CN115359028A - Liquid flow velocity detection method, detection device and storage medium - Google Patents

Abstract

The embodiment of the present application discloses a liquid flow velocity detection method, detection device, and storage medium, which are used in the technical field of velocity detection. The method in the embodiment of the application includes: acquiring a video stream of the flowing liquid, and converting the multi-frame video in the video stream to The image is converted into a multi-frame grayscale image; according to the pixel position offset of the same particle in the liquid between any two adjacent grayscale images, the target particles that do not meet the preset difference conditions are determined, and the target particles are deleted. Multi-frame target grayscale images are obtained from the corresponding pixels in the frame grayscale image; the pixel position offset and interval time of the same particle are substituted into the sparse optical flow algorithm to obtain the target optical flow velocity; scaling according to the corresponding size of the video stream Ratio, the target optical flow velocity is scaled to obtain the target flow velocity of the liquid, which can measure a more accurate liquid flow velocity.

Description

Liquid flow velocity detection method, detection device and storage medium

technical field

The embodiments of the present application relate to the technical field of speed detection, and in particular, to a liquid flow speed detection method, detection device, and storage medium.

Background technique

River flow velocity measurement is one of the important tasks of water flow monitoring in natural rivers and artificial channels. Accurate monitoring of water flow conditions is conducive to preventing geological disasters such as mountain torrents. Liquid velocity is an important description of liquid behavior and one of the important parameters in the study of fluid dynamics. Flow velocity detection is the basis of flow detection. Only when the flow velocity is accurately detected can the flow be accurately measured. Effective detection of flow velocity is very important.

Existing methods for detecting flow velocity generally include rotameters and ultrasonic radar flowmeters. The rotor-type flow meter needs to be placed in the flow field, which will interfere with the flow field body of the liquid, thereby changing its local flow velocity, and is generally used for single-point flow velocity detection; the ultrasonic radar flow meter is based on the ultrasonic principle, using Doppler In the process of contact with the measured liquid, the flow velocity of the measured liquid is measured by using Doppler frequency shift. The ultrasonic wave is unstable during the liquid transmission process, and the transmission distance should not be too far. Generally, the local speed measurement and measurement range is limited.

Therefore, the existing flow velocity detection methods are generally used for single-point flow velocity detection or local velocity measurement, and the measured liquid flow velocity is not accurate enough.

Contents of the invention

Embodiments of the present application provide a liquid flow rate detection method, a detection device, and a storage medium, capable of measuring a relatively accurate liquid flow rate.

The embodiment of the present application provides a liquid flow rate detection method, including:

Obtaining a video stream for capturing a flowing liquid, and converting multiple frames of video images in the video stream into multiple frames of grayscale images;

According to the pixel position offset of the same particle in the liquid between any two adjacent frames of grayscale images, determine the target particles that do not meet the preset difference conditions, and delete the target particles in the grayscale of multiple frames corresponding pixels in the image to obtain multiple frames of target grayscale images;

Determine the pixel point offset of the same particle in two adjacent frames of the target gray image in the multiple frames of the target gray image, and determine the interval between the two adjacent frames of the target gray image; Substituting the pixel position offset and the interval duration into the sparse optical flow algorithm to obtain the target optical flow velocity;

According to the scaling ratio corresponding to the video stream, the target optical flow velocity is scaled to obtain the target flow velocity of the liquid.

Further, said any two adjacent frames of grayscale images include: a first frame of grayscale image and a second frame of grayscale image;

The determining the target particle that does not meet the preset difference condition according to the pixel position offset of the same particle in the liquid between any two adjacent frames of grayscale images includes:

Obtain the first pixel point position offset from the first frame of grayscale image to the second frame of grayscale image, and obtain the second pixel point position offset from the second frame of grayscale image to the first frame of grayscale image Pixel position offset;

Acquiring the first pixel position of any pixel in the first frame of grayscale image;

According to the offset of the first pixel point position and the first pixel point position, obtain the pixel point position corresponding to the grayscale image of the second frame of any pixel point;

Predicting a second pixel position of any pixel in the first frame of grayscale image according to the second pixel position offset and the pixel position of the second frame of grayscale image;

If the distance between the first pixel point position and the second pixel point position is greater than a preset distance threshold, it is determined that the particle corresponding to any pixel point is a target particle that does not satisfy the preset difference condition.

