Liquid flow measurement method and apparatus based on capillary interface optical tracking

Abstract

This application relates to the field of micro / nano-scale flow measurement technology, and discloses a liquid flow measurement method and apparatus based on capillary interface optical tracking. The method includes: determining edge point cloud data of the gas-liquid meniscus in a capillary during a target detection period; performing circular fitting and geometric parameter calculation based on the edge point cloud data to obtain contact data of the gas-liquid meniscus; wherein the contact data includes the contact angle and the contact position between the gas-liquid meniscus and the inner wall of the capillary; and performing flow correction processing based on the contact data to determine the corrected volumetric flow rate during the target detection period. Its beneficial effect is that by using circular fitting and geometric parameter calculation to generate contact data including the contact angle and contact position, the deformation state of the meniscus is quantified, effectively eliminating the volume measurement deviation caused by the meniscus cap deformation, and significantly improving the accuracy and reliability of liquid flow measurement under ultra-low flow and dynamic changing conditions.

Description

Technical Field

[0001] This application relates to the field of micro-nano scale flow measurement technology, and in particular to a liquid flow measurement method and apparatus based on capillary interface optical tracking. Background Technology

[0002] Ultra-microfluid flow measurement is of great significance in insulin pumps, microinfusion pumps, microfluidic systems, and related flow calibration devices. For these devices, flow rates are typically low, measurement times are long, and high accuracy in cumulative infusion volume and instantaneous flow rate is required. Therefore, high-resolution and high-accuracy flow measurement methods are needed.

[0003] In related technologies, the capillary interface (gas-liquid meniscus) displacement method is commonly used to measure the volumetric flow rate of ultrafine liquids. This method typically involves introducing liquid into a transparent capillary, recording the positional changes of the gas-liquid interface within the capillary using an image acquisition system, and calculating the liquid volume change by multiplying the axial displacement of the gas-liquid interface by the capillary cross-sectional area, thereby calculating the flow rate. This method is simple in structure, highly visualized, and suitable for continuous measurement in low flow ranges, and is therefore widely used in ultrafine liquid flow measurement and calibration scenarios.

[0004] However, this method has limitations in terms of low calculation accuracy because it ignores the influence of the meniscus. Therefore, a liquid flow measurement method that can achieve high-precision detection is needed. Summary of the Invention

[0005] This application aims to at least partially solve one of the technical problems in related technologies. To this end, this application proposes a liquid flow measurement method and apparatus based on capillary interface optical tracking. The main technical solutions adopted in this application include: In a first aspect, this application provides a liquid flow measurement method based on capillary interface optical tracking. The method includes: determining edge point cloud data of the gas-liquid meniscus in the capillary during the target detection period; performing circular fitting and geometric parameter calculation processing based on the edge point cloud data to obtain contact data of the gas-liquid meniscus; wherein, the contact data includes the contact angle and the contact position between the gas-liquid meniscus and the inner wall of the capillary; and performing flow correction processing based on the contact data to determine the corrected volumetric flow rate during the target detection period.

[0006] First, edge point cloud data of the meniscus is acquired to accurately reflect its real-time morphological characteristics. Then, circular fitting and geometric parameter calculations are used to generate current contact data including contact angle and contact position, thereby quantifying the deformation state of the meniscus. Finally, volume compensation calculations are performed based on the current contact data, effectively eliminating the volume measurement deviation caused by the meniscus's spherical deformation, and significantly improving the accuracy and reliability of liquid flow measurement under ultra-low flow and dynamic changing conditions.

[0007] Optionally, circular fitting and geometric parameter calculation are performed based on edge point cloud data to obtain contact data of the gas-liquid meniscus, including: performing a fitting operation on the edge point cloud data to generate fitting circular parameters of the gas-liquid meniscus; wherein, the fitting circular parameters include fitting circle center data and fitting circle radius data; and performing geometric analysis based on the fitting circular parameters and capillary structure parameters to obtain contact data.

[0008] First, a least-squares circle fitting operation is performed on the edge point cloud data, transforming discrete edge points into continuous center and radius parameters, thus achieving a geometric parameterized representation of the meniscus profile and laying a stable data foundation for subsequent geometric analysis. Then, geometric analysis is performed based on the fitted circular parameters and capillary structure parameters, ultimately generating accurate current contact data to effectively support subsequent volume compensation calculations. The entire method fully utilizes optical image information, avoids the simplistic assumptions about the meniscus morphology found in traditional methods, and significantly improves adaptability to ultra-low flow and dynamically changing scenarios.

[0009] Optionally, geometric analysis is performed based on the fitted circular parameters and capillary structure parameters to obtain contact data, including: morphological discrimination based on the fitted circle center data to obtain the concavity / convexity identification result of the gas-liquid meniscus; and contact calculation is performed based on the fitted circular parameters, the concavity / convexity identification result, and the capillary structure parameters to obtain contact data.

[0010] By combining the results of concavity and convexity recognition to calculate the contact angle in segments, the corresponding calculation formula can be determined according to the real-time orientation of the meniscus, thereby accurately obtaining the value of the contact angle and obtaining accurate volume compensation calculation results.

[0011] Optionally, morphological discrimination is performed based on the fitted circle center data to obtain the concavity / convexity recognition result of the gas-liquid meniscus, including: calculating the point cloud statistics of the gas-liquid meniscus based on the edge point cloud data; comparing the fitted circle center data with the point cloud statistics to determine the concavity / convexity shape, thereby obtaining the concavity / convexity recognition result of the gas-liquid meniscus.

[0012] The concave-convex shape is determined by calculating the mean of the abscissa of the edge point cloud and comparing it with the abscissa of the center of the fitted circle. This fully utilizes the characteristic that the axial shape difference of the meniscus is the most significant, thus directly using the essential characteristics of the axial distribution of the meniscus to achieve fast and accurate shape recognition. There is no need for complex image segmentation or feature extraction operations. It has high computational efficiency and strong anti-interference ability.

[0013] Optionally, contact calculation is performed based on the fitted circle parameters, the concavity / convexity recognition results, and the capillary structure parameters to obtain contact data, including: performing geometric auxiliary calculation based on the fitted circle radius and the capillary structure parameters to obtain the geometric auxiliary angle; calculating the contact angle based on the geometric auxiliary angle and the concavity / convexity recognition results; and calculating the contact position based on the fitted circle parameters, the concavity / convexity recognition results, and the geometric auxiliary angle.

[0014] By introducing geometric auxiliary angles, complex geometric relationships can be decomposed into simple trigonometric function calculations, which significantly simplifies the derivation process of the current contact angle and the current contact position and improves calculation efficiency.

