光流场是所有像素点的瞬时运动矢量。如果物体运动时，物体的像素亮度保持不变，则亮度的瞬时速度为光流，图像中所有像素的光流构成光流场。Among them, assuming that the motion interval is extremely small, we can regard it as a two-dimensional vector describing the instantaneous velocity of the point

Also known as the optical flow vector. The optical flow field is the instantaneous motion vector of all pixels. If the pixel brightness of the object remains unchanged when the object moves, the instantaneous speed of the brightness is the optical flow, and the optical flow of all pixels in the image constitutes the optical flow field.

The optical flow method is realized by the following three assumptions: (1) brightness consistency; (2) small motion between two frames; (3) spatial consistency.

When carrying out the optical flow constraint equation, the optical flow is also the fluid to be detected in this scheme. Let I(x,y,t) be the brightness value at the point (x,y) at time t, and I(x+dx , y+dy, t+dt) is the brightness value of the corresponding pixel at (x, y) when t+dt, assuming the brightness consistency, there are:

I(x,y,t)=I(x+dx,y+dy,t+dt) (1)

Expand the above right side with Taylor series, since the optical flow defines (u,v)=(dx/dt,dy/dt), (u,v)=(dx/dt,dy/dt) can be used as (u, v) Substitute to get the basic equation of optical flow:

That is, the optical flow constraint equation:

I _x u+I _y v+I _t =0 (3)

However, due to the existence of the aperture problem, only one optical flow constraint equation cannot solve the two unknowns (u, v), so other constraints need to be introduced. Different optical flow methods create different constraints, establish a new constraint equation, and make the solution of the equation unique.

Assuming that the motion vector is constant in a small spatial neighborhood Ω, then assign different weights to each point in the region. Assuming that there are n pixels in the neighborhood Ω, then each pixel should satisfy:

I _xi u+I _yi v+I _ti =0 i=1,2...,n (4)

The basic constraint equation of optical flow becomes:

E _c (u,v)=∫∫[I _x u+I _y v+I _t ] ² dxdy (5)

In the neighborhood Ω, the error of Lucas-Kanade optical flow is formulated as:

E _LK (u,v)＝∫∫W ² (x,y).(I _x u+I _y v+I _t ) ² dxdy (6)

Among them, W(x,y)={ _wi |i=1,2,...,n} is the weight of each point in the neighborhood, and its distribution characteristic is that the further away from the center, the greater the corresponding weight Small.

Discretizing formula (6), we get:

Among them, I _xi , I _yi , and I _ti are the gradient values of each point pixel in the field in the x, y, and t directions, and _wi is the weight of each point. make:

W=diag(w _x1 ,w _x2 ,...,w _xn ) (9)

b=[I _t1 ,I _t2 ,...,I _tn ] ^T (10)

Then formula (7) can be expressed as:

Then the solution of formula (11) can be expressed as:

Among them, A, W and b in the formula are the influencing factors in the formula, and the solution of formula (12) is the obtained optical flow vector data.

In this embodiment, a target image frame is selected from the image frame set, then a corresponding target image is obtained according to the target image frame, and Gaussian filtering is performed on the target image to obtain a vector field image, and then a vector field image is obtained according to the vector field image, and the vector data corresponding to the fluid to be detected is calculated by a preset optical flow algorithm. In the above manner, the position of the fluid to be detected is effectively tracked, so that the speed and direction of the movement of the fluid to be detected can be accurately obtained.

Referring to FIG. 4 , FIG. 4 is a structural block diagram of the first embodiment of the fluid velocity detection device in the microfluidic chip of the present invention.

As shown in FIG. 4 , the fluid velocity detection device in the microfluidic chip proposed by the embodiment of the present invention includes: an extraction module 4001, configured to acquire a flow velocity video corresponding to the fluid to be detected, and perform image frame extraction on the flow velocity video, Obtaining an image frame set; the obtaining module 4002 is used to obtain the fluid motion time corresponding to each frame of the image in the image frame set; the calculation module 4003 is used to obtain the vector data corresponding to the fluid to be detected in the each frame of image, and calculate the velocity of the fluid to be detected according to the fluid movement time and the vector data.

