An intelligent verification system and method for marine filling machine based on machine vision

Abstract

The present invention discloses an intelligent calibration system and method for a marine filling machine based on machine vision. The system comprises a marine filling machine, a calibration device, a machine vision module and an intelligent terminal. The calibration device comprises a calibration liquid adding pipeline and a calibration air return pipeline. The inlet end of the calibration liquid adding pipeline serves as a liquid adding port to be connected with a liquid adding gun. A first ultrasonic flow measurement device is clamped in the calibration liquid adding pipeline. A second ultrasonic flow measurement device is clamped in the calibration air return pipeline, and the outlet end of the calibration air return pipeline serves as an air return port to be connected with the air return gun. The present invention automatically identifies the flow display value of the marine filling machine by utilizing a machine vision module, and automatically controls the flow of the marine filling machine at the same time. Without damaging the pipeline, the ultrasonic LNG flow measurement technology is used to measure the fluid flow in the pipeline as a standard value, so as to realize automatic reading and automatic calibration of the flow of the marine filling machine, prevent calibration risks, ensure the safety of calibration personnel, and improve calibration efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of marine filling machine calibration equipment, and in particular to a marine filling machine intelligent calibration system and method based on machine vision. Background Art

[0002] Marine filling machine is an ultra-low temperature metering system for filling fuel into power ships. Marine filling machine is a metering instrument for trade settlement between fuel supply and demand parties. Its metering accuracy and reliability are key indicators for measuring the quality of filling machine. At present, domestic metrology verification institutions and ship manufacturers have not yet carried out the verification of marine filling machines, lacking safe verification methods and devices. In addition, LNG is flammable and explosive, and there are safety hazards in the on-site verification operation of technicians. Summary of the invention

[0003] The purpose of the present invention is to overcome the deficiencies of the above-mentioned prior art and provide a system and method for intelligent calibration of marine filling machines based on machine vision, so as to realize intelligent calibration of marine filling machines, reduce potential safety hazards and ensure the safety of technicians.

[0004] To achieve the above object, the technical solution of the present invention is:

[0005] In a first aspect, the present invention provides a machine vision-based intelligent verification system for a marine filling machine, comprising a marine filling machine, a verification device, a machine vision module, and an intelligent terminal;

[0006] The marine filling machine comprises a filling pipeline and a filling return pipeline; a liquid mass flowmeter is arranged in the filling pipeline, and a liquid filling gun is installed at the outlet end of the filling pipeline; a pneumatic valve is arranged between the liquid filling gun and the liquid mass flowmeter in the filling pipeline; a gas mass flowmeter is arranged in the return pipeline, and a return gun is installed at the inlet end of the return pipe; the liquid mass flowmeter and the gas mass flowmeter are collectively referred to as mass flowmeters;

[0007] The calibration device comprises a calibration liquid adding pipeline and a calibration air return pipeline; the inlet end of the calibration liquid adding pipeline serves as a liquid adding port to be connected to a liquid adding gun; a first ultrasonic flow measuring device is clamped in the calibration liquid adding pipeline; a second ultrasonic flow measuring device is clamped in the calibration air return pipeline, and the outlet end of the calibration air return pipeline serves as an air return port to be connected to an air return gun; the first ultrasonic flow measuring device and the second ultrasonic flow measuring device are collectively referred to as a standard device;

[0008] The machine vision module is used to identify and obtain data from the mass flow meter, and transmit the obtained data to the intelligent terminal, which controls and adjusts the pneumatic valve to keep the flow stable;

[0009] The data detected by the standard device is transmitted to the smart terminal through the signal receiving and processing module. The smart terminal compares and calculates the data detected by the mass flow meter and the data detected by the standard device to automatically verify the accuracy and repeatability of the filling machine.

[0010] Furthermore, the first ultrasonic flow measurement device and the second ultrasonic flow measurement device have the same structure, and both include two transducers, and the two transducers are respectively installed on opposite sides of the pipeline.

