CN115235948A - Liquid viscosity measurement system based on computer vision identification - Google Patents

Abstract

The invention relates to a liquid viscosity measurement system based on computer vision recognition, which belongs to the technical field of viscosity measurement under the detection and measurement technology, and designs and builds an image acquisition hardware structure, comprising the following steps: the system comprises a constant-temperature water tank, a surface light source, a cloud deck, an industrial camera, a stepping motor and a computer, wherein video images are collected and detected through the industrial camera, cross-platform computer vision library OpenCV image collection processing and computer vision technology are applied under a Windows operating system, after gray processing is carried out on collected images, identification of scale marks is achieved through a straight-line segment detection algorithm, and identification and detection of motion liquid levels are achieved through an improved ViBe motion target detection algorithm. Experiments show that the system is accurate and reliable in detection in a reliable environment and meets the requirement of on-line monitoring on real-time performance; the man-machine interaction performance is good, the intelligent degree and the detection efficiency of the viscometer are effectively improved, and the actual use value is high.

Description

Liquid Viscosity Measurement System Based on Computer Vision Recognition

technical field

The invention belongs to the technical field of viscosity measurement under detection and measurement technology, and in particular relates to a liquid viscosity measurement system based on computer vision recognition technology.

Background technique

The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

Viscosity is the degree of viscosity, which is an inherent property of the fluid. It is used to indicate the difficulty of fluid flow. Under the action of gravity or an external force, the flow process of the fluid will undergo continuous deformation, and the interaction of liquid motion will occur. The force is expressed as the viscosity of the liquid, and the magnitude of the interaction force between the fluid molecules is characterized by the magnitude of the fluid viscosity. Different substances have different viscosities, and the same substance has different viscosities at different temperatures. There are two absolute viscosity measurement methods for viscosity measurement: dynamic viscosity and kinematic viscosity. The absolute measurement method requires very precise processing of the viscometer, and the measurement operation process is cumbersome and the measurement accuracy is low. Capillary viscometer is based on the capillary method to measure fluid viscosity, which is the most widely used measurement method in many viscosity measurement methods. The capillary method is based on the basic principle of Poiseuille's law (the basic equation of the capillary viscosity method, referred to as Poiseuille's law), that is, the viscous fluid drops in the capillary under the action of its own gravity, and the time it takes for a certain volume of fluid to flow through the capillary is related to the viscosity of the fluid. There is a corresponding relationship between them. The kinematic viscosity of the fluid is obtained by measuring the product of the time the fluid flows through the capillary multiplied by the instrument constant for which the capillary is calibrated. In the use of this type of viscometer, the structure will be slightly different due to different conditions and different measurement objects. The most common ones are U-type and Finn-type capillary viscometers.

When measuring a liquid with a high viscosity, the liquid descends slowly in the tube, and the time is too long, and the measurement result needs to be calculated manually, which will seriously consume the operator's energy and time. Therefore, viscosity measurement is gradually developing in the direction of automatic measurement.

SUMMARY OF THE INVENTION

In order to overcome the above deficiencies, the present invention provides a suitable viscosity measurement system based on computer vision recognition technology, designs image acquisition, image acquisition processing, timing calculation system, realizes detection and tracking of the moving liquid level through a computer, and determines whether the liquid level flows through the The upper and lower scale lines of the viscometer timing ball are used to obtain the timing time, and then the relative method is used to calculate it according to the formula v=C*t, where v represents the kinematic viscosity of the liquid; C represents the viscometer constant; t represents a certain volume the time it takes for the liquid to flow through the capillary. The system can reduce the workload of the experimenter, improve the verification accuracy and detection efficiency, and at the same time increase the economic benefit, which has high innovation and practical significance.