[0015] Optionally, the contact position is calculated based on the fitted circle parameters, the concavity / convexity recognition results, and the geometric auxiliary angle, including: determining the axial offset distance of the gas-liquid meniscus according to the fitted circle radius data and the geometric auxiliary angle; wherein, the axial offset distance refers to the axial distance between the contact point between the meniscus and the inner wall of the capillary relative to the center of the fitted circle; and performing contact positioning processing based on the axial offset distance, the fitted circle radius, and the concavity / convexity recognition results to generate the contact position.

[0016] By first calculating the axial offset distance and then determining the offset direction based on the concave-convex identification results (adding for convex liquid surfaces and subtracting for concave liquid surfaces), the contact point can be accurately located to the left or right of the fitted circle center, thereby obtaining the true axial coordinates of the contact section between the meniscus and the pipe wall, providing an accurate positional reference for subsequent volume compensation.

[0017] Optionally, the target detection period has a detection time interval; the gas-liquid meniscus is characterized by a spherical cap model; flow correction processing is performed based on contact data to determine the corrected volumetric flow rate for the target detection period, including: performing spherical cap geometry calculation based on geometric auxiliary angles to obtain spherical cap parameters; wherein, the spherical cap parameters include spherical cap height and spherical cap volume; and determining the corrected volumetric flow rate based on the detection time interval, spherical cap parameters, and contact position.

[0018] By employing a spherical cap model to accurately quantify meniscus deformation, the curvature change of the meniscus can be converted into a calculable volume, enabling frame-by-frame compensation for meniscus deformation. This compensation method eliminates the need for pre-calibration of contact angle or meniscus morphology, adapting to liquids with varying wettability and different flow states. Especially in ultra-low flow rates (nL / h), during start-up and shutdown phases, and in measurement scenarios with significant wettability changes, it can significantly reduce the relative error caused by meniscus deformation, greatly improving the accuracy and reliability of liquid flow measurement.

[0019] Optionally, the corrected volumetric flow rate is determined based on the detection time interval, the spherical cap parameters, and the contact position, including: calculating the compensated volume data based on the spherical cap parameters, the contact position, and the concavity / convexity identification results; wherein, the compensated volume data refers to the liquid volume after meniscus deformation compensation; and determining the corrected volumetric flow rate based on the compensated volume data and the detection time interval.

[0020] By first calculating the total volume of the liquid in a single frame and then obtaining the compensated volume data through difference calculation, the following two types of volume changes can be considered simultaneously: first, the volume change of the cylinder caused by the displacement of the meniscus contact point; and second, the volume change of the spherical cap caused by the deformation of the meniscus itself. This completely covers all sources of liquid volume change, fundamentally solving the systematic error problem caused by traditional methods that only calculate the volume increment of the cylinder.

[0021] Optionally, determining the edge point cloud data of the gas-liquid meniscus in the capillary during the target detection period includes: acquiring the original image data of the gas-liquid meniscus in the capillary during the target detection period; performing target detection processing based on the original image data to extract the region of interest where the gas-liquid meniscus is located; and performing edge detection processing on the region of interest to obtain the edge point cloud data of the gas-liquid meniscus.

[0022] First, by acquiring raw image data, the real-time dynamic changes of the meniscus can be completely recorded. Then, by extracting the region of interest, the image processing scope can be effectively narrowed, reducing computational load and minimizing irrelevant interference. Finally, edge detection processing can accurately extract the contour information of the meniscus, providing a reliable data foundation for subsequent circular fitting and parameter calculation.

[0023] Secondly, this application provides a liquid flow measurement device based on capillary interface optical tracking. The device includes: a point cloud data acquisition module for determining the edge point cloud data of the gas-liquid meniscus in the capillary during the target detection period; a contact data acquisition module for performing circular fitting and geometric parameter calculation processing based on the edge point cloud data to obtain the contact data of the gas-liquid meniscus; wherein the contact data includes the contact angle and the contact position between the gas-liquid meniscus and the inner wall of the capillary; and a liquid flow correction module for performing flow correction processing based on the contact data to determine the corrected volumetric flow rate during the target detection period. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0025] Figure 1a This is a flowchart of a liquid flow measurement method based on capillary interface optical tracking according to an embodiment of this application; Figure 1b This is a schematic diagram of an optical tracking structure provided according to an embodiment of this application; Figure 1c This is a schematic diagram of a gas-liquid meniscus provided according to an embodiment of this application; Figure 2a This is a flowchart of a method for determining contact data according to an embodiment of this application; Figure 2b This is a schematic diagram of a fitted circle provided according to an embodiment of this application; Figure 2c This is a schematic diagram of a convex liquid surface provided according to an embodiment of this application; Figure 2d This is a schematic diagram of a concave liquid surface according to an embodiment of this application; Figure 3 This is a flowchart of a method for determining a modified volumetric flow rate according to an embodiment of this application; Figure 4 This is a structural block diagram of a liquid flow measurement device based on capillary interface optical tracking according to an embodiment of this application. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] It should be noted that most meniscus displacement methods in related technologies are based on the ideal cylindrical volume model, assuming that the gas-liquid interface only undergoes translation, or that the impact of meniscus morphology changes on volume calculation is negligible. However, in reality, the gas-liquid interface within a capillary is usually not planar, but rather a meniscus with a certain curvature. The morphology of the meniscus is affected by factors such as liquid wettability, the condition of the capillary wall, flow conditions, and dynamic contact angle changes, and undergoes continuous changes during measurement. Especially in ultra-low flow rates (nL / h), during start-up and shutdown phases, and in liquids with significant changes in wettability, the changes in meniscus curvature and contact angle are more pronounced, and the resulting spherical cap volume changes become a non-negligible source of error.

[0028] However, traditional methods do not pay enough attention to this source of error, and in most cases still use the cylindrical approximation of "interface displacement × cross-sectional area" for volume conversion. When the meniscus is a concave or convex liquid surface, if the additional volume introduced by interface deformation is not distinguished and compensated, it will lead to deviations in the calculation of volume increment, thus affecting the accuracy of volumetric flow rate. Especially under ultra-low flow conditions, the interface displacement itself is very small per unit time, while the proportion of volume error caused by meniscus deformation increases, making it difficult for traditional methods to meet the requirements of high-precision measurement. Therefore, it is necessary to propose a method that can combine meniscus image information, dynamic contact angle changes, and spherical cap volume model to perform real-time, arbitrary frame, or frame-by-frame compensation of the traditional capillary displacement method, so as to improve the accuracy and applicability of ultra-low liquid flow measurement.