[0011] Furthermore, the first ultrasonic flow measurement device and the second ultrasonic flow measurement device both detect and obtain flow data in the pipeline by using ultrasonic LNG flow measurement technology:

[0012] The ultrasonic signal is emitted by a transducer installed on one side of the pipeline, and is received by another transducer after passing through the opposite pipe wall. The ultrasonic signal is emitted alternately in the downstream and upstream directions. The ultrasonic signal propagates in the flowing medium, and the propagation time in the downstream direction is shorter than the propagation time in the upstream direction. The average flow velocity of the fluid on the ultrasonic signal propagation path is determined by measuring the propagation time difference. The measured path average flow velocity is corrected according to the cross-sectional distribution state of the fluid flow velocity in the pipeline, and the cross-sectional average flow velocity proportional to the volume flow rate, that is, the flow data in the pipeline, can be obtained.

[0013] Furthermore, the propagation time of the ultrasonic signal in the downstream direction is (1-1)

[0014] The propagation time of ultrasonic signal in the countercurrent direction (1-2)

[0015] Propagation time difference for: (1-3)

[0016] Ultrasound mean propagation time (1-4)

[0017] Speed ​​of sound (1-5)

[0018] Flow rate (1-6)

[0019] flow (1-7)

[0020] In formulas (1-1) to (1-7), P is the ultrasonic wave propagation path length; L is the distance between the two ultrasonic flow measurement devices; C is the speed of sound under the measurement conditions; It is the angle between the pipe and the vertical pipe section.

[0021] Furthermore, the machine vision-based intelligent verification system for marine filling machines also includes a temperature sensor and a pressure sensor to detect the temperature and pressure of LNG in the filling pipeline to compensate for the temperature and pressure parameters in a timely manner; under standard working conditions, the flow rate of the marine filling machine is compared with the flow rate of the standard device. Under standard conditions, the mass flow rate is:

[0022] (1-8)

[0023] Where:

[0024] is the mass flow rate; is the molar mass; It is the stress of working conditions; is the thermodynamic temperature under operating conditions; is the compression factor under working conditions; is the universal gas constant.

[0025] Furthermore, a check valve is provided in the filling return air pipeline.

[0026] Furthermore, the marine filling machine also includes an electronic control and display system, which is connected to the liquid mass flow meter and the gas mass flow meter respectively to display the detected flow data.

[0027] Furthermore, the machine vision module identifies and obtains the data of the mass flow meter in the following manner:

[0028] Obtain a mass flow meter reading image, and use an improved threading method to identify the reading numbers in the image. The improved threading method includes:

[0029] First scan from the midpoint of the upper boundary of the number from top to bottom, then scan from the midpoint of 3 / 4 of the left boundary of the number and the midpoint of 1 / 4 of the left boundary from left to right respectively. Determine the type of number based on the results of the three scans to obtain the cumulative flow indication of the mass flow meter.

[0030] Furthermore, the method of determining the digital type based on the results of three scans to obtain the cumulative flow indication of the mass flow meter includes:

[0031] The three scans respectively obtain the corresponding counting variable values ​​count1, count2 and count3. Under the initial conditions, count1, count2 and count3 are all assigned 0; in the first scan, when the pixel value of the first row is 0 or the pixel value of the previous row is 255 and the pixel value of this row is 0, the counting variable count1 is increased by 1; in the second scan, when the pixel value of the first column is 0, the counting variable count2 is increased by 1, and when the pixel value of the previous column is 255 and the pixel value of this column is 0, the counting variable count2 is increased by 10; in the third scan, when the pixel value of the first column is 0, the counting variable count3 is increased by 1, and when the pixel value of the previous column is 255 and the pixel value of this column is 0, the counting variable count3 is increased by 10; according to the counting rules of the above variables and the characteristics of 0-9 numbers, the type of number is judged to obtain the cumulative flow indication of the mass flowmeter.