Further, the step of substituting the pixel position offset of the same particle and the interval duration into the sparse optical flow algorithm to obtain the target optical flow velocity includes:

得到所述目标光流速度

其中所述

所述

x、y表示所述同一粒子在x、y方向的像素点位置偏移量，t表示所述间隔时长，Ix,Iy,It分别表示x、y方向的像素点位置偏移量的偏导数以及间隔时长t的偏导数。According to the sparse optical flow algorithm

Get the target optical flow velocity

which stated

said

x, y represent described same particle in x, the pixel point position offset of y direction, t represents described interval duration, Ix, Iy, It represent respectively the partial derivative of the pixel point position offset of x, y direction and The partial derivative of the interval duration t.

Further, the step of substituting the pixel position offset of the same particle and the interval duration into the sparse optical flow algorithm to obtain the target optical flow speed includes:

By continuously down-sampling the two adjacent frames of target grayscale images, the two adjacent frames of target grayscale images obtained by each sampling are constructed into an image pyramid; wherein, the bottom layer of the image pyramid is the two adjacent frames The frame target grayscale image, the highest layer is two adjacent frames of target grayscale images sampled at the lowest sampling rate;

Substituting the pixel position offset of the same particle in the highest layer image in the image pyramid, and the interval between two adjacent frames of target grayscale images in the highest layer image into the sparse optical flow algorithm, the highest image pyramid is obtained. The first optical flow velocity of the layer image;

Iteratively calculate the optical flow velocity of each layer image of the image pyramid based on the first optical flow velocity until the optical flow velocity of the bottom image is obtained, and the iteration is completed, and the optical flow velocity of the bottom image is used as the target Optical flow velocity.

Further, before performing scaling processing on the target optical flow velocity according to the scaling ratio corresponding to the video stream to obtain the target flow velocity of the liquid, the method further includes:

Calculate the linear relationship between the flow distance of the liquid within the interval time and the displacement distance of the pixel point within the interval time, and use the linear relationship as the size scaling ratio.

Further, the two adjacent frames of target grayscale images are multiple groups, and multiple groups of two adjacent frames of target grayscale images correspond to multiple target optical flow velocities;

The step of scaling the target optical flow velocity to obtain the target flow velocity of the liquid according to the scaling ratio corresponding to the video stream includes:

determining an optical flow velocity according to a plurality of target optical flow velocities;

converting the optical flow velocity into a target flow velocity of the liquid according to the linear relationship.

Further, the two adjacent frames of target grayscale images are multiple groups, and multiple groups of two adjacent frames of target grayscale images correspond to multiple target optical flow velocities;

The step of scaling the target optical flow velocity to obtain the target flow velocity of the liquid according to the scaling ratio corresponding to the video stream includes:

converting a plurality of the target optical flow velocities into a plurality of flow velocities of the liquid according to the linear relationship;

A target flow rate of the liquid is determined according to a plurality of flow rates of the liquid.

The embodiment of the present application also provides a liquid flow rate detection device, including:

An acquisition unit, configured to acquire a video stream for capturing a flowing liquid, and convert multiple frames of video images in the video stream into multiple frames of grayscale images;

The deletion unit is used to determine the target particles that do not meet the preset difference conditions according to the pixel position offset of the same particle in the liquid between any two adjacent frames of grayscale images, and the target particles in the multi-frame corresponding pixels in the grayscale image to obtain multiple frames of target grayscale images;

A determining unit, configured to determine the pixel point offset of the same particle in two adjacent frames of the target gray-scale image in the multi-frame target gray-scale image, and determine the interval between the two adjacent frames of the target gray-scale image; execute A unit for substituting the pixel position offset of the same particle and the interval duration into the sparse optical flow algorithm to obtain the target optical flow velocity;

The processing unit is configured to perform scaling processing on the target optical flow velocity to obtain the target flow velocity of the liquid according to the scaling ratio corresponding to the video stream.

The embodiment of the present application also provides a liquid flow rate detection device, including:

Central processing unit, memory, input and output interface, wired or wireless network interface, power supply;

The memory is a temporary storage memory or a persistent storage memory;

The central processing unit is configured to communicate with the memory, and to execute instruction operations in the memory on the control plane functional entity to execute the above-mentioned flow velocity detection method.