The extraction module 4001 obtains the flow velocity video corresponding to the fluid to be detected, and performs image frame extraction on the flow velocity video to obtain an image frame set.

It should be noted that the execution body of the method in this embodiment is a calculator, that is, a calculation method that can receive the flow velocity video corresponding to the fluid to be detected, process the flow velocity video to obtain vector data, and then calculate the vector data. device.

In this solution, the fluid to be detected is a flowable liquid, which may be ink or clear water. The microfluidic chip is a chip selected by the user and used in the experiment. The control chip adopts a PIE-14-007 hybrid chip, and another liquid or chip can also be used during the test, which is not limited in this embodiment.

The step of acquiring and acquiring the flow velocity video corresponding to the fluid to be detected, performing image frame extraction on the flow velocity video, and obtaining a set of image frames, includes acquiring the flow velocity video corresponding to the fluid to be detected, and performing image processing on the flow velocity video. Frame extraction, obtaining an initial image frame, extracting a to-be-processed image frame from the initial image frame according to a preset ROI area, and constructing an image frame set according to the to-be-processed image frame.

Wherein, the flow velocity video is the video of the fluid motion detection in the microfluidic chip under a given pump speed provided by the user, and then the flow velocity video is extracted, and the initial image frame of the flow velocity video is obtained, and then according to The user-defined area is to extract the image frame to be processed from the initial image frame. There may be blank frames corresponding to the fluid to be detected and motion frames corresponding to the fluid to be detected in the initial image frame. The processed image frames are motion frames corresponding to the fluid to be detected, and finally the motion frames corresponding to the fluid to be detected are put together to construct an image frame set.

The obtaining module 4002 obtains the operation of the fluid motion time corresponding to each frame of the image in the set of image frames.

It should be noted that, each image frame in the image frame set is in an adjacent relationship, and each frame in the image frame also has a corresponding fluid motion time.

The calculation module 4003 obtains the vector data corresponding to the fluid to be detected in each frame of the image, and calculates the velocity of the fluid to be detected according to the fluid movement time and the vector data.

The obtaining of the vector data corresponding to the fluid to be detected in each frame of image includes selecting a target image frame from the image frame set, then obtaining a corresponding target image according to the target image frame, and analyzing the target image. Gaussian filtering is performed on the image to obtain a vector field image, and finally vector data corresponding to the fluid to be detected is calculated by a preset optical flow algorithm according to the vector field image.

The above-mentioned target image is that the user uses a calculator to select a plurality of adjacent target image frames from the image frame set, then obtains the corresponding target image according to the target image frame, and converts the target image into grayscale. After that, Gaussian filtering is performed on the grayscale image to obtain a smooth image corresponding to the fluid to be detected, and the optical flow field corresponding to the fluid to be detected is calculated according to the smooth image, and the optical flow field is determined according to the optical flow field. The optical flow field detection area corresponding to the fluid to be detected is divided into regions to obtain a vector field image.

It should be understood that the smooth image is divided into grids to calculate the optical flow field, and then the isolated optical flow vector caused by the noise of the image is filtered out through threshold segmentation, and finally the selected optical flow vector is converted into two real lines and columns. Value image and solve. These values are averaged as the motion distance between the two frames.

The calculating, according to the vector field image, the vector data corresponding to the fluid to be detected by using a preset optical flow algorithm includes determining the coordinate value corresponding to the fluid to be detected according to the smooth image, and determining the coordinate value corresponding to the fluid to be detected according to the vector field image. Determine the pixel value corresponding to the fluid to be detected, obtain the effective weight corresponding to the fluid to be detected through a normal distribution algorithm according to the coordinate value, and derive the pixel value to obtain the corresponding value of the fluid to be detected. The directional gradient value is calculated according to the effective weight and the directional gradient value, and the corresponding vector data of the fluid to be detected is calculated.