[0029] Based on this, according to the embodiments of this application, an embodiment of a liquid flow measurement method based on capillary interface optical tracking is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0030] This embodiment provides a liquid flow measurement method based on capillary interface optical tracking, such as... Figure 1a As shown, the method includes the following steps: S110. Determine the edge point cloud data of the gas-liquid meniscus in the capillary during the target detection period.

[0031] The target detection period can refer to a continuous time interval used to calculate the liquid flow rate, such as the time interval between the start sampling time (which can be denoted as the first sampling time) and the end sampling time (which can be denoted as the second sampling time).

[0032] For example, if the time interval between two adjacent image acquisitions is taken as the target detection time period, then the first sampling time can be the acquisition time of the previous frame image, and the second sampling time can be the acquisition time of the current frame image.

[0033] Among them, edge point cloud data can refer to data that can be used to describe the interface contour morphology of the gas-liquid meniscus, and can be a set of points consisting of a series of discrete points located on the light-dark boundary line of the meniscus.

[0034] Specifically, the edge point cloud data of the gas-liquid meniscus in the capillary can be determined in the following way: First, the original image data of the gas-liquid meniscus in the capillary during the target detection period is acquired; then, target detection processing is performed based on the original image data to extract the region of interest where the gas-liquid meniscus is located; then, edge detection processing is performed on the region of interest to obtain the edge point cloud data of the gas-liquid meniscus.

[0035] The original image data can refer to two-dimensional image data that can describe the real-time morphology of the gas-liquid meniscus in the capillary at each sampling moment during the target detection period. The image data acquired at each sampling moment can include the meniscus outline, capillary wall and background information at that sampling moment.

[0036] It should be noted that since the shape of the gas-liquid meniscus changes dynamically with the flow of liquid, the meniscus parameters at different sampling times are different. Therefore, in order to accurately measure the liquid flow rate, the original image data at each sampling time within the time period can be obtained according to a preset sampling frequency (such as sampling once per frame).

[0037] Specifically, raw image data of the gas-liquid meniscus in the capillary can be continuously acquired during the target detection period using an industrial camera or microscopic imaging device.

[0038] In order to obtain a distortion-free and high-resolution image of the meniscus and ensure the accuracy of subsequent parameter calculations, this embodiment can use a specially designed capillary interface optical tracking structure for image acquisition.

[0039] For example, such as Figure 1b As shown, the optical tracking structure includes a high-resolution industrial camera 1, a telecentric lens 2, a parallel light source 3, a capillary tube 4, and a linear displacement platform 5.

[0040] The system comprises a high-resolution industrial camera 1, which continuously captures clear images of the meniscus at preset sampling times, ensuring the recognizability of edge details. A telecentric lens 2 eliminates perspective distortion, ensuring consistent image size of the meniscus at different locations. A parallel light source 3 provides uniform background illumination, enhancing the contrast between the meniscus and the background. A capillary tube 4 serves as a channel for liquid flow and meniscus formation. A linear displacement platform 5 simultaneously supports the capillary tube and the telecentric lens, keeping them relatively stationary and moving synchronously to ensure the meniscus remains within the image's clear range at all times.

[0041] It should be noted that there are no special restrictions on the specifications of the capillary tube or the liquid to be tested.

[0042] For example, the capillary tube used in this embodiment has an inner diameter of less than 2 mm and can be made of transparent materials such as glass. The inner wall of the capillary tube can be surface-treated as needed. The surface treatment includes, but is not limited to, changing the hydrophilicity / hydrophobicity and uniformity of the inner wall through physical methods (such as micro-nano coating technology and plasma cleaning) or chemical methods (such as silanization treatment) to adapt to the measurement requirements of different liquids to be tested.

[0043] The liquid to be tested can preferably be a non-volatile liquid that can form a recognizable capillary interface in a transparent capillary, such as water, oil, insulin solution, and ionic solution.

[0044] Furthermore, after setting up the optical tracking structure, it can first be debugged to ensure that the telecentric lens can clearly capture the meniscus shape and that the capillary tube is kept parallel to the linear displacement platform to achieve follow-up tracking. After debugging, the liquid to be tested can be introduced into the capillary tube to form a stable gas-liquid meniscus. Then, during the target detection period, meniscus images are continuously acquired according to the preset sampling time (e.g., one sample per frame) to form a sequence of image frames arranged in chronological order, which serves as the raw image data.

[0045] Subsequently, target detection processing can be performed on the original image data corresponding to each sampling time to extract the region of interest at each sampling time.

[0046] It should be noted that, for the convenience of subsequent calculations, a coordinate system can be introduced. The lower left corner of the original image data is defined as the origin, the horizontal direction of the image is taken as the x-axis, with the direction from the liquid segment to the gas segment as the positive x-direction; the vertical direction of the image is taken as the y-axis, with the upward direction as the positive y-direction. Under this coordinate system, template matching or convolutional neural networks can be used to perform target detection on the original image data, locate the position of the gas-liquid meniscus in the image, and then extract the corresponding region of interest.

[0047] The region of interest (ROI) can refer to a local image region located at the meniscus interface in the original image data; that is, a region cropped from the complete original image data, retaining only the shape of the gas-liquid meniscus and its surrounding background. For example, ROI 101 can be as follows: Figure 1c As shown.

[0048] Furthermore, after obtaining the region of interest, edge detection processing can be performed on the region of interest.

[0049] Specifically, in the cropped region of interest, methods such as grayscale thresholding, gradient edge detection (e.g., Sobel or Canny operators), or sub-pixel edge extraction can be used to identify grayscale abrupt changes between the gas-liquid meniscus and the background, thereby obtaining a series of continuous or discrete edge points. The set of coordinates of these edge points constitutes the edge point cloud data of the meniscus in that frame. Repeating the above operation for each frame within the target detection period yields the edge point cloud data corresponding to each sampling time. For example, the edge point set 103 at a certain sampling time can be... Figure 1c As shown, its coordinate values ​​are the edge point cloud data at that sampling time.

[0050] First, by acquiring raw image data at each sampling moment within the target detection period, the real-time dynamic changes of the meniscus during that period can be completely recorded. Then, extracting the region of interest at each sampling moment effectively narrows the image processing scope, reduces computational load, and minimizes irrelevant interference. Finally, edge detection processing accurately extracts the contour information of the meniscus at each sampling moment, providing a reliable data foundation for subsequent circular fitting and parameter calculation.

[0051] S120. Based on the edge point cloud data, perform circular fitting and geometric parameter calculation to obtain the contact data of the gas-liquid meniscus.

[0052] It is understandable that the original image data is an image sequence continuously acquired at a preset sampling frequency during the target detection period, and each image in the sequence corresponds to the real-time shape of the meniscus at a sampling moment. Therefore, for any sampling moment within the target detection period, the corresponding edge point cloud data can be extracted first, and then circular fitting and geometric parameter calculation can be performed on it to finally obtain the contact data of the gas-liquid meniscus at that sampling moment.