The embodiment of the present application also provides a computer-readable storage medium, the computer-readable storage medium includes instructions, and when the instructions are run on a computer, the computer is made to execute the above method for detecting flow velocity.

It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

In the embodiment of the present application, the video stream of the flowing liquid is obtained, and the multi-frame video images in the video stream are converted into multi-frame grayscale images; according to the pixel position of the same particle in the liquid between any two adjacent grayscale images Offset, determine the target particles that do not meet the preset difference conditions, and delete the corresponding pixels of the target particles in the multi-frame grayscale image to obtain a multi-frame target grayscale image; The pixel position offset of the same particle in the target grayscale image of the frame determines the interval between two adjacent frames of the target grayscale image; the pixel position offset and the interval of the same particle are substituted into the sparse optical flow algorithm to obtain the target Optical flow velocity: according to the scaling ratio corresponding to the video stream, the target optical flow velocity is scaled to obtain the target flow velocity of the liquid. The target flow velocity of the liquid is obtained by the pixel position offset of the same particle in two adjacent frames of the target grayscale image in multiple frames of the target grayscale image, and the velocity of the liquid flow surface can be measured globally, and the measured liquid flow velocity is more accurate.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings that need to be used in the description of the embodiments. Obviously, the drawings in the following description are only some implementations recorded in the application. For example, those skilled in the art can also obtain other drawings based on these drawings.

Fig. 1 is a flow chart of the flow rate detection of a liquid disclosed in the embodiment of the present application;

Fig. 2 is a flow chart of the flow rate detection of another liquid disclosed in the embodiment of the present application;

Fig. 3 is a schematic diagram of a liquid layered velocity measurement disclosed in the embodiment of the present application;

FIG. 4 is a schematic diagram of a flow rate detection device disclosed in an embodiment of the present application;

Fig. 5 is a schematic diagram of another flow velocity detection device disclosed in the embodiment of the present application.

Detailed ways

In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiment of the application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiment of the application. Obviously, the described embodiment is only It is a part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

In the description of the embodiments of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer " and other indicated orientations or positional relationships are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the embodiments of the present application and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, Constructed and operative in a particular orientation and therefore should not be construed as limiting to the embodiments of the present application.

In the description of the embodiments of this application, it should be noted that unless otherwise specified and limited, the terms "installation", "connection", and "connection" should be understood in a broad sense, for example, it can be a fixed connection or a A detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be a direct connection or an indirect connection through an intermediary, and it may be an internal communication between two components. Those of ordinary skill in the art can understand the specific meanings of the above terms in the embodiments of the present application in specific situations.

Existing methods for detecting the flow velocity of liquids generally include rotor-type flow meters and ultrasonic radar flow meters. The rotor-type flow meters need to be placed in the flow field, which will interfere with the flow field body of the liquid, thereby changing its local flow velocity, and It is generally used for single-point flow velocity detection; the ultrasonic radar flowmeter is based on the principle of ultrasound, using the Doppler effect, and using the Doppler frequency shift to measure the flow velocity of the liquid under test during the contact process with the liquid under test. There is instability in the process of liquid transmission, and the transmission distance should not be too far. Generally, the velocity is measured locally, and the measurement range is limited. It can be seen that the existing flow velocity detection methods are generally used for single-point flow velocity detection or local velocity measurement, and the measured liquid flow velocity is not accurate enough. Therefore, the embodiment of the present application provides a liquid flow rate detection method, which can make the measured liquid flow rate more accurate, as shown in Figure 1, the specific steps are as follows:

101. Acquire a video stream for capturing a flowing liquid, and convert multiple frames of video images in the video stream into multiple frames of grayscale images.

The flow velocity detection device can acquire a video stream of shooting liquid, and convert multiple frames of video images in the video stream into multiple frames of grayscale images. The liquid (body of water) includes water and substances dissolved therein, and is primarily a non-turbid liquid. The flow velocity detection device can control the laser to emit a flat laser to the flowing liquid. After the laser is injected into the liquid, the impurities in the liquid will reflect the laser light; the flow velocity detection device controls the camera to capture the reflected particle image and collect the laser reflection point of the liquid. The image gets a video stream of the liquid. Next, the rgb2gray() function or RGB three-channel method can be used to convert multiple frames of video images in the video stream into multiple frames of grayscale images, which are not limited here.

