CN113740207A - Machine vision-based viscosity detection system and method - Google Patents

Abstract

The invention discloses a viscosity detection system and method based on machine vision, which can solve the problems of high labor intensity, low efficiency and the like in manual judgment and timing in the process of capillary viscometer verification and kinematic viscosity measurement. Aiming at the verification process and the kinematic viscosity measurement process of a flatfish viscometer, an automatic viscosity detection system based on machine vision is established, a dynamic image of fluid is analyzed into continuous static pictures through a high-speed camera, a timing starting point and a timing end point are identified through a high-precision positioning template matching mode, collected starting point and end point pictures are processed, and automatic measurement is realized through gray processing, denoising, binarization processing and identification. The scheme provided by the invention improves the verification efficiency and kinematic viscosity measurement efficiency of the viscometer, and has better application prospects in the aspects of quantity value transmission of a metering technical mechanism and viscosity measurement of an oil product inspection mechanism.

Description

Machine vision-based viscosity detection system and method

Technical Field

The invention relates to the technical field of viscosity detection, in particular to a system and a method for detecting viscosity based on machine vision.

Background

Viscosity is one of the important thermophysical parameters of fluid, and the size of the viscosity directly determines the fluidity of the fluid. The fluid with high viscosity has high internal friction force, and the fluid with lower viscosity needs more power when the same flow rate is generated. The viscosity is accurately measured, and the prediction is used for guiding the production process control and the logistics transportation of petroleum products and high polymer materials; determining the working conditions of an internal combustion engine, a printing machine and a hydraulic machine; the health state of the human body fluid circulation and the like are significant. The viscosity of the liquid is related to factors such as composition, temperature and pressure, and the viscosity is mainly measured by a capillary method, a rotation method, a vibration method, a falling body method, an optical method, an ultrasonic method and the like. Among them, the capillary method is the most widely used method for measuring liquid viscosity due to its characteristics of high measurement accuracy, simple structure, etc.

At present, the verification of a capillary viscometer and the measurement of kinematic viscosity are mainly manual measurement, the measurement time is required to be continuously (200-2000) s every time, the labor intensity of operators is high, and the mental stress is easy to occur, so the working efficiency is difficult to improve.

Disclosure of Invention

In view of the above, there is a need to provide a viscosity detection system and method based on machine vision.

The embodiment of the invention provides a viscosity detection system based on machine vision, which comprises:

the camera is used for acquiring continuous images at an upper timing marking and continuous images at a lower timing marking of the viscometer;

a computer, comprising:

the timing control module is used for controlling the timing device to start timing when the fluid exists in the continuous images at the upper timing marking line; when the fluid exists in the continuous images at the lower timing marking line, controlling the timing device to end timing;

the flow time determining module is used for determining the flow time of the fluid with known viscosity in the capillary tube of the viscometer or the flow time of the fluid with unknown viscosity in the capillary tube of the viscometer according to the starting timing time and the ending timing time;

the viscometer constant determination module is used for inputting the flow time of the known viscosity fluid in a capillary of the viscometer and the nominal value of the kinematic viscosity of the fluid into a kinematic viscosity formula of the fluid to obtain the viscometer constant; and

and the kinematic viscosity value determining module is used for inputting the constant of the viscometer and the flow time of the unknown viscosity fluid in the capillary of the viscometer into a fluid kinematic viscosity formula to obtain the kinematic viscosity value of the unknown viscosity fluid.

The viscosity detection system based on machine vision provided by the embodiment of the invention also comprises a light source, a controller and a controller, wherein the light source is arranged on the side of the viscometer; and the light source adopts a 300mm 600mm LED white light source.

Further, the viscometer employs a flat capillary viscometer.

Furthermore, the camera adopts a CMOS photosensitive chip high-speed infrared camera.

The viscosity detection system based on machine vision provided by the embodiment of the invention also comprises a fluid tracking device, which specifically comprises:

the guide rail is arranged on the side of the viscometer;

the screw thread penetrates through the screw rod, and the screw seat is matched in the guide rail in a sliding manner; the camera is fixed on the threaded seat;

and a rotating shaft of the stepping motor is fixedly connected with one end of the screw rod.