The above step of obtaining the effective weights corresponding to the fluid to be detected according to the coordinate values through a normal distribution algorithm includes, according to the coordinate values, calculating the initial weights corresponding to the fluid to be detected through a normal distribution algorithm, The initial weights are screened to obtain effective weights corresponding to the fluid to be detected, which further includes: judging whether the initial weights belong to a preset weight threshold range, if the initial weights belong to the preset weights. If the weight threshold range is set, the initial weight is used as the effective weight corresponding to the fluid to be detected.

In addition, the preset optical flow algorithm mentioned above is to calculate the motion information of the corresponding pixels between two frames, and calculate the motion information of the small target pixels, so as to obtain the optical flow vector.

也称之为光流矢量。光流场是所有像素点的瞬时运动矢量。如果物体运动时，物体的像素亮度保持不变，则亮度的瞬时速度为光流，图像中所有像素的光流构成光流场。Among them, assuming that the motion interval is extremely small, we can regard it as a two-dimensional vector describing the instantaneous velocity of the point

Also known as the optical flow vector. The optical flow field is the instantaneous motion vector of all pixels. If the pixel brightness of the object remains unchanged when the object moves, the instantaneous speed of the brightness is the optical flow, and the optical flow of all pixels in the image constitutes the optical flow field.

The optical flow method is realized by the following three assumptions: (1) brightness consistency; (2) small motion between two frames; (3) spatial consistency.

When carrying out the optical flow constraint equation, the optical flow is also the fluid to be detected in this scheme. Let I(x,y,t) be the brightness value at the point (x,y) at time t, and I(x+dx , y+dy, t+dt) is the brightness value of the corresponding pixel at (x, y) when t+dt, assuming the brightness consistency, there are:

I(x,y,t)=I(x+dx,y+dy,t+dt) (1)

Expand the above right side with Taylor series, since the optical flow defines (u,v)=(dx/dt,dy/dt), (u,v)=(dx/dt,dy/dt) can be used as (u, v) Substitute to get the basic equation of optical flow:

That is, the optical flow constraint equation:

I _x u+I _y v+I _t =0 (3)

However, due to the existence of the aperture problem, only one optical flow constraint equation cannot solve the two unknowns (u, v), so other constraints need to be introduced. Different optical flow methods create different constraints, establish a new constraint equation, and make the solution of the equation unique.

Assuming that the motion vector is constant in a small spatial neighborhood Ω, then assign different weights to each point in the region. Assuming that there are n pixels in the neighborhood Ω, then each pixel should satisfy:

I _xi u+I _yi v+I _ti =0 i=1,2...,n (4)

The basic constraint equation of optical flow becomes:

E _c (u,v)=∫∫[I _x u+I _y v+I _t ] ² dxdy (5)

In the neighborhood Ω, the error of Lucas-Kanade optical flow is formulated as:

E _LK (u,v)＝∫∫W ² (x,y).(I _x u+I _y v+I _t ) ² dxdy (6)

Among them, W(x,y)={ _wi |i=1,2,...,n} is the weight of each point in the neighborhood, and its distribution characteristic is that the further away from the center, the greater the corresponding weight Small.

Discretizing formula (6), we get:

Among them, I _xi , I _yi , and I _ti are the gradient values of each point pixel in the field in the x, y, and t directions, and _wi is the weight of each point. make:

W=diag(w _x1 ,w _x2 ,...,w _xn ) (9)

b=[I _t1 ,I _t2 ,...,I _tn ] ^T (10)

Then formula (7) can be expressed as:

Then the solution of formula (11) can be expressed as:

Among them, A, W and b in the formula are the influencing factors in the formula, and the solution of formula (12) is the obtained optical flow vector data.