102. Delete the pixel points corresponding to the target particles in the multiple frames of grayscale images to obtain multiple frames of target grayscale images.

The flow velocity detection device can determine the target particles that do not meet the preset difference conditions according to the pixel position offset of the same particle in the liquid between any two adjacent gray-scale images, and delete the target particles corresponding to the multi-frame gray-scale images. pixels to obtain multiple frames of target grayscale images. Specifically, any two adjacent frames of grayscale images include: a first frame of grayscale image and a second frame of grayscale image; wherein, according to the pixel points of the same particle in the liquid between any two adjacent frames of grayscale images Position offset, determining the target particles that do not meet the preset difference conditions includes: obtaining the first pixel position offset from the first frame of grayscale image to the second frame of grayscale image, and obtaining the second frame of grayscale image to The second pixel position offset of the first frame grayscale image; it can be understood that a plurality of liquid particles can be arbitrarily taken to correspond to multiple pixel positions of the first frame grayscale image and the second frame grayscale image, which can be Taking multiple pixel positions of the first frame of grayscale image as the starting point, and taking multiple pixel positions of the second frame of grayscale image as the end point, obtain multiple offsets of multiple pixel point positions in the x and y directions , taking the mode or average value of the plurality of offsets, the first pixel position offset from the first frame grayscale image to the second frame grayscale image can be obtained; the second pixel position offset Again, a plurality of liquid particles can be arbitrarily selected to correspond to multiple pixel positions of the first frame grayscale image and the second frame grayscale image, and the second frame grayscale image can be obtained starting from the multiple pixel point positions of the second frame grayscale image The second pixel position offset from the grayscale image to the first frame of grayscale image. At the same time, the first pixel point position of any pixel point in the first frame of the grayscale image can be obtained, and the first pixel point position is generally a real pixel point position.

According to the first pixel position offset and the first pixel position, any pixel corresponding to the pixel position of the second frame grayscale image is obtained; it can be understood that the first pixel position offset is generally The offset in the x and y directions, according to the position offset of the first pixel point, any pixel point is offset in the x and y directions, so that any pixel point corresponding to the second frame of the grayscale image can be obtained Pixel location. According to the second pixel position offset and the pixel position of the second frame grayscale image, predict the second pixel position of any pixel in the first frame grayscale image; it can be understood that according to the second pixel Point position offset After offsetting the pixel point position of the second frame grayscale image, the second pixel point position in the first frame grayscale image of any pixel point can be predicted. Generally, the first frame grayscale image The distance between the first pixel point and the second pixel point of any pixel point on the image is close to or at the same position. If the distance between the first pixel point position and the second pixel point position is greater than the preset distance threshold, it is determined that the particle corresponding to any pixel point is a target particle that does not meet the preset difference condition, and the preset distance threshold can be 10 cm or 20 cm, which is not limited here; the distance between the first pixel position and the second pixel position is too far, and the deviation of any pixel on the surface in the grayscale image is large, and any pixel needs to be deleted The point corresponds to the pixel point of the target particle in the multi-frame grayscale image.

In an achievable scheme, as shown in Figure 2, the first frame image is obtained from the video stream, and the first frame image is converted into a grayscale image, and the trajectory array Trajectories is created, and the first frame grayscale image The corner position (the corner position is the pixel position of the liquid particle in the grayscale image) is added to the array Trajectories, the second frame image is obtained from the video stream, and converted into a grayscale image; according to the first frame grayscale image to The pixel position offset of the second frame of grayscale image and the corner position of the first frame of grayscale image obtain the corner position of the second frame of grayscale image; according to the second frame of grayscale image to the first frame of grayscale image Obtain the corner position of the first frame gray image; subtract the obtained first frame corner point from the corresponding point in the array Trajectories to find The distance between the two; determine whether the distance between the two is less than the preset distance threshold, if not, delete the corner point; if so, add the corner point of the second frame of grayscale image to the array new-Trajectories, and assign the points in new-Trajectories Given the array Trajectories, the corner trajectory can be obtained, and the optical flow velocity can be calculated by using the LK algorithm.