Further, the minimum moving amount of the camera moving on the guide rail is 0.1 mm.

Further, the timing device employs an electric stopwatch.

Further, the fluid level at the upper and lower timing marks is meniscus shaped.

Further, the kinematic viscosity formula of the fluid is as follows:

ν＝C·τ

wherein v is the kinematic viscosity of the fluid and has a nominal value of 10.06mm²S and 19.67mm²S; c is a viscometer constant;τ is the flow time of a fluid of known viscosity in the capillary of the viscometer, or the flow time of a fluid of unknown viscosity in the capillary of the viscometer.

The embodiment of the invention also provides a machine vision-based viscosity detection method, which comprises the following steps:

acquiring continuous images at an upper timing marking and a lower timing marking of the viscometer;

when the fluid exists in the continuous images at the upper timing marking line, controlling a timing device to start timing; when the fluid exists in the continuous images at the lower timing marking line, controlling the timing device to end timing;

determining the flow time of the fluid with known viscosity in a capillary tube of the viscometer or the flow time of the fluid with unknown viscosity in the capillary tube of the viscometer according to the starting timing time and the ending timing time;

inputting the flow time of the known viscosity fluid in a capillary tube of a viscometer and a nominal value of the kinematic viscosity of the fluid into a kinematic viscosity formula of the fluid to obtain a viscometer constant; and

and inputting the constant of the viscometer and the flow time of the fluid with unknown viscosity in a capillary of the viscometer into a fluid kinematic viscosity formula to obtain a kinematic viscosity value of the fluid with unknown viscosity.

Compared with the prior art, the viscosity detection system and method based on machine vision provided by the embodiment of the invention have the following beneficial effects:

aiming at the problems of high labor intensity, low efficiency and the like of manual judgment and timing in the current capillary viscometer verification and kinematic viscosity measurement processes, the invention designs a set of viscosity detection system; the method comprises the steps of firstly building a hardware system, then analyzing four common capillary viscometer, carrying out capillary viscometer verification and kinematic viscosity measurement tests on the representative of the viscosity viscometer by using a Pink viscosity meter, and showing that the time repeatability of the capillary viscometer in the detection can be as low as 0.017% and the kinematic viscosity measurement repeatability can be as low as 0.043% based on a viscosity detection system of a machine vision technology, is superior to a manual measurement result and is less than corresponding regulations and standard requirements, can replace a manual measurement process, and obviously improves the working efficiency.

Drawings

FIG. 1 is a schematic diagram of a conventional capillary viscometer provided in an embodiment of the invention;

FIG. 2 is a block diagram of a flat capillary viscometer structure according to one embodiment of the invention;

FIG. 3 is a schematic diagram of the shape of the fluid level provided in an embodiment of the present invention;

FIG. 4 is a machine vision inspection system provided in an embodiment of the present invention;

fig. 5 is a schematic diagram of a timing start image and a gray scale and binarization processing result provided in an embodiment of the present invention;

fig. 6 is a schematic diagram of a timing-terminated image and a result of a gray scale and binarization processing according to another embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

In one embodiment, a machine vision-based viscosity detection system is provided, the system comprising:

1. fluid kinematic viscosity equation

According to Poiseuille's law, when a certain volume of Newtonian fluid passes through a given capillary under a certain pressure gradient, the fluid is incompressible, flows in a stable laminar flow state, and does not slide along the wall of the pipe; under the conditions that the capillary tube is linear, the inner diameter is uniform, and the ratio of the length to the diameter is large enough, the dynamic viscosity eta of the capillary tube can be obtained by calculating through measuring the flow rate of fluid and the pressure difference generated by flowing through the capillary tube:

wherein Δ p represents the pressure difference across the capillary; l, R denotes capillary length and radius; q represents the volumetric flow rate through the capillary tube per unit time. When a glass capillary viscometer is used for measurement, Δ p can be expressed by hydrostatic pressure ρ gh, and the kinematic viscosity ν of the measured fluid can be calculated according to the density ρ of the measured fluid:

wherein g is the gravity acceleration, h is the height of the measured fluid column, V is the volume of the fluid, and tau is the flowing time of the fluid in the capillary. Under the condition that the capillary and the measured fluid are determined, R, h, V and L are all constant values, and the order is that

Can obtain the product

ν＝C·τ (3)

Equation (3) above is the equation for measuring kinematic viscosity of fluid with a capillary viscometer, and C is called viscometer constant.

