Evaporation tank liquid level detection method based on machine vision

Abstract

The invention belongs to the technical field of sugar manufacturing processes, and particularly relates to an evaporation tank liquid level detection method based on machine vision, which comprises the following steps: S1, acquiring an original image of a sight glass of an evaporation tank; s2, carrying out noise reduction processing on the original image; s3, calculating the RGB value of each pixel in the divided region, converting the RGB value into an XYZ value, and converting the XYZ value into an LAB value; s4, according to the algorithm model of liquid level detection and identification, studying and judging whether each pixel is a liquid identifier or a liquid-free identifier; s5, calculating the ratio of the number of the pixels with the liquid marks to the number of the total pixels; s6, comprehensively calculating the highest division area of the liquid of the single sight glass, and finding out the calibrated liquid level scale of the area; s7, circulating the steps until the liquid level scales reached by all the sight glasses on the evaporation tank are analyzed; and S8, comprehensively calculating the liquid level scales reached by all sight glasses. The problems that the liquid level calculated through a traditional liquid level measuring mode does not meet the requirement, and the liquid level numerical value cannot be accurately obtained can be solved, and good market application prospects are achieved.

Description

Technical Field

[0001] The invention belongs to the technical field of sugar making process, and in particular relates to a method for detecting the liquid level of an evaporation tank based on machine vision. Background Art

[0002] The sugar industry is the basic industry of the food industry, and it is also the raw material industry for a variety of products such as papermaking, chemical industry, fermentation, medicine, building materials, furniture, etc. It occupies an important position in the national economy. The sugar production process includes raw material processing, juice extraction, clarification and purification, evaporation and concentration, sugar boiling and crystallization, separation and drying, packaging and storage. Evaporation and concentration is an important part of the sugar production process. The working principle is to heat the clarified sugar juice with steam in the evaporator to evaporate the water and increase the concentration of the sugar juice. In order to improve the evaporation efficiency, low liquid level evaporation is required, and the liquid level is calculated according to the actual liquid and the height of the generated bubbles. Since the evaporation process uses steam heating to heat the sugar juice to a boiling state, the water vaporizes and generates a large number of bubbles. Most of the liquid level gauges currently used in the sugar industry measure the pressure difference and then calculate the actual liquid level in combination with the density. The liquid level calculated by this method does not meet the requirements. It is necessary to design a method that can measure the liquid level including the bubble height.

[0003] Traditional methods of measuring liquid level mainly include pressure difference, ultrasound, float, radar, capacitance, etc., which can be used normally in general scenarios, but there are some problems with the above methods in the evaporation tank, and it is impossible to accurately obtain the liquid level value. A normal evaporation tank is equipped with a steam drum, and the sugar solution boils in the steam drum. The evaporation tank is high temperature and high pressure. Ultrasonic, float, radar and other measurement methods cannot be installed. Capacitive sensors may misjudge due to reasons such as boiling and splashing of liquid. The double flange level gauge can obtain the liquid level through the pressure difference between the upper and lower parts, but due to reasons such as boiling, the measured pressure may also deviate, resulting in inaccurate measurement.

[0004] In real scenes, a sight glass is usually installed on the tank body, and the position of the liquid in the sight glass is manually observed to determine the liquid level. This application provides a liquid level detection method for an evaporator based on machine vision, which uses a camera instead of human eyes to obtain a real-time liquid level image, and then obtains the final liquid level value through a video analysis algorithm to provide a basis for control.

[0005] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgment or any form of suggestion that the information constitutes the prior art already known to a person skilled in the art. Summary of the invention

[0006] The purpose of the present invention is to provide an evaporator liquid level detection method based on machine vision, which uses a camera instead of human eyes to obtain a real-time liquid level picture, and then obtains the final liquid level value through a video analysis algorithm, so as to solve the problem that the liquid level calculated by the traditional liquid level measurement method does not meet the requirements and the liquid level value cannot be accurately obtained.

[0007] In order to achieve the above object, the present invention provides the following technical solutions:

[0008] A method for detecting the liquid level of an evaporator tank based on machine vision comprises the following steps:

[0009] S1, obtaining the original image of the evaporation tank mirror;

[0010] S2, perform noise reduction processing on the original image, obtain the pixels of the monitoring area of ​​the original image, and divide the detection area according to the calibration;

[0011] S3, calculating the RGB value of each pixel in the divided area, converting the RGB value into an XYZ value, and then converting the XYZ value into a LAB value;

[0012] S4. According to the algorithm model of liquid level detection and recognition, each pixel is judged as having liquid or no liquid mark, and the number of pixels with liquid mark is counted;

[0013] S5, calculating the ratio of the number of pixels with liquid marks to the total number of pixels, and when the ratio reaches the liquid level recognition threshold, the divided area is regarded as a liquid level reaching area;

[0014] S6. Comprehensively calculate the highest divided area reached by the liquid in a single sight glass, and find the liquid level scale calibrated in the area;

[0015] S7, repeat the above steps until the liquid level scales reached by all the sight glasses on the evaporation tank are analyzed;

[0016] S8. Comprehensively calculate the liquid level scales reached by all sight glasses, and judge the current liquid level value in combination with the liquid level values ​​of the previous three tests.