[0053] Contact data can refer to relevant geometric parameters that describe the contact state between the meniscus and the inner wall of the capillary, including the contact angle and the contact position between the gas-liquid meniscus and the inner wall of the capillary.

[0054] The contact position can refer to the axial position of the cross section where the gas-liquid meniscus contacts the inner wall of the capillary. It can be represented by the coordinates of the contact point between the meniscus and the inner wall of the capillary in the axial (x-direction).

[0055] The contact angle refers to the angle between the solid-liquid interface and the liquid-gas interface on the liquid side at the contact point of the gas-liquid-solid three phases.

[0056] For example, please continue to refer to Figure 1c It displays the contact angle at a certain sampling time (e.g., denoted as the i-th frame). And the axial position 105 of the section where the contact point P between the gas-liquid meniscus and the inner wall of the capillary is located.

[0057] Specifically, for each sampling moment within the target detection period, the edge point cloud data corresponding to that sampling moment is first processed using the least squares circle fitting method to obtain the center and radius data of the fitted circle at that sampling moment. Based on this, a geometric auxiliary angle can be calculated using the radius of the fitted circle and the inner diameter of the capillary. This geometric auxiliary angle can be the angle between the center of the fitted circle and the contact point P, and a straight line passing through the center of the fitted circle and parallel to the inner wall of the capillary. Combining this geometric auxiliary angle with the concavity / convexity of the meniscus, the contact angle can be calculated. This contact angle can be output independently as an additional result for use in other applications requiring dynamic contact angle parameters. Simultaneously, based on the aforementioned geometric auxiliary angle, combined with the coordinates of the fitted circle's center, radius, and inner diameter of the capillary, the axial offset distance of the meniscus contact point relative to the center of the fitted circle can be calculated. Then, combined with the concavity / convexity recognition result of the meniscus, the offset direction is determined, ultimately obtaining the axial position of the contact point between the meniscus and the inner wall of the capillary at that sampling moment, which is taken as the contact position.

[0058] S130. Perform flow correction processing based on contact data to determine the corrected volumetric flow rate during the target detection period.

[0059] The corrected volumetric flow rate can refer to the volumetric flow rate of liquid passing through the capillary cross-section during the target detection period after meniscus deformation compensation. For example, if the detection time interval between the first sampling time and the second sampling time is taken as the target detection period, then the corrected volumetric flow rate reflects the liquid volume after meniscus deformation correction during that period.

[0060] Specifically, the corrected volumetric flow rate can be calculated using the following steps: First, for each sampling moment within the target detection period, based on the capillary cross-sectional area and the contact data at that sampling moment, calculate the total liquid volume after meniscus deformation compensation at that sampling moment, which is used as the compensated volume data for that sampling moment. Next, select the compensated volume data for two different sampling moments within the target detection period (e.g., the first sampling moment and the second sampling moment), calculate the difference between the two, and obtain the actual liquid volume change within that period. Finally, divide this liquid volume change by the detection time interval between the two sampling moments; the resulting ratio is the corrected volumetric flow rate for the target detection period.

[0061] In the above implementation, firstly, acquiring edge point cloud data of the meniscus accurately reflects its real-time morphological characteristics. Then, using circular fitting and geometric parameter calculation, contact data including contact angle and contact position is generated, thereby quantifying the deformation state of the meniscus. Finally, flow correction processing is performed based on the contact data, effectively eliminating the volume measurement deviation caused by the meniscus's spherical deformation, and significantly improving the accuracy and reliability of liquid flow measurement under ultra-low flow and dynamic changing conditions.

[0062] In some implementation methods, please refer to the appendix. Figure 2a Based on edge point cloud data, circular fitting and geometric parameter calculation are performed to obtain the contact data of the gas-liquid meniscus, including: S210. Perform a fitting operation on the edge point cloud data to generate fitting circular parameters for the gas-liquid meniscus.

[0063] It should be noted that for small-sized capillaries (diameter less than 2 mm), the gas-liquid meniscus morphology is dominated by interfacial tension, and the influence of gravity is negligible. It is approximately an axisymmetric structure, so its two-dimensional profile can be regarded as part of a standard circular arc. By performing circular fitting on the edge point cloud, the discrete point set can be transformed into two geometric parameters: the center coordinates and the radius. Thus, the overall morphology of the meniscus can be described with a concise geometric model.

[0064] Among them, the fitting circle parameters can refer to the relevant geometric parameters that can characterize the equivalent spherical profile of the gas-liquid meniscus, and can include the fitting circle center data and the fitting circle radius data.

[0065] Specifically, the least squares circle fitting method can be used to perform a fitting operation on the edge point cloud data to generate the fitting circular parameters of the gas-liquid meniscus.

[0066] For example, for edge point cloud data at any sampling time (denoted as frame i) within the target detection period, a fitted circle is determined using a least squares circle fitting algorithm, and the center coordinates of the fitted circle in frame i are obtained. (As the center data of the fitted circle) and the radius of the fitted circle in the i-th frame The fitting results can be as follows: Figure 2b As shown in the figure, the radius of the fitted circle 201 in the i-th frame is displayed. Coordinates of the center O .

[0067] By performing the fitting operation sequentially for each sampling moment within the target detection period in the above manner, the fitting circular parameters corresponding to all sampling moments within that period can be obtained.

[0068] S220. Geometric analysis is performed based on the fitted circular parameters and capillary structure parameters to obtain contact data.

[0069] Specifically, geometric analysis can be performed in the following way: First, shape discrimination is performed based on the fitted circle center data to obtain the concavity and convexity identification results of the gas-liquid meniscus; then, contact calculation is performed based on the fitted circle parameters, the concavity and convexity identification results, and the capillary structure parameters to obtain contact data.

[0070] The concavity / convexity recognition result can refer to the determination result that can characterize the orientation of the gas-liquid meniscus at each sampling time, that is, the convex direction of the meniscus relative to the liquid phase body, which can include three types: convex liquid surface, concave liquid surface, and flat liquid surface.

[0071] Specifically, morphology discrimination may include the following steps: first, calculate the point cloud statistics of the gas-liquid meniscus based on the edge point cloud data; then, compare the center data of the fitted circle with the point cloud statistics to determine the concave-convex shape, thereby obtaining the concave-convex recognition result of the gas-liquid meniscus.

[0072] Among them, point cloud statistics can refer to the statistical distribution value of all edge points of the gas-liquid meniscus in a certain direction at the current sampling time.