And calibrating the viscometer by using the national standard substance of the standard viscosity liquid according to the verification regulation requirement of JJG155-2016 working capillary viscometer. Common capillary viscometers include ubpellohde (ubpellohde), pingi (Pinkievich), fen (Fenske), and counterflow. Referring to fig. 1, and the labels in fig. 1 are: 1-main tube, 2-wide tube, 3-side tube, A-measuring ball, B-liquid storage ball, C-buffer ball, D-suspension horizontal ball, E-upper measuring ball, F-upper liquid storage ball, R-capillary, G-liquid containing marking line, m₁Upper timing mark, m₂-lower timing marker.

Among them, the viscosity meter of the Ping-shi has a simple structure, is relatively convenient to operate, has a wide viscosity measurement range, and is commonly used as a capillary viscosity meter. The invention takes the Ping viscosity as a representative, and researches an automatic viscosity detection system. The structure of the viscometer is described in detail with reference to fig. 2, and the symbols in fig. 2 are: a-a lower receptacle; c-a timing ball; d-an upper reservoir; e-timing marked lines are arranged; f, lower timing marking; l-gripping the tube; an N-vent pipe; an O-branch pipe; p-connecting pipe; r-working capillary.

A certain amount of standard viscosity liquid with known kinematic viscosity is taken, a timing marking of a capillary viscometer is passed through according to requirements, the time length of the viscosity liquid passing through the timing marking is measured, and then the constant of the viscometer can be calculated. Accordingly, the kinematic viscosity value can be calculated by measuring the time period for which the unknown fluid passes through the timing mark with a viscometer whose constant is known.

2. Machine vision inspection demand analysis

According to the formula of the capillary viscometer for measuring the kinematic viscosity of the fluid, the automatic detection system can accurately identify the time when the fluid flows through the upper and lower timing marked lines and calculate the flow time of the fluid. As a result of the viscosity, the fluid is in a steady laminar state when flowing through the reticle, the liquid surface of which is in the shape of a meniscus, see in particular fig. 3, and the labels in fig. 3 are: c-timer, D-upper storage device, E-upper timing mark line, F-lower timing mark line, M-stepping motor, P-camera, R-working capillary.

The machine vision inspection requirements of the automatic viscosity inspection system obtained by simulating the manual measurement process are as follows:

1) real-time detection of fluid flow through the time scale line is accomplished.

2) Ensuring that the detection visual angle and the time scale line are on the same horizontal plane.

3) Accurate identification of colored fluid and transparent reticle is achieved.

3. Construction of machine vision detection system

According to the machine vision detection requirement of the measuring system, the machine vision detection system is determined to be composed of a camera, a fluid tracking device, a light source and a timing device, and is shown in figure 4.

Before measurement, the camera is driven to reciprocate for a circle by a stepping motor, and the upper and lower timing marked lines of the viscometer to be measured are identified by utilizing a template matching mode to realize the tracking of the measured fluid.

When the measurement is started, the camera runs to the horizontal plane of the upper timing marking, the trigger pulse is output to start timing after the fluid flow signal is detected, then the camera runs to the horizontal plane of the lower timing marking, the stop pulse is output to end timing after the fluid flow signal is detected again, and the time interval of two pulses is used as the fluid flow time.

4. Selection of visual system

1) Video camera

In order to accurately acquire fluid motion images, the system selects a CMOS photosensitive chip high-speed infrared camera, the pixel number is more than 1M, the image acquisition speed is 120 frames per second, and a USB2.0 data interface is adopted for image data transmission.