To achieve the above object, one or more embodiments of the present invention provide the following technical solutions:

A liquid viscosity measurement system based on computer vision recognition, comprising: a constant temperature water tank, a surface light source, a pan/tilt, an industrial camera, a stepping motor and a computer, the surface light source and the pan/tilt are located on opposite sides of the constant temperature water tank; the industrial camera passes through the The movable bracket is slidably installed on the PTZ to form an image acquisition system. The lower part of the PTZ is equipped with a stepper motor, which drives the movable bracket to slide up and down, and the industrial camera moves left and right on the horizontal plane of the movable bracket; the capillary viscometer is fixed by the measuring bracket In the transparent circulating constant temperature water tank, the frame image captured by the industrial camera is transmitted to the computer, and the image acquisition processing and timing calculation control unit module is installed on the computer. Position identification; when the liquid to be tested is in free fall in the capillary viscometer and passes the upper scale line on the timing ball from top to bottom, the computer makes a judgment and starts timing; when the liquid to be tested passes through the lower scale line on the timing ball When the interval time is calculated, stop the timing; after calculating the interval time, calculate the kinematic viscosity of the liquid to be tested according to the formula.

In a further technical solution, the computer obtains the frame image by creating a camera, and then converts it into an image format recognizable by Open CV, converts the color image into a grayscale image, delineates the area of interest, and uses the Canny operator in the area of interest to compare the grayscale. The edge detection is carried out on the processed image, and then the detection of the scale line and the detection of the moving liquid level are carried out.

Further technical solutions, for the identification and detection of the scale line at the timing ball of the capillary viscometer, the algorithm used is a linear segment detection algorithm; for the detection and tracking of the moving liquid level in the capillary viscometer, the algorithm used is the improved ViBe moving target detection algorithm. . ViBe (Visual Background extaoctor) is a pixel-level video background modeling or foreground detection algorithm. Its detection effect is better than that of mainstream detection algorithms such as inter-frame difference method, and it occupies relatively less computer memory. However, when the moving speed of the moving target differs greatly, ghost images may appear, resulting in incomplete target detection. Therefore, it should be appropriately improved to reduce the occurrence of missed detection and false detection.

For a further technical solution, firstly, using the improved ViBe moving target detection algorithm, using multiple frames of images and using the improved three-frame difference method to construct the background model prototype of the ViBe algorithm, and adding the adjacent element change values to the prototype. The background model and the threshold are adjusted according to the difference between the current frame image and the background model as an adjustment factor for the complexity of the background. The detected results are combined with the improved three-frame difference method to perform logical AND operation and post-processing operations to complete the detection of the moving liquid level in the viscometer.

In a further technical solution, the surface light source and the constant temperature water tank are integrated, the pan/tilt is not in contact with the constant temperature water tank during setting, and the frame images captured by the industrial camera are transmitted to the computer by a gigabit network cable.

In a further technical solution, the industrial camera preferably uses a Hikvision industrial area scan camera as the image acquisition device, and the kinematic viscosity of the liquid to be tested is displayed on the display screen of the computer.

In a further technical solution, the capillary viscometer is a viscometer that has passed the test and has a permanent mark.

One or more of the above technical solutions have the following beneficial effects:

The technical scheme of the present invention can complete the timing work of viscosity measurement through a computer, and can meet the index requirements in the national measurement regulations. Through computer vision technology, the upper and lower scale lines of the capillary viscometer can be accurately identified and calibrated, providing a valuable reference for the verification process. Through the identification and tracking of the moving liquid level, the time when the liquid level passes through the upper and lower timing lines of the capillary viscometer is detected, and the time interval is obtained to calculate the viscosity of the test liquid.

The detection environment and detection method established by the technical solution of the present invention have standardized operation requirements. The experimenter can fix the capillary viscometer containing the liquid to be tested in the constant temperature water tank, and then adjust the position of the industrial camera through the PC to start the detection, without the need for re-testing. Adjust the environment, etc.

The image acquisition and processing part of the technical solution of the present invention adopts the open source code computer vision class library OpenCV as the core part of the design. All codes of the library are optimized, and the calculation efficiency is high and the operation efficiency is faster. The image acquisition and processing technology meets the requirements of this design and realizes the functions of image acquisition and processing and computer vision.

The target detection algorithm of the technical solution of the present invention adopts the improved ViBe algorithm, which not only has the advantages of easy implementation and high operation efficiency of the original ViBe algorithm, but also can eliminate false targets (ghost images), target and background fusion and other problems in this solution.