It should be noted that in recent years, moving target detection algorithms have emerged one after another, but the more classic algorithms can be divided into three types: frame difference method, background subtraction and optical flow method. Although the frame difference method has a simple algorithm and strong adaptability, it is usually difficult to obtain the complete outline of the moving object, and the phenomenon of "hollow" is prone to occur; the background subtraction method is to establish a background model through video frames, and subtract the background model from the input video sequence. Get the foreground object. Although background subtraction is relatively simple and has high real-time performance, it has high requirements on background quality and is very sensitive. Compared with these two algorithms, optical flow method has higher complexity and higher requirements for brightness stability. But the optical flow method can not only know the position of the moving object, but also get the speed and direction of the moving object clearly. The optical flow method does not require background modeling and background updating, and the resulting moving objects will not appear "hollow". In this experiment, the brightness of the background light source is high, the stability is high, and the moving distance between different frames is short, which is suitable for the use of the optical flow method.

It should be understood that the above are only examples, and do not constitute any limitation to the technical solutions of the present invention. In specific applications, those skilled in the art can make settings as required, which is not limited by the present invention.

In this embodiment, the flow velocity video corresponding to the fluid to be detected is acquired, the image frame is extracted from the flow velocity video, the initial image frame is obtained, and then the to-be-processed image frame is extracted from the initial image frame according to the preset ROI area, and Build an image frame set according to the to-be-processed image frames, obtain the fluid motion time corresponding to each frame of the image in the image frame set, select a target image frame from the image frame set, and obtain the corresponding image frame according to the target image frame target image, and then Gaussian filtering is performed on the target image to obtain a vector field image. According to the vector field image, the vector data corresponding to the fluid to be detected is calculated by a preset optical flow algorithm, and according to the fluid motion time and the vector data to calculate the velocity of the fluid to be detected. Through the above method, image processing algorithms such as preset optical flow algorithm and Gaussian filter algorithm can be used, which can effectively track the flowing liquid, and can accurately obtain the speed and direction of the fluid to be detected, with good stability and high efficiency. Lay the foundation for real-time control of all-position flow velocity in the tube.

It should be noted that the above-described workflow is only illustrative, and does not limit the protection scope of the present invention. In practical applications, those skilled in the art can select some or all of them to implement according to actual needs. The purpose of the solution in this embodiment is not limited here.

In addition, for technical details that are not described in detail in this embodiment, reference may be made to the method for detecting fluid velocity in a microfluidic chip provided by any embodiment of the present invention, which will not be repeated here.

Furthermore, it should be noted that, herein, the terms "comprising", "comprising" or any other variation thereof are intended to encompass non-exclusive inclusion, such that a process, method, article or system comprising a series of elements includes not only those elements, but also other elements not expressly listed or inherent to such a process, method, article or system. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in the process, method, article or system that includes the element.

The above-mentioned serial numbers of the embodiments of the present invention are only for description, and do not represent the advantages or disadvantages of the embodiments.

From the description of the above embodiments, those skilled in the art can clearly understand that the method of the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is better implementation. Based on such understanding, the technical solutions of the present invention can be embodied in the form of software products in essence or the parts that contribute to the prior art, and the computer software products are stored in a storage medium (such as a read-only memory (Read Only Memory). , ROM)/RAM, magnetic disk, optical disk), including several instructions to make a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) execute the methods described in the various embodiments of the present invention.

The above are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the description and drawings of the present invention, or directly or indirectly applied in other related technical fields , are similarly included in the scope of patent protection of the present invention.

Claims (10)

1. A method for detecting the speed of fluid in a microfluidic chip is characterized in that the microfluidic chip contains fluid to be detected, and the method comprises the following steps:

acquiring a flow speed video corresponding to the fluid to be detected, and performing image frame extraction on the flow speed video to obtain an image frame set;

acquiring fluid motion time corresponding to each frame of image in the image frame set;

and acquiring vector data corresponding to the fluid to be detected in each frame of image, and calculating the speed of the fluid to be detected according to the fluid motion time and the vector data.

2. The method of claim 1, wherein the step of obtaining a flow rate video corresponding to the fluid to be detected, performing image frame extraction on the flow rate video, and obtaining an image frame set comprises:

acquiring a flow speed video corresponding to the fluid to be detected, and performing image frame extraction on the flow speed video to obtain an initial image frame;

and extracting an image frame to be processed from the initial image frame according to a preset ROI (region of interest), and constructing an image frame set according to the image frame to be processed.