2) Fluid tracking device

In the measuring process, the stepping motor is used for driving the camera to move in the vertical direction, the minimum moving amount of the linear guide rail can reach 0.1mm, the camera and the reticle are always in the same horizontal plane and are kept in a static state, and the influence caused by image distortion of the camera is effectively avoided.

3) Light source

The background light source is a 300mm multiplied by 600mm LED white light source which is matched with an infrared camera for use, and the colorless time scale line and the colored fluid can be accurately distinguished.

4) Time-piece

The system adopts an electric stopwatch for timing, the measuring range is 0.0001 s-9999.99 s, and the measuring error is less than +/- (5 multiplied by 10)^-6τ)。

5. Realization of automatic viscosity detection system

During the detection process, the fluid in the capillary viscometer is in a motion state, and a dynamic image is obtained by the detection system. And analyzing the dynamic image into continuous static pictures through a high-speed camera, and identifying a timing starting point and a timing end point through a high-precision positioning template matching mode. The detection software system is compiled by C # language, video image acquisition input and output are carried out by adopting an AForge library, an OpenCVSharp library is used for processing the acquired image, and pixels in the image are processed and identified by adopting the OpenCVSharp library together with the levels of scale marks and liquid level through gray processing, denoising, binarization, morphological opening and closing operation, expansion and corrosion processing, so that automatic measurement is realized (fig. 5 and 6).

Example 1:

capillary viscometer constant measurement

The constructed automatic detection system is utilized, according to the requirements of the national metrological verification regulation JJJG 155-2016 working capillary viscometer, under the constant temperature condition of (20 +/-0.01) DEG C, the nominal values are respectively 10.06mm²S and 19.67mm²The capillary constant of a standard viscosity liquid/s is measured by a flat viscometer with the inner diameter of a capillary being 0.8mm, and compared with the manual measurement result, the result is shown in the following table 1:

TABLE 1 comparison of capillary viscometer constant measurements

As can be seen from Table 1, the time reproducibility and the constant reproducibility of the capillary constant measured by the automatic detection system are both significantly better than those of the manual measurement results.

Example 2:

kinematic viscosity measurements

By utilizing the built automatic detection system, according to the requirements of the national standard GB/T10247-2008 viscosity measurement method, under the constant temperature condition of (20 +/-0.01) DEG C, a flat viscometer with the capillary inner diameter of 0.6mm and 0.8mm is selected to measure the kinematic viscosity of a No. 0 diesel oil sample, and the kinematic viscosity is compared with the manual measurement result, and the result is shown in the following table 2:

TABLE 2 kinematic viscosity measurement comparison

As can be seen from Table 2, the repeatability and reproducibility of the kinematic viscosity values of the samples measured by the automatic detection system are both significantly better than those of the manual measurement results.

In conclusion, the automatic viscosity detection system based on machine vision designed by the invention is used for measuring the capillary constant of the viscometer and the kinematic viscosity of the No. 0 diesel oil sample, and the results show that the automatic detection system is obviously superior to the manual measurement results, and meanwhile, the automatic detection system can simplify the timing and calculation processes, and can play a role in improving the measurement accuracy and releasing the human power. The invention establishes the automatic viscosity detection system based on machine vision aiming at the verification process and the kinematic viscosity measurement process of the Ping-Marek viscometer, improves the verification and kinematic viscosity measurement efficiency of the viscometer, and has better application prospect in the aspects of quantity value transmission of a metering technical mechanism and viscosity measurement of an oil product inspection mechanism.

In one embodiment, a machine vision-based viscosity detection method is provided, which includes:

successive images at the upper timing marker and successive images at the lower timing marker of the viscometer are acquired.

When the fluid exists in the continuous images at the upper timing marking line, controlling a timing device to start timing; and controlling the timing device to finish timing when the fluid exists in the continuous images at the lower timing marked line.

And determining the flow time of the fluid with known viscosity in the capillary tube of the viscometer or the flow time of the fluid with unknown viscosity in the capillary tube of the viscometer according to the starting timing time and the ending timing time.