The technical scheme of the invention has the advantages of simple measurement device, strong adaptability to the environment, few error influencing factors, and high measurement accuracy can be achieved by relying on advanced machine vision and image acquisition and processing algorithms.

Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will become apparent from the description which follows, or may be learned by practice of the invention.

Description of drawings

The accompanying drawings forming a part of the present invention are used to provide further understanding of the present invention, and the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, and do not constitute an improper limitation of the present invention.

Fig. 1 is the schematic diagram of the capillary viscometer used in the embodiment of the present invention;

2 is a schematic structural diagram of a liquid viscosity measurement system according to an embodiment of the present invention;

3 is a schematic diagram of an image area captured by an industrial camera according to an embodiment of the present invention;

FIG. 4 is a block diagram of an image acquisition processing flow according to an embodiment of the present invention;

5 is a block diagram of an improved ViBe moving target detection algorithm according to an embodiment of the present invention;

In the figure, 1 is a capillary viscometer, 2 is a constant temperature water tank, 3 is a surface light source, 4 is a gimbal, 5 is an industrial camera, 6 is a stepping motor, 7 is a computer, 8 is an upper scale line, and 9 is a lower scale line , 10 is the timing ball.

Detailed ways

It should be noted that the following detailed description is exemplary and intended to provide further explanation of the invention. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

It should be noted that the terminology used herein is for the purpose of describing specific embodiments only, and is not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly dictates otherwise, the singular is intended to include the plural as well, furthermore, it is to be understood that when the terms "comprising" and/or "including" are used in this specification, it indicates that There are features, steps, operations, devices, components and/or combinations thereof.

As shown in FIG. 1 , which is a schematic diagram of the capillary viscometer used in the embodiment of the present invention, the capillary viscometer 1 is a viscometer based on the capillary method to measure the viscosity of fluids, and is a viscometer that has passed the verification and has a permanent mark.

As shown in FIG. 2 , it is a schematic structural diagram of a liquid viscosity measurement system according to an embodiment of the present invention. A liquid viscosity measurement system based on computer vision recognition, comprising: a constant temperature water tank 2, a surface light source 3, a pan/tilt 4, an industrial camera 5, a stepping motor 6 and a computer 7, the surface light source 3 and the pan/tilt 4 are respectively located in the constant temperature water tank 2 the opposite sides. The surface light source 3 and the constant temperature water tank 2 are integrated in structure, and the bottom shares a bottom panel to ensure the stability of the surface light source 3 and the constant temperature water tank 2 during the test process. The surface light source 3 and the constant temperature water tank 2 constitute the environment required by the liquid viscosity measurement system. Both the surface light source 3 and the constant temperature water tank 2 can be disassembled and replaced independently to adapt to different environmental requirements.

The industrial camera 5 is slidably installed on the gimbal 4 through the movable bracket to form an image acquisition system. The lower part of the gimbal 4 is provided with a stepper motor 6, and the stepper motor 6 drives the movable bracket to slide up and down to adjust the height of the industrial camera 5. The camera 5 can be moved left and right on the horizontal plane of the movable bracket, which is convenient for taking accurate images of the capillary viscometer 1 of various specifications and styles. Limited by the length of different types of capillary viscometers 1 and the different positions of timing balls 10, the fixed image acquisition system has limited adaptability, so this design develops an adjustable real-time image acquisition that can adapt to different types of capillary viscometers 1 system. For industrial camera 5, Hikvision industrial area scan camera is preferred as the image acquisition device. Compared with ordinary cameras, this Hikvision industrial camera has a large storage capacity, a large adjustment range of the camera's field of view, and can be directly called; the resolution of Hikvision industrial camera is 500 10,000, the acquisition frame rate is 24 frames/s. Since the circulating constant temperature water has a slight vibration when the constant temperature water tank 2 is working, in order to avoid the influence of this slight vibration on the image acquisition, the pan/tilt 4 is not in contact with the constant temperature water tank 2 during setting. Debug and fix the height of the pan/tilt 4 and the distance between the industrial camera 5 and the capillary viscometer 1, so that the image acquisition device can collect a clear image covering the complete range of the timing ball 10. The size of a single image is 2448 pixels × 2048 pixel. As shown in FIG. 3 , it is a schematic diagram of an image area captured by an industrial camera according to an embodiment of the present invention. By adjusting the height and horizontal position, it is ensured that the industrial camera 5 can completely capture the upper scale line 8 and the lower scale line on the timing ball 10 Image at 9.