[0032] In a second aspect, the present invention provides an intelligent verification method for a marine filling machine based on machine vision, the method is the above-mentioned on-site verification system, and the method comprises:

[0033] (1) Connect the liquid filling gun and air return gun of the marine filling machine in series with the inlet end of the calibration liquid filling pipeline and the outlet end of the calibration air return pipeline respectively;

[0034] (2) Power on and preheat the marine filling machine equipment;

[0035] (3) Start the calibration device to perform a cyclic precooling process, and the cryogenic liquid flows through the standard device and the mass flow meter for precooling;

[0036] (4) During the pre-cooling cycle, observe the cryogenic liquid flow parameters displayed by the electronic control and display system. When the test conditions are met, adjust the circulation flow of the calibration device to the set flow value according to the set test flow point, and run until the fluid state is stable;

[0037] (5) After the fluid state is stable, the flow signals of the standard device and the mass flow meter are collected synchronously, and the cumulative mass flow output by the standard device and the mass flow meter under test is recorded within one minute. The cumulative mass flow value output by the standard device is compared with the cumulative mass flow value of the mass flow meter under test.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] The present invention proposes an intelligent calibration system and method for a marine filling machine based on machine vision, which automatically identifies the flow display value of the marine filling machine by using a machine vision module, and automatically controls the flow of the marine filling machine. Without damaging the pipeline, the ultrasonic measurement of LNG flow technology uses the ultrasonic propagation velocity difference method to indirectly measure the fluid flow in the pipeline as a standard value, thereby realizing automatic reading and intelligent calibration of the flow of the marine filling machine, preventing calibration risks, ensuring personnel safety, improving calibration efficiency, liberating calibration personnel, and saving time and cost. It avoids the safety risks of directly measuring the fluid by damaging the pipeline, and the shortcomings of the traditional calibration process in which the readings of the instrument being tested cannot be automatically collected, and manual reading, recording, and calculation are required. The original records and certificates are prone to errors, the calibration process is time-consuming, the calibration capacity is insufficient, and the labor cost is high. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A schematic diagram of the composition of the machine vision-based intelligent detection system for marine filling machines provided by an embodiment of the present invention when in use;

[0041] Figure 2 This is the working principle diagram of ultrasonic flow measurement;

[0042] Figure 3 This is a schematic diagram of online temperature and pressure compensation and data acquisition;

[0043] In the figure: 1. Marine filling machine; 2. Calibration device; 3. Machine vision module; 4. Intelligent terminal; 11. Liquid mass flow meter; 12. Filling gun; 13. Pneumatic valve; 14. Gas mass flow meter; 15. Check valve; 16. Electronic control and display system; 21. First ultrasonic flow measurement device; 22. Second ultrasonic flow measurement device; 23. Signal receiving and processing module; 200. Transducer. DETAILED DESCRIPTION

[0044] Example:

[0045] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0046] See also Figure 1 As shown, the composition diagram of a machine vision-based intelligent verification system for a marine filling machine provided in this embodiment when in use mainly includes a marine filling machine 1, a verification device 2, a machine vision module 3 and an intelligent terminal 4. In this embodiment, the intelligent terminal is a laptop computer.

[0047] The marine filling machine comprises a filling pipeline and a filling return air pipeline; a liquid mass flowmeter 11 is arranged in the filling pipeline, and a liquid filling gun 12 is installed at the outlet end of the filling pipeline; in the filling pipeline, a pneumatic valve 13 is arranged between the liquid filling gun 12 and the liquid mass flowmeter 11, and the amount of gas added is adjusted by controlling the pneumatic valve 13; a gas mass flowmeter 14 is arranged in the return air pipeline, and a return air gun 15 is installed at the inlet end of the return air pipe. In order to avoid cumbersome description, the liquid mass flowmeter 11 and the gas mass flowmeter 14 are collectively referred to as mass flowmeters.

[0048] The calibration device includes a calibration liquid adding pipeline and a calibration air return pipeline; the inlet end of the calibration liquid adding pipeline serves as a liquid adding port to communicate with the liquid adding gun 12, and a first ultrasonic flow measuring device 21 is clamped in the calibration liquid adding pipeline; a second ultrasonic flow measuring device 22 is clamped in the calibration air return pipeline, and the outlet end of the calibration air return pipeline serves as an air return port to communicate with the air return gun 15; that is, the calibration device and the marine filling machine are connected in series. Similarly, in order to avoid redundancy, the first ultrasonic flow measuring device 21 and the second ultrasonic flow measuring device 22 are collectively referred to as standard devices.