[0073] Understandably, under the current coordinate system definition, since the difference in the concavity and convexity of the meniscus is mainly reflected in the positional distribution along the axial direction (x direction), and the vertical coordinate only reflects the radial position, it has no significant impact on the determination of the concavity and convexity. Therefore, the mean (or median) of the horizontal coordinates of all edge points at this sampling time can be used as the statistical data of the point cloud.

[0074] For example, taking the average value of the horizontal axis as the point cloud statistics, the point cloud statistics at any sampling time (the i-th frame) can be calculated using the following formula: In the formula, The total number of gas-liquid meniscus edge points contained in the i-th frame; This represents the x-coordinate of the k-th edge point in the i-th frame.

[0075] After obtaining the point cloud statistics, the concavity / convexity shape determination operation can be performed: for any sampling time in the target detection period, the center data of the fitted circle at that sampling time is obtained. Point cloud statistics By comparing the results, the concavity and convexity of the gas-liquid meniscus can be identified.

[0076] For example, taking the direction from the liquid to the gas as the positive x-axis, for any sampling moment within the target detection period: when the center data of the fitted circle at that sampling moment... Greater than point cloud statistics At that time, it is determined that the liquid bulges towards the gas side, and the concavity / convexity recognition result of the gas-liquid meniscus being the convex liquid surface is obtained; when the center data of the fitted circle at that sampling time is... Smaller than point cloud statistics When the gas bulges towards the liquid side, the concavity / convexity recognition result of the gas-liquid meniscus is obtained, indicating that the gas-liquid meniscus is a concave liquid surface.

[0077] Furthermore, for ease of subsequent calculations, interface shape symbol parameters can be defined based on the concavity / convexity recognition results. When the concavity / convexity identification result is a convex liquid surface, define When the concavity / convexity identification result is a concave liquid surface, define... .

[0078] Furthermore, noise interference may occur during image acquisition and edge detection, causing slight fluctuations in point cloud statistics between adjacent sampling times. Therefore, a preset judgment threshold can be set to suppress the impact of noise on shape determination.

[0079] Specifically, for two adjacent sampling times, if the difference between the x-coordinate of the fitted circle's center and the point cloud statistics at the current sampling time and the difference between the x-coordinate of the fitted circle's center and the point cloud statistics at the previous sampling time are less than a preset judgment threshold, it means that the meniscus shape changes very little or is in a stable state at the two adjacent sampling times. To avoid frequent switching of concavity and convexity recognition results causing computational instability, the concavity and convexity recognition result at the previous sampling time can be used as the concavity and convexity recognition result at the current sampling time, thereby improving the robustness of the method.

[0080] The concave-convex shape is determined by calculating the mean of the abscissa of the edge point cloud and comparing it with the abscissa of the center of the fitted circle. This fully utilizes the characteristic that the axial shape difference of the meniscus is the most significant, thus directly using the essential characteristics of the axial distribution of the meniscus to achieve fast and accurate shape recognition. There is no need for complex image segmentation or feature extraction operations. It has high computational efficiency and strong anti-interference ability.

[0081] Furthermore, after obtaining the concave-convex recognition results, they can be combined with the fitted circular parameters and capillary structure parameters to perform contact calculations to obtain contact data.

[0082] Here, capillary structural parameters can refer to fixed parameters that characterize the geometric dimensions and structural features of a capillary. Specifically, these capillary structural parameters may include the capillary inner diameter, such as... Figure 2b As shown in the figure, the inner diameter D of the capillary can be used as a structural parameter of the capillary.

[0083] Specifically, contact calculation can be performed in the following ways: First, geometric auxiliary calculations are performed based on the fitted circle radius and capillary structure parameters to obtain the geometric auxiliary angle; then, the contact angle can be calculated based on the geometric auxiliary angle and the concavity / convexity recognition results; at the same time, the contact position can also be calculated based on the fitted circle parameters, the concavity / convexity recognition results, and the geometric auxiliary angle.

[0084] It should be noted that since the meniscus is approximately a spherical cap due to surface tension inside the capillary, there is a clear geometric relationship between its radius of curvature, dynamic contact angle, and inner diameter of the capillary. In order to facilitate the calculation of the contact angle and contact position, a geometric auxiliary angle can be introduced as an auxiliary parameter.

[0085] For example, such as Figure 2b As shown, it displays geometric auxiliary angles that characterize the curvature of the gas-liquid meniscus at the current sampling time. .

[0086] Specifically, the geometric auxiliary angle at any sampling time (denoted as the i-th frame) during the target detection period can be calculated using the arcsine function: In the formula, represents the fitted circle radius data for the i-th frame; D represents the capillary structure parameters.

[0087] By introducing geometric auxiliary angles, complex geometric relationships can be decomposed into simple trigonometric function calculations, which significantly simplifies the derivation process of the current contact angle and the current contact position and improves calculation efficiency.

[0088] Furthermore, in determining the geometric auxiliary angle Then, the axial position of the cross section where the gas-liquid meniscus contacts the inner wall of the capillary can be further determined, i.e., the contact position.

[0089] Specifically, the contact position can be calculated as follows: First, determine the axial offset distance of the gas-liquid meniscus based on the fitted circle radius data and geometric auxiliary angles; then, perform contact positioning processing based on the axial offset distance, fitted circle radius, and concavity / convexity recognition results to generate the contact position.

[0090] Here, the axial offset distance can refer to the axial distance between the contact point between the meniscus and the inner wall of the capillary relative to the center of the fitted circle, i.e. Figure 2c and Figure 2d In .

[0091] For example, the axial offset distance of the gas-liquid meniscus at any sampling time (denoted as the i-th frame) during the target detection period can be calculated using the Pythagorean theorem and trigonometric functions, by the following formula. : In the formula, The fitted circle radius data represents the i-th frame; D represents the capillary structure parameters. Represents the geometric auxiliary angle of the i-th frame.

[0092] Subsequently, contact positioning can be performed using the following formula to obtain the contact position at the sampling moment. That is, the axial x-coordinate value of the contact point between the meniscus and the inner wall of the capillary: In the formula, The data representing the center of the fitted circle in the i-th frame is the x-coordinate of the fitted circle's center. Represents the concavity / convexity recognition result of the i-th frame; This represents the axial offset distance of the i-th frame.

[0093] By first calculating the axial offset distance and then determining the offset direction based on the concave-convex identification results (adding for convex liquid surfaces and subtracting for concave liquid surfaces), the contact point can be accurately located to the left or right of the fitted circle center, thereby obtaining the true axial coordinates of the contact section between the meniscus and the pipe wall, providing an accurate positional reference for subsequent volume compensation.