103. Determine the pixel position offset and interval duration of the same particle in two adjacent frames of the target grayscale image.

The flow velocity detection device can determine the pixel position offset of the same particle in two adjacent frames of the target gray image in the multi-frame target gray image, and determine the interval between two adjacent frames of the target gray image. It can be understood that the same particle refers to a plurality of particles, and the pixel position offset of the same particle in two adjacent frames of target grayscale images is generally equal to the number of particles corresponding to the two adjacent frames of target grayscale images. Multiple pixel position offsets of the frame target grayscale image. The pixel point offset of the same particle can be determined by the pixel point position of the same particle in two adjacent frames of target grayscale images. The pixel position offset generally refers to the pixel moving a distance of (dx, dy) to the next frame.

104. Substitute the pixel position offset and interval duration of the same particle into the sparse optical flow algorithm to obtain the target optical flow velocity.

The flow velocity detection device can substitute the pixel position offset and interval time of the same particle into the sparse optical flow algorithm (LK algorithm) to obtain the target optical flow velocity. The trajectory of the particle is obtained by analyzing and processing the grayscale image, specifically: the light intensity of the pixel point of the same particle in the first frame of grayscale image is I(x,y,t), and the distance moved by (dx,dy) To the second frame of grayscale image, dt time is used. x and y represent the pixel position offset of the same particle in the x and y directions, and t represents the interval duration; assuming that the light intensity of the pixel is constant before and after the movement, then:

I(x,y,t)=I(x+δx,y+δy,t+δt)

Carry out Taylor expansion of the formula on the right side of the equal sign, ignore the infinitesimal term, and divide dt on both sides at the same time, you can get:

Let u and v be the velocity vectors of the optical flow along the x-axis and y-axis respectively, then

Ix, Iy, It respectively represent the partial derivative of the pixel position offset in the x and y directions and the partial derivative of the interval time t, that is

In summary, I _x u+I _y v+I _t =0.

Assuming that the optical flow is a constant in the neighborhood of the pixel, the optical flow equation is solved for all the pixels in the neighborhood using the least square method. Introducing additional constraints, assuming that the optical flow of the image is a constant value in a window with a size of m×m (n=m ² ), then a system of equations can be obtained:

记

采用最小二乘法，可以得到：

即目标光流速度

可以理解的是，得到的目标光流速度一般为光流矢量。Expressed in matrix form as:

remember

Using the method of least squares, we can get:

target optical flow velocity

It can be understood that the obtained target optical flow velocity is generally an optical flow vector.

Furthermore, when the speed of the object is high, the error of the algorithm is relatively large. In order to reduce the speed of the object in the image, one method is to reduce the size of the image. Assuming that when the image is 400×400, the object speed is [1616], then when the image is reduced to 200×200, the speed becomes [8,8]. When scaling down to 100×100, the speed is reduced to [4,4]. So you can slow down the speed by reducing the image size to get more accuracy. Therefore, by continuously down-sampling the two adjacent frames of the target gray-scale images, the two adjacent frames of the target gray-scale images obtained by each sampling can be constructed as an image pyramid; The target grayscale image, the highest layer is two adjacent frames of target grayscale images sampled at the lowest sampling rate; the pixel position offset of the same particle in the highest layer image in the image pyramid, and the Substitute the interval time of the frame target grayscale image into the sparse optical flow algorithm to obtain the first optical flow velocity of the image at the highest level of the image pyramid; based on the first optical flow velocity, iteratively calculate the optical flow velocity of each layer of the image pyramid until the final The optical flow velocity of the underlying image ends the iteration, and the optical flow velocity of the bottom image is used as the target optical flow velocity. In an achievable solution, (1) build a pyramid. Let I0=I be the image of the 0th layer, which is the image with the highest resolution in the pyramid image, and the width and height of the image are defined as nx0=nx and ny0=ny respectively. Build the pyramid in a recursive manner: compute I1 from I0, compute I2 from I1.... (2) Pyramid-based tracking. First, the optical flow and affine transformation matrix are calculated on the image of the highest layer; the calculation result of the previous layer is passed to the image of the next layer as the initial value, and the image of this layer is calculated on the basis of this initial value. The optical flow and affine change matrix of this layer; then pass the optical flow and affine matrix of this layer as the initial value to the image of the next layer until it is passed to the last layer, that is, the original image layer, and this layer calculates The resulting optical flow and affine transformation matrix are the results of the final optical flow and affine transformation matrix. (3) Iterative process. At each level of the pyramid, the goal is to compute the optical flow and affine transformation matrix to minimize the error. The result of each layer is iterated layer by layer, and finally the target optical flow speed is obtained.