[0017] Preferably, before the liquid level detection, the evaporator liquid level detection application system is manually operated to capture images of all mirrors, and the liquid level identification area and liquid level scale on the mirror image are calibrated, so that all mirrors in the same evaporator form a complete liquid level scale, and an algorithm model for liquid level detection and identification is established, and a liquid level color value is calculated by the following formula, and the liquid level color value is compared with the set value; when the liquid level color value is greater than the set value, it is considered that the pixel point has a liquid level. The calculation formula for the liquid level color value is as follows:

[0018] Liquid level color value = coefficient 1*L value + coefficient 2*A value + coefficient 3*B value

[0019] Among them, coefficient 1, coefficient 2, and coefficient 3 are set constant values; L value, A value, and B value are LAB values ​​calculated by the method described above.

[0020] Preferably, a large number of original images need to be collected before liquid level detection, the LAB values ​​are analyzed, and the algorithm model for liquid level detection and recognition is trained with the LAB values ​​and liquid and liquid-free classification calibration as parameters.

[0021] Preferably, three sight glasses are installed on the wall of the evaporator, a camera bracket is installed on the side of the evaporator, and the industrial camera is installed on the camera bracket; an industrial camera is installed in front of each sight glass, and the camera lens is facing the sight glass, 30cm~40cm away from the sight glass; the industrial camera is connected to the network environment where the evaporator liquid level detection application system is located through a network cable, and a fill light is installed next to the industrial camera.

[0022] Compared with the prior art, the present invention has the following beneficial effects: the evaporator liquid level detection method based on machine vision of the present invention realizes non-contact evaporator liquid level recognition, solves the problem of inaccurate liquid level recognition caused by complex evaporation conditions, and reduces the difficulty and economic cost of construction, providing an accurate control indicator basis for the evaporation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a schematic diagram of the positions of the evaporation tank sight glass and the industrial camera of the present invention;

[0024] Figure 2 It is a schematic flow chart of the method of the present invention;

[0025] Figure 3 It is a schematic diagram of dividing the detection area of ​​the present invention;

[0026] Figure 4 It is a schematic diagram for calibrating the liquid level recognition area and the liquid level scale on the mirror image. DETAILED DESCRIPTION

[0027] The following is a clear and complete description of the technical solution of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0028] In the description of the present invention, it should be noted that the orientations or positional relationships indicated by terms such as "center", "up", "down", "left", "right", "vertical", "horizontal", "inside" and "outside" are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the invention.

[0029] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0030] See attached Figure 2 , a method for detecting liquid level of an evaporator tank based on machine vision, comprising the following steps:

[0031] S1, obtaining the original image of the evaporation tank mirror;

[0032] S2, perform noise reduction processing on the original image, remove interference pixels, obtain pixels of the monitoring area of ​​the original image, and divide the detection area according to the calibration;

[0033] S3, calculating the RGB value of each pixel in the divided area, converting the RGB color space value into an XYZ value, and then converting the XYZ value into a LAB color space value;

[0034] S4, inputting the LAB color space value of the calibrated area on the sight glass into the algorithm model of liquid level detection and recognition, determining whether the pixel is a pixel with liquid or without liquid, and counting the number of pixels with liquid identification;

[0035] S5, calculating the pixel ratio of each area demarcated by the mirror with liquid, and if it exceeds the set ratio threshold, it is marked as having liquid, and if it does not exceed the set ratio threshold, it is marked as having no liquid;

[0036] S6. Comprehensively calculate the highest divided area reached by the liquid in a single sight glass, and find the liquid level scale calibrated in the area;

[0037] S7, repeat the above steps until the liquid level scales reached by all the sight glasses on the evaporation tank are analyzed;

[0038] S8. Calculate the liquid level scales reached by all sight glasses comprehensively, and judge the current liquid level value in combination with the liquid level values ​​of the previous three tests. The previous three tests here refer to the last test, the last test before, and the last test before that based on the current test.

[0039] In this embodiment, three sight glasses are installed on the wall of the evaporation tank at different levels. Figure 1 As shown, it is convenient to manually observe the liquid level and evaporation in the tank. A camera bracket is installed on the side of the evaporation tank, and an industrial camera is installed on the camera bracket. An industrial camera is installed in front of each mirror, and the camera lens is facing the mirror, and the camera lens is 30cm~40cm away from the mirror. The industrial camera is connected to the network environment where the evaporation tank liquid level detection application system that implements the above liquid level detection logic is located through a network cable, and the fill light is installed next to the camera. The industrial camera in this embodiment adopts the Hikvision network camera DS-2CD2625FFV4-LZS 2.7-12mm.