[0094] Understandably, although the contact angle is not involved in the calculation of the contact position in this step, it can still be calculated using the geometric auxiliary angle and the concavity / convexity recognition results, and output as an independent result for use in other application scenarios that require the contact angle.

[0095] After determining the geometric auxiliary angle at that sampling moment, the contact angle at that sampling moment can then be calculated. .

[0096] It should be noted that when the meniscus is approximately planar, the deformation volume is zero, and there is no need to perform contact angle-related compensation calculations. Therefore, it can be handled separately in subsequent flow compensation. The flat liquid surface case will not be derived in detail here.

[0097] For example, when the concavity / convexity identification result is a convex liquid surface, please refer to... Figure 2c Angle A1 and geometric auxiliary angle Since angles A1 and B1 are vertical angles, angle B1 is also equal to the geometric auxiliary angle. They are equal. Furthermore, angles B1 and C are complementary angles, and angle C is equal to the current contact angle. Since they are complementary angles, the contact angle for a convex liquid surface can be derived from this. for .

[0098] When the concavity / convexity recognition result in the i-th frame is a concave liquid surface, please refer to... Figure 2d Angle A2 and geometric auxiliary angle Since angles A2 and B2 are vertically opposite, angle B2 is also equal to the geometric auxiliary angle. They are equal. And angle B2 is equal to the current contact angle. Since they are complementary angles, the contact angle for a concave liquid surface can be derived from this. for .

[0099] That is, the contact angle of the i-th frame. This can be expressed by the following formula: In the formula, This represents the concavity / convexity recognition result of the i-th frame.

[0100] By combining the results of concavity and convexity recognition to calculate the contact angle in segments, the corresponding calculation formula can be determined according to the real-time orientation of the meniscus, thereby accurately obtaining the value of the contact angle and obtaining accurate volume compensation calculation results.

[0101] Meanwhile, the segmented calculation method naturally adapts to the continuous or abrupt transitions between the convex and concave shapes of the meniscus. Even if there are large fluctuations in the contact angle, this method can still output accurate contact angle values ​​stably and will not cause calculation abnormalities or decrease in accuracy due to shape switching.

[0102] Furthermore, the process relies solely on the radius of the fitted circle and the concavity / convexity identification results at that sampling moment, without depending on historical accumulated information. This avoids error propagation and drift issues, further enhancing the reliability and robustness in ultra-low flow measurement scenarios.

[0103] In the above implementation, a least-squares circle fitting operation is first performed on the edge point cloud data, which transforms discrete edge points into continuous circle center and radius parameters, realizing the geometric parameterization of the meniscus contour and laying a stable data foundation for subsequent geometric analysis. Then, geometric analysis is performed based on the fitted circle parameters and capillary structure parameters, ultimately generating accurate current contact data, effectively supporting subsequent volume compensation calculations. The entire method fully utilizes optical image information, avoids the simplistic assumptions about the meniscus morphology in traditional methods, and significantly improves adaptability to ultra-low flow and dynamically changing scenarios.

[0104] In some implementations, the target detection period has a detection time interval; the gas-liquid meniscus is characterized by a spherical cap model; please refer to the appendix. Figure 3 Based on contact data, flow correction processing is performed to determine the corrected volumetric flow rate for the target detection period, including: S310. Perform geometric calculations on the spherical cap based on the geometric auxiliary angle to obtain the spherical cap parameters.

[0105] It is understandable that, similar to the approximate circular arc profile of the meniscus inside a small-sized capillary, due to the dominant effect of surface tension, the meniscus in three-dimensional space presents a spherical cap shape that is rotationally symmetrical about the capillary axis. Therefore, a spherical cap model can be used to describe its geometry, and the three-dimensional meniscus can be equivalent to a spherical cap model, whose two-dimensional cross-section is the circular arc part obtained by fitting a circle.

[0106] Based on this spherical cap model, the parameters of the spherical cap that characterize the degree of meniscus deformation can be obtained through geometric calculations.

[0107] The spherical cap parameter can be a geometric quantity used to quantify the degree of meniscus deformation; that is, a value extracted from the spherical cap model that reflects the extent of the meniscus's bulge or depression and the size of the space it occupies. For example, please refer to... Figure 2c and Figure 2d The parameters of the sphere can include the height of the sphere. and the volume of the spherical crown (All images are presented in two dimensions).

[0108] Specifically, since the arc of the fitted circle corresponds to the boundary of the spherical cap, and there is a clear trigonometric function relationship between the height of the spherical cap and the geometric auxiliary angle, the height of the spherical cap can be calculated using the cosine value of the geometric auxiliary angle and the radius data of the fitted circle when solving for the parameters of the spherical cap.

[0109] For example, for any sampling time within the target detection period (denoted as the i-th frame), its spherical crown height It can be calculated using the following formula: In the formula, Represents the geometric auxiliary angle of the i-th frame; This represents the fitted circle radius data for the i-th frame.

[0110] Furthermore, the volume of the spherical cap at that sampling moment can be calculated using pi, the radius data of the fitted circle, and the height of the spherical cap.

[0111] For example, the volume of the spherical cap in the i-th frame It can be calculated using the following formula: S320, Determine the corrected volumetric flow rate based on the detection time interval, spherical cap parameters, and contact position.

[0112] It is understood that the target detection period has a detection time interval, that is, the target detection period consists of multiple sampling moments collected at a preset sampling frequency, and the time length between any two sampling moments is the detection time interval. For example, if the target detection period is the time interval between the first sampling moment and the second sampling moment, the detection time interval is the time difference between the second sampling moment and the first sampling moment.

[0113] Specifically, after obtaining the spherical cap parameters at each sampling time, the corrected volumetric flow rate can be determined as follows: First, calculate the compensated volumetric data based on the spherical cap parameters, contact position, and concavity / convexity identification results; then, determine the corrected volumetric flow rate based on the compensated volumetric data and the detection time interval.

[0114] The compensated volume data refers to the total liquid volume after meniscus deformation compensation at any sampling time during the target detection period. It is the actual total liquid volume value after superimposing or reducing the volume compensation amount brought about by meniscus deformation on the basis of the traditional cylindrical model (contact position multiplied by capillary cross-sectional area). It combines the contributions of both meniscus deformation and axial displacement.

[0115] Specifically, by using the volume of the spherical cap, the current contact position, and the concavity / convexity recognition results, the compensated volume data at any sampling time (denoted as the i-th frame) can be determined.

[0116] For example, the compensated volume data of the i-th frame The following formula can be used for calculation: In the formula, Represents the volume of the spherical cap in the i-th frame; The contact position of the i-th frame is represented by the axial x-coordinate of the point of contact between the meniscus and the inner wall of the capillary. Represents the concavity / convexity recognition result of the i-th frame; A represents the capillary cross-sectional area, and has .