105. According to the scaling ratio corresponding to the video stream, the target optical flow velocity is scaled to obtain the target flow velocity of the liquid.

After the target optical flow velocity is obtained, the target optical flow velocity can be scaled according to the scaling ratio corresponding to the video stream to obtain the target flow velocity of the liquid. It can be understood that when the video stream is shot, there is a certain deviation between the displacement distance of the pixels in the video image obtained according to different shooting methods and the actual distance. The actual liquid flow velocity can be obtained by the speed change of the pixel and the actual distance in the top view angle, which is linearly related. Specifically, the linear relationship between the flow distance of the liquid within the time interval and the displacement distance of the pixel point within the time interval can be calculated, and the linear relationship can be used as a scaling ratio. The linear relationship can be obtained through the proportional relationship between the flow distance of the liquid and the displacement distance of the pixel point within the time interval between two adjacent frames of images.

It can be understood that when two adjacent frames of target grayscale images are in multiple groups, multiple groups of adjacent two frames of target grayscale images can correspond to multiple target optical flow velocities; One optical flow velocity is determined from one target optical flow velocity, and the optical flow velocity is transformed to obtain the target flow velocity of the liquid; it is also possible to convert multiple target optical flow velocities into multiple liquid flow velocities first, and then from multiple target optical flow velocities Determine the target flow rate of the liquid from the flow rates of the liquids. Specifically, determining the optical flow velocity according to multiple target optical flow velocities may be to obtain the optical flow velocity by taking the mode or average value of the multiple target optical flow velocities, which is not specifically limited here. Next, the optical flow velocity is converted into the target flow velocity of the liquid according to the linear relationship. It can be understood that the target flow velocity of the liquid can be obtained by multiplying the optical flow velocity by the ratio of the linear relationship. Further, the multiple target optical flow velocities can be converted into multiple flow velocities of the liquid according to the linear relationship; and then the target flow rate of the liquid is determined according to the multiple flow velocities of the liquid. The target flow rate of the liquid can be obtained by taking the mode or averaging the multiple flow rates of the liquid.

In the embodiment of the present application, the video stream of the flowing liquid is obtained, and the multi-frame video images in the video stream are converted into multi-frame grayscale images; according to the pixel position of the same particle in the liquid between any two adjacent grayscale images Offset, determine the target particles that do not meet the preset difference conditions, and delete the corresponding pixels of the target particles in the multi-frame grayscale image to obtain a multi-frame target grayscale image; The pixel position offset of the same particle in the target grayscale image of the frame determines the interval between two adjacent frames of the target grayscale image; the pixel position offset and the interval of the same particle are substituted into the sparse optical flow algorithm to obtain the target Optical flow velocity: according to the scaling ratio corresponding to the video stream, the target optical flow velocity is scaled to obtain the target flow velocity of the liquid. The target flow velocity of the liquid is obtained by the pixel position offset of the same particle in two adjacent frames of the target grayscale image in multiple frames of the target grayscale image, and the velocity of the liquid flow surface can be measured globally, and the measured liquid flow velocity is more accurate.

In an achievable way, as shown in Figure 3, when it is necessary to obtain the liquid flow velocity of different flow layers, the laser can be controlled to move to the specified flow layer, which refers to the water surface to the bottom of the liquid. any plane. Next, the obtained liquid video flow of each flow layer is passed through the above steps, and the corresponding liquid flow velocity of each flow layer can be obtained. It can be seen that the liquid is projected and irradiated by the laser beam, and the particles in the liquid produce diffuse reflection on the laser, and then the particles are imaged from the camera, and the trajectory of the particles between different frames is tracked, and the trajectory is captured by the camera, and its movement The velocity is analyzed and calculated to obtain the velocity of the particle, and the velocity of the particle is used to represent the velocity of the liquid to realize the measurement of the velocity of the liquid. The benefits of using the above method are: the reflective properties of the particles in the water are excited by the laser, and their movement trajectories can be displayed under the optical camera; the movement of the particles in the water is captured by the camera, and the flow rate of the liquid is realized by analyzing the key corners of the video. The measurement of the liquid flow rate of different flow layers can be realized by moving the laser transmitter; the accurate measurement of the flow rate of different flow layers can be accurate and the liquid flow rate of the entire section can be obtained.