The capillary viscometer 1 is fixed in the transparent circulating constant temperature water tank 2 through the measuring bracket, the surface light source 3 with stable brightness provides suitable background light, and the industrial camera 5 can be freely adjusted vertically through the control of the stepping motor 6 to adapt to different heights. The frame image captured by the industrial camera 5 is transmitted to the computer 7 by a gigabit network cable. After the computer 7 completes the measurement, timing and calculation, the viscosity of the liquid to be tested is obtained.

An image acquisition processing and timing calculation control unit module is installed on the computer 7. As a control unit for image acquisition processing and timing calculation, the computer 7 receives the frame image captured by the industrial camera 5, and carries out the determination of the scale line and the identification of the moving liquid level. . When the liquid to be tested does a free fall motion in the capillary viscometer 1 and passes the upper scale line 8 on the timing ball 10 from top to bottom, the computer 7 makes a judgment and starts timing; when the liquid to be tested passes through the lower scale on the timing ball 10 At line 9, stop timing; after calculating the interval time, calculate the kinematic viscosity of the liquid to be tested according to the formula, and present it on the display screen of the computer 7.

As shown in FIG. 4 , it is a block diagram of an image acquisition processing flow according to an embodiment of the present invention. The image acquisition processing and timing calculation unit is mainly implemented in Visual Studio 2019 through C++ algorithm statements. Computer 7 obtains a frame image by creating a camera, then converts it to an image format recognizable by Open CV, converts the color image to a grayscale image, defines a region of interest (ROI), and uses the Canny operator in the region of interest to compare the grayscale. The processed image is subjected to edge detection, which can accurately obtain the edge of the liquid level image, and then perform the detection of the scale line and the detection of the moving liquid level.

For the identification and detection of the tick marks at the timing ball 10 of the capillary viscometer 1, the algorithm used is the linear segment detection algorithm (LSD algorithm, Line Segment Detector). The algorithm can obtain high-precision straight-line segment detection results in a short period of time, with fast detection speed and without parameter adjustment. The method of error control is used to improve the accuracy of straight-line detection.

For the detection and tracking of the moving liquid level in the capillary viscometer 1, the improved ViBe moving target detection algorithm is used, as shown in FIG. 5, which is a block diagram of the improved ViBe moving target detection algorithm according to the embodiment of the present invention. Based on the collected video sequence, the background model prototype of ViBe algorithm is constructed by using multi-frame images and the improved three-frame difference method, and the change value of adjacent elements is added to the prototype. According to the difference between the current frame image and the background model as the adjustment factor of the background complexity, to adjust the background model and threshold, and perform median filtering; the improved three-frame difference method includes: three-frame difference processing, expansion processing, logical OR operation and morphology learn filtering. Foreground detection is the process of identifying moving objects (foreground objects) and static parts in a video image sequence, that is, identifying the moving liquid surface and static background in the timing ball in the image sequence. Logical AND operation and post-processing operation (morphological filtering) are performed on the detected results and the three-frame difference method to complete the detection of the moving liquid level in the viscometer.

In the detection process, the present invention can guarantee the experimental environment, under the premise of less noise or no noise, there is no need for excessive preprocessing, resulting in excessive data calculation amount and errors caused by insufficient computer computing power.

Those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by the computing device, so that they can be stored in a storage device. The device is executed by a computing device, or they are separately fabricated into individual integrated circuit modules, or multiple modules or steps in them are fabricated into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Although the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, they do not limit the scope of protection of the present invention. Those skilled in the art should understand that on the basis of the technical solutions of the present invention, those skilled in the art do not need to pay creative efforts. Various modifications or deformations that can be made are still within the protection scope of the present invention.