[0117] Furthermore, after obtaining the compensated volume data at a single sampling time, the compensated volume difference at the detection time interval can be calculated using the detection time interval between any two sampling times and the compensated volume data at the corresponding sampling time. Finally, the corrected volumetric flow rate can be calculated based on this compensated volume difference.

[0118] For example, for any two sampling times (denoted as frame j and frame i) within the target detection period, the compensated volume difference over the detection time interval can be calculated using the following formula: In the formula, This represents the compensated volume difference between frame j and frame i. Represents the volume of the spherical cap in the i-th frame; Represents the contact position in the i-th frame; Represents the concavity / convexity recognition result of the i-th frame; Represents the volume of the spherical cap in the j-th frame; Represents the contact position in the j-th frame; This represents the concavity / convexity recognition result of the j-th frame.

[0119] Furthermore, assuming that the two sampling times mentioned above are adjacent and that their gas-liquid menisci maintain the same shape, i.e., the concavity and convexity recognition results of the j-th frame and the i-th frame are the same, then the calculation formula for the compensated volume difference under this detection time interval can be further simplified.

[0120] Specifically, when the concavity / convexity recognition results in both the j-th and i-th frames are convex liquid surfaces, since Then we have the following formula: In the formula, This represents the change in contact position between frame j and frame i, directly reflecting the translational distance between the liquid columns; This represents the change in the volume of the spherical cap between frame j and frame i.

[0121] When the concave / convexity recognition results in both the j-th and i-th frames are concave liquid surfaces, due to Then we have the following formula: After obtaining the compensated volume difference value at the detection time interval, the corrected volumetric flow rate can be determined using the compensated volume difference value and the detection time interval.

[0122] It should be noted that the compensated volume difference already includes the contributions of both the change in contact position and the change in the volume of the spherical cap. Therefore, in actual calculations, the compensated volume difference can be directly divided by the detection time interval to obtain the final flow rate.

[0123] For example, for any two sampling times during the target detection period (denoted as frame j and frame i), the corrected volumetric flow rate can be calculated using the following formula: In the formula, Represents the corrected volumetric flow rate; This represents the compensated volume difference between frame j and frame i. This represents the detection time interval.

[0124] By first calculating the total volume of the liquid in a single frame and then obtaining the compensated volume data through difference calculation, we can simultaneously consider the following two types of volume changes: first, the volume change of the liquid column (approximately a cylinder) caused by the displacement of the meniscus contact point; and second, the volume change of the spherical cap caused by the deformation of the meniscus itself. This completely covers all sources of liquid volume change and fundamentally solves the systematic error problem caused by traditional methods that only calculate the volume increment of the cylinder.

[0125] In the above implementation, by employing a spherical cap model to accurately quantify the meniscus deformation, the curvature change of the meniscus can be converted into a calculable volume, achieving frame-by-frame compensation for the meniscus deformation. This compensation method does not require pre-calibration of the contact angle or meniscus morphology and can adapt to liquids with different wettability and different flow states. Especially in ultra-low flow rates at the nL / h level, during start-up and shutdown phases, and in measurement scenarios with significant changes in wettability, it can significantly reduce the relative error caused by meniscus deformation and greatly improve the accuracy and reliability of liquid flow measurement.

[0126] It should be understood that although the steps in the flowchart above are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart above may include multiple steps or stages, which are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps.

[0127] This specification also provides a liquid flow measurement device 400 based on capillary interface optical tracking, such as... Figure 4 As shown, it includes: a point cloud data acquisition module 410, a contact data acquisition module 420, and a liquid flow calculation module 430, wherein: The point cloud data acquisition module 410 is used to determine the edge point cloud data of the gas-liquid meniscus in the capillary during the target detection period.

[0128] The contact data acquisition module 420 is used to perform circular fitting and geometric parameter calculation based on edge point cloud data to obtain the contact data of the gas-liquid meniscus; wherein, the contact data includes the contact angle and the contact position between the gas-liquid meniscus and the inner wall of the capillary.

[0129] The liquid flow calculation module 430 is used to perform flow correction processing based on contact data to determine the corrected volumetric flow rate during the target detection period.

[0130] In some embodiments, the contact data acquisition module 420 is further configured to perform a fitting operation on the edge point cloud data to generate fitting circular parameters for the gas-liquid meniscus; wherein the fitting circular parameters include fitting circle center data and fitting circle radius data; and perform geometric analysis processing based on the fitting circular parameters and capillary structure parameters to obtain contact data.

[0131] In some implementations, the contact data acquisition module 420 is also used to perform morphological discrimination based on the fitted circle center data to obtain the concavity and convexity identification results of the gas-liquid meniscus; and to perform contact calculation based on the fitted circle parameters, the concavity and convexity identification results and the capillary structure parameters to obtain contact data.

[0132] In some implementations, the contact data acquisition module 420 is also used to calculate the point cloud statistics of the gas-liquid meniscus based on the edge point cloud data; compare the fitted circle center data with the point cloud statistics to determine the concave-convex shape, thereby obtaining the concave-convex recognition result of the gas-liquid meniscus.

[0133] In some embodiments, the contact data acquisition module 420 is further configured to perform geometric auxiliary calculations based on the fitted circle radius and capillary structure parameters to obtain a geometric auxiliary angle; calculate the contact angle based on the geometric auxiliary angle and the concavity / convexity recognition results; and calculate the contact position based on the fitted circle parameters, the concavity / convexity recognition results, and the geometric auxiliary angle.

[0134] In some embodiments, the contact data acquisition module 420 is further configured to determine the axial offset distance of the gas-liquid meniscus based on the fitted circle radius data and the geometric auxiliary angle; wherein, the axial offset distance refers to the axial distance between the contact point between the meniscus and the inner wall of the capillary relative to the center of the fitted circle; and to perform contact positioning processing based on the axial offset distance, the fitted circle radius and the concavity / convexity identification results to generate the contact position.

[0135] In some implementations, the target detection period has a detection time interval; the gas-liquid meniscus is characterized by a spherical cap model; the liquid flow calculation module 430 is also used to perform spherical cap geometry calculation based on the geometric auxiliary angle to obtain spherical cap parameters; wherein, the spherical cap parameters include spherical cap height and spherical cap volume; the corrected volumetric flow rate is determined based on the detection time interval, spherical cap parameters and contact position.

[0136] In some implementations, the liquid flow calculation module 430 is also used to calculate the compensated volume data based on the spherical cap parameters, contact position, and concavity / convexity identification results; wherein, the compensated volume data refers to the liquid volume after meniscus deformation compensation; and the corrected volume flow rate is determined based on the compensated volume data and the detection time interval.