The embodiment of the present application also provides a liquid flow rate detection device, as shown in Figure 4, including:

The acquiring unit 401 is configured to acquire a video stream of shooting a flowing liquid, and convert multiple frames of video images in the video stream into multiple frames of grayscale images;

The deletion unit 402 is used to determine the target particles that do not meet the preset difference conditions according to the pixel position offset of the same particle in the liquid between any two adjacent frames of grayscale images, and determine the target particles in multiple Frame corresponding pixels in the grayscale image to obtain multiple frames of target grayscale images;

A determining unit 403, configured to determine the pixel position offset of the same particle in two adjacent frames of the target gray-scale image in the multiple frames of the target gray-scale image, and determine the interval between the two adjacent frames of the target gray-scale image; Executing unit 404, configured to substitute the pixel position offset of the same particle and the interval duration into the sparse optical flow algorithm to obtain the target optical flow velocity;

The processing unit 405 is configured to perform scaling processing on the target optical flow velocity to obtain the target flow velocity of the liquid according to the scaling ratio corresponding to the video stream.

The embodiment of the present application also provides a liquid flow velocity detection device 500, as shown in FIG. 5, including:

Central processing unit 501, memory 502, input and output interface 503, wired or wireless network interface 504, power supply 505;

The memory 502 is a temporary storage memory or a persistent storage memory;

The central processing unit 501 is configured to communicate with the memory 502, and execute instructions in the memory 502 on the control plane functional entity to execute the above-mentioned flow velocity detection method.

Those skilled in the art can clearly understand that for the convenience and brevity of the description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiment, which will not be repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or May be integrated into another system, or some features may be ignored, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

If the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or part of the contribution to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), magnetic disk or optical disk, and other media that can store program codes.

Claims (10)

1. A liquid flow velocity detection method, characterized in that, comprising: Obtaining a video stream for capturing a flowing liquid, and converting multiple frames of video images in the video stream into multiple frames of grayscale images; According to the pixel position offset of the same particle in the liquid between any two adjacent frames of grayscale images, determine the target particles that do not meet the preset difference conditions, and delete the target particles in the grayscale of multiple frames corresponding pixels in the image to obtain multiple frames of target grayscale images; Determine the pixel point offset of the same particle in two adjacent frames of the target gray image in the multiple frames of the target gray image, and determine the interval between the two adjacent frames of the target gray image; Substituting the pixel position offset and the interval duration into the sparse optical flow algorithm to obtain the target optical flow velocity; According to the scaling ratio corresponding to the video stream, the target optical flow velocity is scaled to obtain the target flow velocity of the liquid. 2. The flow velocity detection method according to claim 1, wherein said any two adjacent frames of grayscale images comprise: a first frame of grayscale image and a second frame of grayscale image; The determining the target particle that does not meet the preset difference condition according to the pixel position offset of the same particle in the liquid between any two adjacent frames of grayscale images includes: Obtain the first pixel point position offset from the first frame of grayscale image to the second frame of grayscale image, and obtain the second pixel point position offset from the second frame of grayscale image to the first frame of grayscale image Pixel position offset; Acquiring the first pixel position of any pixel in the first frame of grayscale image; According to the offset of the first pixel point position and the first pixel point position, obtain the pixel point position corresponding to the grayscale image of the second frame of any pixel point; Predicting a second pixel position of any pixel in the first frame of grayscale image according to the second pixel position offset and the pixel position of the second frame of grayscale image; If the distance between the first pixel point position and the second pixel point position is greater than a preset distance threshold, it is determined that the particle corresponding to any pixel point is a target particle that does not satisfy the preset difference condition. 3. The flow velocity detection method according to claim 1, wherein said substituting the pixel position offset of the same particle and the interval duration into the sparse optical flow algorithm to obtain the target optical flow velocity comprises: According to the sparse optical flow algorithm