Claims (8)

1. The liquid viscosity measurement system based on computer vision discernment, characterized by, includes: the device comprises a constant-temperature water tank (2), a surface light source (3), a cradle head (4), an industrial camera (5), a stepping motor (6) and a computer (7), wherein the surface light source (3) and the cradle head (4) are respectively positioned on two opposite sides of the constant-temperature water tank (2); the industrial camera (5) is slidably mounted on the cradle head (4) through the movable support to form an image acquisition system, the stepping motor (6) is arranged at the lower part of the cradle head (4), the movable support is driven by the stepping motor (6) to slide up and down, and the industrial camera (5) moves left and right on the horizontal plane of the movable support; the capillary viscometer (1) is fixed in a transparent circulating constant-temperature water tank (2) through a measuring bracket, a frame image shot by an industrial camera (5) is transmitted to a computer (7), an image acquisition processing and timing calculation control unit module is installed on the computer (7), and the computer (7) receives the frame image shot by the industrial camera (5) and judges scale lines and identifies a motion liquid level; when the liquid to be tested does free falling motion in the capillary viscometer (1) and passes through an upper scale mark (8) on the timing ball (10) from top to bottom, the computer (7) makes a judgment and starts timing; when the liquid to be tested passes through the lower scale mark (9) on the timing ball (10), stopping timing; and after the interval time is obtained through calculation, calculating the kinematic viscosity of the liquid to be tested according to a formula.

2. The liquid viscosity measurement system based on computer vision recognition according to claim 1, wherein the computer (7) obtains a frame image by creating a camera, converts the frame image into an image format recognizable by Open CV, converts a color image into a gray scale image, delimits an area of interest, performs edge detection on the image after gray scale processing in the area of interest by using Canny operator, and then performs detection of scale marks and detection of moving liquid level.

3. The computer vision recognition-based liquid viscosity measurement system according to claim 2, characterized in that for the recognition detection of the scale lines at the timing ball (10) of the capillary viscometer (1), the algorithm used is a straight line segment detection algorithm; for the detection and tracking of the moving liquid level in the capillary viscometer (1), the adopted algorithm is an improved ViBe moving target detection algorithm; the ViBe, namely background modeling, is an algorithm for pixel-level video background modeling or foreground detection, and when the motion speed of a moving target is greatly different, ghost images can appear to cause incomplete target detection, so that the ViBe is improved to reduce the occurrence of missed detection and error detection.

4. The computer vision recognition-based liquid viscosity measurement system according to claim 3, wherein an improved ViBe moving object detection algorithm is applied, a background model prototype of the ViBe algorithm is constructed by using a multi-frame image and an improved three-frame difference method, and adjacent element change values are added to the prototype; taking the difference value of the current frame image and the background model as an adjusting factor of the background complexity degree to adjust the background model and the threshold value; and carrying out logical AND operation and post-processing operation on the detected result and an improved three-frame difference method to finish the detection of the motion liquid level in the viscometer.

5. The computer vision recognition-based liquid viscosity measurement system of claim 4, wherein the improved three-frame difference method comprises: three-frame difference processing, dilation processing, logical or operations, and morphological filtering.

6. The computer vision recognition-based liquid viscosity measurement system according to claim 1, wherein the surface light source (3) and the constant temperature water tank (2) are of an integral structure, the pan-tilt (4) is not in contact with the constant temperature water tank (2) when being set, and a frame image shot by the industrial camera (5) is transmitted to the computer (7) through a gigabit network.

7. The computer vision recognition based liquid viscosity measurement system according to claim 1, characterized in that an industrial camera (5), preferably a Haikang industrial area-array camera, is used as an image acquisition device, and the kinematic viscosity of the liquid to be tested is presented on a display screen of a computer (7).

8. The computer vision recognition based liquid viscosity measurement system according to claim 1, characterized in that the capillary viscometer (1) is a viscometer that has passed certification and has a permanent identification.

Images