[0137] In some implementations, the point cloud data acquisition module 410 is further used to acquire the original image data of the gas-liquid meniscus in the capillary during the target detection period; perform target detection processing based on the original image data to extract the region of interest where the gas-liquid meniscus is located; and perform edge detection processing on the region of interest to obtain the edge point cloud data of the gas-liquid meniscus.

[0138] For specific limitations regarding a liquid flow measurement device based on capillary interface optical tracking, please refer to the limitations of a liquid flow measurement method based on capillary interface optical tracking mentioned above, which will not be repeated here. Each module in the aforementioned liquid flow measurement device based on capillary interface optical tracking can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0139] In this embodiment, a liquid flow measurement device based on capillary interface optical tracking is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0140] The apparatus, module, or unit described in the above embodiments can be implemented by a computer chip or entity, or by a product having a certain function. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0141] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0142] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0143] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus, and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0144] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0145] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

[0146] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0147] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the apparatus embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0148] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

[0149] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A liquid flow measurement method based on capillary interface optical tracking, characterized in that, The method includes: Determine the edge point cloud data of the gas-liquid meniscus in the capillary during the target detection period; A fitting operation is performed on the edge point cloud data to generate fitting circular parameters for the gas-liquid meniscus; wherein, the fitting circular parameters include fitting circle center data and fitting circle radius data; Based on the fitted circle center data, morphological discrimination is performed to obtain the concavity and convexity recognition results of the gas-liquid meniscus; Geometric auxiliary calculations are performed based on the radius of the fitted circle and the capillary structure parameters to obtain the geometric auxiliary angle; wherein, the geometric auxiliary angle is the angle between the center of the fitted circle and the contact point and the straight line that is overfitted to the center of the fitted circle and parallel to the inner wall of the capillary. The contact angle is calculated based on the geometric auxiliary angle and the concavity / convexity recognition result; wherein, the contact angle refers to the angle between the solid-liquid interface and the liquid-gas interface on the liquid side at the gas-liquid-solid three-phase contact point. Based on the fitted circular parameters, the concave-convex recognition results, and the geometric auxiliary angle, the contact position between the gas-liquid meniscus and the inner wall of the capillary is calculated; wherein, the contact position and the contact angle constitute contact data; The flow correction process is performed based on the contact data to determine the corrected volumetric flow rate for the target detection period. This includes: calculating the compensated volume data based on the capillary cross-sectional area and the contact data; calculating the liquid volume change based on the compensated volume data at two different sampling times within the target detection period; and dividing the liquid volume change by the detection time interval between the two sampling times to obtain the corrected volumetric flow rate for the target detection period.

2. The method according to claim 1, characterized in that, The step of performing morphological discrimination based on the fitted circle center data to obtain the concavity / convexity recognition result of the gas-liquid meniscus includes: Calculate the point cloud statistics of the gas-liquid meniscus based on the edge point cloud data; The center data of the fitted circle is compared with the statistical data of the point cloud to determine the concave and convex shape, thereby obtaining the concave and convex recognition result of the gas-liquid meniscus.

3. The method according to claim 1, characterized in that, The step of calculating the contact position based on the fitted circular parameters, the concavity / convexity recognition result, and the geometric auxiliary angle includes: The axial offset distance of the gas-liquid meniscus is determined based on the fitted circle radius data and the geometric auxiliary angle; wherein, the axial offset distance refers to the axial distance between the contact point between the meniscus and the inner wall of the capillary relative to the center of the fitted circle; The contact positioning process is performed based on the axial offset distance, the fitted circle radius, and the concavity / convexity recognition result to generate the contact position.

4. The method according to claim 1, characterized in that, The target detection period has a detection time interval; the gas-liquid meniscus is characterized by a spherical cap model; the flow correction processing based on the contact data to determine the corrected volumetric flow rate for the target detection period includes: Geometric calculations of the spherical cap are performed based on geometric auxiliary angles to obtain the spherical cap parameters; wherein, the spherical cap parameters include the height and volume of the spherical cap. The corrected volumetric flow rate is determined based on the detection time interval, the spherical cap parameters, and the contact position.

5. The method according to claim 4, characterized in that, Determining the corrected volumetric flow rate based on the detection time interval, the spherical cap parameters, and the contact position includes: The compensated volume data is calculated based on the spherical cap parameters, the contact position, and the concavity / convexity recognition results; wherein, the compensated volume data refers to the liquid volume after meniscus deformation compensation; The corrected volumetric flow rate is determined based on the compensated volumetric data and the detection time interval.

6. The method according to claim 1, characterized in that, The determined edge point cloud data of the gas-liquid meniscus in the capillary during the target detection period includes: Obtain the original image data of the gas-liquid meniscus in the capillary during the target detection period; Target detection processing is performed based on the original image data to extract the region of interest where the gas-liquid meniscus is located; Edge detection processing is performed on the region of interest to obtain edge point cloud data of the gas-liquid meniscus.

7. A liquid flow measurement device based on capillary interface optical tracking, characterized in that, The device includes: The point cloud data acquisition module is used to determine the edge point cloud data of the gas-liquid meniscus in the capillary during the target detection period; A contact data acquisition module is used to perform a fitting operation on the edge point cloud data to generate fitting circular parameters for the gas-liquid meniscus; wherein, the fitting circular parameters include fitting circle center data and fitting circle radius data; morphological discrimination is performed based on the fitting circle center data to obtain the concavity / convexity recognition result of the gas-liquid meniscus; geometric auxiliary calculation is performed based on the fitting circle radius and the capillary structure parameters to obtain a geometric auxiliary angle; wherein, the geometric auxiliary angle is the angle between the fitting circle center and the contact point and a straight line passing through the fitting circle center and parallel to the inner wall of the capillary; the contact angle is calculated based on the geometric auxiliary angle and the concavity / convexity recognition result; wherein, the contact angle is the angle between the solid-liquid interface and the liquid-gas interface on the liquid phase side at the gas-liquid-solid three-phase contact point; the contact position between the gas-liquid meniscus and the inner wall of the capillary is calculated based on the fitting circular parameters, the concavity / convexity recognition result, and the geometric auxiliary angle; wherein, the contact position and the contact angle constitute contact data; A liquid flow correction module is used to perform flow correction processing based on the contact data to determine the corrected volumetric flow rate for the target detection period; including: calculating the compensated volume data based on the capillary cross-sectional area and the contact data; calculating the liquid volume change based on the compensated volume data at two different sampling times within the target detection period; and dividing the liquid volume change by the detection time interval between the two sampling times to obtain the corrected volumetric flow rate for the target detection period.

Images