Get the target optical flow velocity

which stated

said

x, y represent described same particle in x, the pixel point position offset of y direction, t represents described interval duration, Ix, Iy, It represent respectively the partial derivative of the pixel point position offset of x, y direction and The partial derivative of the interval duration t. 4. The flow velocity detection method according to claim 1, wherein said substituting the pixel position offset of the same particle and the interval duration into the sparse optical flow algorithm to obtain the target optical flow velocity comprises: By continuously down-sampling the two adjacent frames of target grayscale images, the two adjacent frames of target grayscale images obtained by each sampling are constructed into an image pyramid; wherein, the bottom layer of the image pyramid is the two adjacent frames The frame target grayscale image, the highest layer is two adjacent frames of target grayscale images sampled at the lowest sampling rate; Substituting the pixel position offset of the same particle in the highest layer image in the image pyramid, and the interval between two adjacent frames of target grayscale images in the highest layer image into the sparse optical flow algorithm, the highest image pyramid is obtained. The first optical flow velocity of the layer image; Iteratively calculate the optical flow velocity of each layer image of the image pyramid based on the first optical flow velocity until the optical flow velocity of the bottom image is obtained, and the iteration is completed, and the optical flow velocity of the bottom image is used as the target Optical flow velocity. 5. The flow velocity detection method according to claim 1, characterized in that, before scaling the target optical flow velocity to obtain the target flow velocity of the liquid according to the scaling ratio corresponding to the video stream, further include: Calculate the linear relationship between the flow distance of the liquid within the interval time and the displacement distance of the pixel point within the interval time, and use the linear relationship as the size scaling ratio. 6. The flow velocity detection method according to claim 5, wherein the two adjacent frames of target grayscale images are multiple groups, and multiple groups of adjacent two frames of target grayscale images correspond to obtain a plurality of target optical flow velocities; The step of scaling the target optical flow velocity to obtain the target flow velocity of the liquid according to the scaling ratio corresponding to the video stream includes: determining an optical flow velocity according to a plurality of target optical flow velocities; converting the optical flow velocity into a target flow velocity of the liquid according to the linear relationship. 7. The flow velocity detection method according to claim 5, wherein the two adjacent frames of target grayscale images are multiple groups, and multiple groups of adjacent two frames of target grayscale images correspond to obtain a plurality of target optical flow velocities; The step of scaling the target optical flow velocity to obtain the target flow velocity of the liquid according to the scaling ratio corresponding to the video stream includes: converting a plurality of the target optical flow velocities into a plurality of flow velocities of the liquid according to the linear relationship; A target flow rate of the liquid is determined according to a plurality of flow rates of the liquid. 8. A liquid flow rate detection device, characterized in that, comprising: An acquisition unit, configured to acquire a video stream for capturing a flowing liquid, and convert multiple frames of video images in the video stream into multiple frames of grayscale images; The deletion unit is used to determine the target particles that do not meet the preset difference conditions according to the pixel position offset of the same particle in the liquid between any two adjacent frames of grayscale images, and the target particles in the multi-frame corresponding pixels in the grayscale image to obtain multiple frames of target grayscale images; A determining unit, configured to determine the pixel point offset of the same particle in two adjacent frames of the target gray-scale image in the multi-frame target gray-scale image, and determine the interval between the two adjacent frames of the target gray-scale image; execute A unit for substituting the pixel position offset of the same particle and the interval duration into the sparse optical flow algorithm to obtain the target optical flow velocity; The processing unit is configured to perform scaling processing on the target optical flow velocity to obtain the target flow velocity of the liquid according to the scaling ratio corresponding to the video stream. 9. A liquid flow rate detection device, characterized in that it comprises: Central processing unit, memory, input and output interface, wired or wireless network interface, power supply; The memory is a temporary storage memory or a persistent storage memory; The central processing unit is configured to communicate with the memory, and execute instructions in the memory on the control plane functional entity to execute the flow velocity detection method described in any one of claims 1-7. 10. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises instructions, and when the instructions are run on a computer, the computer is made to execute the flow velocity detection method according to claims 1-7.

Images