3. The method of claim 1, wherein the step of obtaining vector data corresponding to the fluid to be detected in each of the frames of images comprises:

selecting a target image frame from the set of image frames;

acquiring a corresponding target image according to the target image frame;

converting the target image into a gray image, and performing Gaussian filtering processing on the gray image to obtain a smooth image corresponding to the fluid to be detected;

calculating an optical flow field corresponding to the fluid to be detected according to the smooth image;

determining an optical flow field detection area corresponding to the fluid to be detected according to the optical flow field;

carrying out region division on the optical flow field detection region to obtain a vector field image;

and calculating vector data corresponding to the fluid to be detected by a preset optical flow algorithm according to the vector field image.

4. The method according to claim 3, wherein said step of calculating the vector data corresponding to said fluid to be detected by a preset optical flow algorithm based on said vector field image comprises:

determining coordinate values corresponding to the fluid to be detected according to the smooth image;

determining a pixel value corresponding to the fluid to be detected according to the vector field image;

obtaining an effective weight value corresponding to the fluid to be detected through a normal distribution algorithm according to the coordinate value;

deriving the pixel values to obtain a direction gradient value corresponding to the fluid to be detected;

and calculating the vector data corresponding to the fluid to be detected according to the effective weight and the direction gradient value.

5. The method according to claim 4, wherein the step of obtaining the effective weight corresponding to the fluid to be detected by a normal distribution algorithm according to the coordinate value comprises:

calculating an initial weight corresponding to the fluid to be detected through a normal distribution algorithm according to the coordinate value;

and screening the initial weight to obtain an effective weight corresponding to the fluid to be detected.

6. The method of claim 5, wherein the step of screening the initial weight to obtain the effective weight corresponding to the fluid to be detected comprises:

judging whether the initial weight value belongs to a preset weight value threshold range or not;

and if the initial weight value belongs to the preset weight value threshold range, taking the initial weight value as an effective weight value corresponding to the fluid to be detected.

7. A device for detecting the speed of fluid in a microfluidic chip, which is characterized in that the microfluidic chip contains fluid to be detected, the device comprises:

the extraction module is used for acquiring a flow speed video corresponding to the fluid to be detected, and performing image frame extraction on the flow speed video to obtain an image frame set;

the acquisition module is used for acquiring fluid motion time corresponding to each frame of image in the image frame set;

and the calculation module is used for acquiring the vector data corresponding to the fluid to be detected in each frame of image and calculating the speed of the fluid to be detected according to the fluid motion time and the vector data.

8. The device according to claim 7, wherein the extracting module is further configured to obtain a flow rate video corresponding to the fluid to be detected, and perform image frame extraction on the flow rate video to obtain an initial image frame;

the extraction module is further configured to extract an image frame to be processed from the initial image frame according to a preset ROI region, and construct an image frame set according to the image frame to be processed.

9. The apparatus of claim 7, wherein the computing module is further configured to select a target image frame from the set of image frames;

the computing module is further used for acquiring a corresponding target image according to the target image frame;

the calculation module is further configured to perform gaussian filtering processing on the target image to obtain a vector field image;

and the calculation module is also used for calculating vector data corresponding to the fluid to be detected through a preset optical flow algorithm according to the vector field image.

10. An apparatus for detecting the velocity of a fluid in a microfluidic chip, the apparatus comprising: the method comprises the following steps of a CCD camera, a microscope, a backlight plate, a chip fixing clamp, a vibration isolation platform, a display, a servo injection controller, an injection pump, a memory, a processor and a fluid speed detection program in the microfluidic chip, wherein the fluid speed detection program is stored in the memory and can run on the processor, and the fluid speed detection program in the microfluidic chip is configured to realize the steps of the fluid speed detection method in the microfluidic chip according to any one of claims 1 to 6.

Images