And inputting the flow time of the known viscosity fluid in a capillary tube of the viscometer and the nominal value of the kinematic viscosity of the fluid into a kinematic viscosity formula of the fluid to obtain a viscometer constant. And

and inputting the constant of the viscometer and the flow time of the fluid with unknown viscosity in a capillary of the viscometer into a fluid kinematic viscosity formula to obtain a kinematic viscosity value of the fluid with unknown viscosity.

It should be noted that, for specific limitations of the machine vision-based viscosity detection method, reference may be made to the above limitations of the machine vision-based viscosity detection system, and details thereof are not repeated here.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features. Furthermore, the above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims (10)

1. A machine vision-based viscosity detection system, comprising:

the camera is used for acquiring continuous images at an upper timing marking and continuous images at a lower timing marking of the viscometer;

a computer, comprising:

the timing control module is used for controlling the timing device to start timing when the fluid exists in the continuous images at the upper timing marking line; when the fluid exists in the continuous images at the lower timing marking line, controlling the timing device to end timing;

the flow time determining module is used for determining the flow time of the fluid with known viscosity in the capillary tube of the viscometer or the flow time of the fluid with unknown viscosity in the capillary tube of the viscometer according to the starting timing time and the ending timing time;

the viscometer constant determination module is used for inputting the flow time of the known viscosity fluid in a capillary of the viscometer and the nominal value of the kinematic viscosity of the fluid into a kinematic viscosity formula of the fluid to obtain the viscometer constant; and

and the kinematic viscosity value determining module is used for inputting the constant of the viscometer and the flow time of the unknown viscosity fluid in the capillary of the viscometer into a fluid kinematic viscosity formula to obtain the kinematic viscosity value of the unknown viscosity fluid.

2. The machine-vision-based viscosity detection system of claim 1, further comprising a light source disposed to the side of the viscometer; and the light source adopts a 300mm 600mm LED white light source.

3. The machine-vision-based viscosity detection system of claim 1, wherein the viscometer is a flat capillary viscometer.

4. The machine-vision-based viscosity detection system of claim 1, wherein the camera is a CMOS-chip high-speed infrared camera.

5. The machine-vision-based viscosity detection system of claim 1, further comprising a fluid tracking device, which specifically comprises:

the guide rail is arranged on the side of the viscometer;

the screw thread penetrates through the screw rod, and the screw seat is matched in the guide rail in a sliding manner; the camera is fixed on the threaded seat;

and a rotating shaft of the stepping motor is fixedly connected with one end of the screw rod.

6. The machine-vision-based viscosity detection system of claim 5 wherein the minimum amount of movement of the camera on the guide rail is 0.1 mm.

7. The machine vision-based viscosity detection system of claim 1 wherein said timing device employs an electric stopwatch.

8. The machine vision-based viscosity detection system of claim 1 wherein the fluid level at the upper and lower timing marks is meniscus shaped.

9. The machine-vision-based viscosity sensing system of claim 1, wherein the fluid kinematic viscosity formula is:

ν＝C·τ

wherein v is the kinematic viscosity of the fluid and has a nominal value of 10.06mm²S and 19.67mm²S; c is a viscometer constant; τ is the flow time of a fluid of known viscosity in the capillary of the viscometer, or the flow time of a fluid of unknown viscosity in the capillary of the viscometer.

10. A machine vision-based viscosity detection method is characterized by comprising the following steps:

acquiring continuous images at an upper timing marking and a lower timing marking of the viscometer;

when the fluid exists in the continuous images at the upper timing marking line, controlling a timing device to start timing; when the fluid exists in the continuous images at the lower timing marking line, controlling the timing device to end timing;

determining the flow time of the fluid with known viscosity in a capillary tube of the viscometer or the flow time of the fluid with unknown viscosity in the capillary tube of the viscometer according to the starting timing time and the ending timing time;

inputting the flow time of the known viscosity fluid in a capillary tube of a viscometer and a nominal value of the kinematic viscosity of the fluid into a kinematic viscosity formula of the fluid to obtain a viscometer constant; and

and inputting the constant of the viscometer and the flow time of the fluid with unknown viscosity in a capillary of the viscometer into a fluid kinematic viscosity formula to obtain a kinematic viscosity value of the fluid with unknown viscosity.

Images