[0040] The evaporator tank liquid level detection application system is connected to an industrial camera. Before the liquid level detection, the application system is manually operated to capture images of all mirrors, and the liquid level recognition area and liquid level scale on the mirror image are calibrated, so that all mirrors in the same evaporator tank form a complete liquid level scale, and an algorithm model for liquid level detection and recognition is established.

[0041] During the detection process, the evaporator level detection application system automatically controls the camera to capture the original images of all mirrors in real time, and uploads them to the evaporator level detection application system, analyzes the liquid level picture, performs noise reduction on the picture, analyzes the RGB values ​​of all pixels in the picture, and converts them into LAB color space. Then, the height position of the liquid level in each mirror is calculated according to the algorithm model of liquid level detection and recognition. Finally, according to certain rules, the liquid level data on all mirrors are combined to obtain the final liquid level, and the final liquid level recognition height and the captured mirror picture are pushed to the front-end page of the evaporator level detection application system for display. The rule mentioned here means: compare all mirrors with liquid levels, record the highest mirror liquid level at the installation position, and then compare it with the liquid levels detected in the previous three times to calculate the deviation; if the deviation is less than the set value, the liquid level will be used as the final liquid level of this detection; if the deviation is greater than the set value, the next highest mirror liquid level will be taken for comparison, and so on.

[0042] Aiming at the actual working conditions during the use of the evaporator, the evaporator liquid level detection method based on machine vision of the present invention uses machine vision technology to realize the measurement of the liquid level height of the evaporator, and according to the actual situation on site, creatively uses color difference as a condition for judging whether there is a liquid level, and designs a corresponding judgment algorithm to obtain a better measurement effect.

[0043] The machine vision-based evaporation tank liquid level detection method of the present invention can realize non-contact evaporation tank liquid level recognition, solve the problem of inaccurate liquid level recognition caused by complex evaporation conditions, and reduce the difficulty and economic cost of construction, providing an accurate control indicator basis for the evaporation process.

[0044] The foregoing description of specific exemplary embodiments of the present invention is for the purpose of illustration and demonstration. These descriptions are not intended to limit the present invention to the precise form disclosed, and it is clear that many changes and variations can be made based on the above teachings. The purpose of selecting and describing the exemplary embodiments is to explain the specific principles of the present invention and its practical application, so that those skilled in the art can realize and utilize various different exemplary embodiments of the present invention and various different selections and changes. The scope of the present invention is intended to be limited by the claims and their equivalents.

Claims

1. A method for detecting liquid level of an evaporator based on machine vision, characterized in that: The following steps are involved: S1, obtaining the original image of the evaporation tank mirror; S2, perform noise reduction processing on the original image, obtain the pixels of the monitoring area of ​​the original image, and divide the detection area according to the calibration; S3, calculating the RGB value of each pixel in the divided area, converting the RGB value into an XYZ value, and then converting the XYZ value into a LAB value; S4. According to the algorithm model of liquid level detection and recognition, each pixel is judged as having liquid or no liquid mark, and the number of pixels with liquid mark is counted; S5, calculating the ratio of the number of pixels with liquid marks to the total number of pixels, and when the ratio reaches the liquid level recognition threshold, the divided area is regarded as a liquid level reaching area; S6. Comprehensively calculate the highest divided area reached by the liquid in a single sight glass, and find the liquid level scale calibrated in the area; S7, repeat the above steps until the liquid level scales reached by all the sight glasses on the evaporation tank are analyzed; S8. Comprehensively calculate the liquid level scales reached by all sight glasses, and judge the current liquid level value in combination with the liquid level values ​​of the previous three tests.

2. The method for detecting liquid level of an evaporator based on machine vision according to claim 1, characterized in that: Before liquid level detection, the evaporator liquid level detection application system is manually operated to capture images of all mirrors, and the liquid level recognition area and liquid level scale on the mirror image are calibrated, so that all mirrors in the same evaporator form a complete liquid level scale, and an algorithm model for liquid level detection and recognition is established.

3. The method for detecting liquid level of an evaporator based on machine vision according to claim 2, characterized in that: Before liquid level detection, a large number of original images need to be collected, the LAB value needs to be analyzed, and the algorithm model for liquid level detection and recognition needs to be trained using the LAB value and liquid / liquid classification calibration as parameters.

4. The method for detecting liquid level of an evaporator based on machine vision according to claim 2, characterized in that: Three sight glasses are installed on the wall of the evaporator tank, a camera bracket is installed on the side of the evaporator tank, and the industrial camera is installed on the camera bracket; an industrial camera is installed in front of each sight glass, and the camera lens faces the sight glass and is 30cm~40cm away from the sight glass; the industrial camera is connected to the network environment where the evaporator tank liquid level detection application system is located through a network cable, and the fill light is installed next to the industrial camera.

Images