[0049] The machine vision module 3 is used to identify and obtain the data of the gas mass flowmeter and the liquid mass flowmeter, and transmit the obtained data back to the intelligent terminal 4, which controls and adjusts the pneumatic valve 13 to keep the flow stable. In other words, the machine vision module 3 is used to automatically identify the flow display value of the marine filling machine, and automatically control the flow of the marine filling machine. The data detected by the liquid standard mass flowmeter and the gas standard mass flowmeter are transmitted to the intelligent terminal. The intelligent terminal can automatically calibrate the accuracy and repeatability of the filling machine based on the data detected by the mass flowmeter and the data detected by the standard device, thereby improving the calibration efficiency, liberating the calibration personnel, saving time and cost, and avoiding the shortcomings of the inability to automatically collect the readings of the instrument being tested in the traditional calibration process, requiring manual reading, recording, and calculation, and the original records and certificates are prone to errors. The calibration process is time-consuming, the calibration capacity is insufficient, and the labor cost is high.

[0050] In a specific embodiment, if Figure 2 As shown, the first ultrasonic flow measurement device 21 and the second ultrasonic flow measurement device 22 have the same structure, and both include two transducers 200, which are respectively installed on opposite sides of the pipeline. The first ultrasonic flow measurement device and the second ultrasonic flow measurement device both detect and obtain flow data in the pipeline in the following manner:

[0051] The ultrasonic signal is emitted by a transducer installed on one side of the pipeline, and is received by another transducer after passing through the opposite side of the pipeline wall. The ultrasonic signal is emitted alternately in the downstream and upstream directions. The ultrasonic signal propagates in the flowing medium, and the propagation time in the downstream direction is shorter than the propagation time in the upstream direction. By measuring the propagation time difference, the average flow velocity of the fluid on the ultrasonic signal propagation path is determined. The measured average flow velocity of the path is corrected according to the cross-sectional distribution state of the fluid flow velocity in the pipeline, and the cross-sectional average flow velocity proportional to the volume flow rate, that is, the flow data in the pipeline, can be obtained. In this way, there is no need to contact with the medium, no pressure loss, easy installation and maintenance, safe for LNG measurement, and bidirectional measurement can be performed with high accuracy and high repeatability.

[0052] Specifically, Figure 3 As shown in the figure, the propagation time of the ultrasonic signal in the downstream direction is (1-1)

[0053] The propagation time of ultrasonic signal in the countercurrent direction (1-2)

[0054] Propagation time difference for: (1-3)

[0055] Ultrasound mean propagation time (1-4)

[0056] Speed ​​of sound (1-5)

[0057] Flow rate (1-6)

[0058] flow (1-7)

[0059] In formulas (1-1) to (1-7), P is the ultrasonic wave propagation path length; L is the distance between the two ultrasonic flow measurement devices; C is the speed of sound under the measurement conditions; It is the angle between the pipe and the vertical pipe section.

[0060] Thus, the above calculation is performed in the signal receiving and processing module 23, so that the flow data in the pipeline can be accurately obtained.

[0061] The temperature and pressure of LNG in the actual pipeline vary to varying degrees, and the temperature and pressure parameters need to be compensated in time. To this end, the intelligent verification system for marine filling machines based on machine vision also includes temperature sensors and pressure sensors to detect the temperature and pressure of LNG in the filling pipeline to compensate for the temperature and pressure parameters in time; under standard working conditions, the flow rate of the marine filling machine is compared with the flow rate of the standard device. Under standard conditions, the mass flow rate is:

[0062] (1-8)

[0063] Where:

[0064] is the mass flow rate; is the molar mass; It is the stress of working conditions; is the thermodynamic temperature under operating conditions; is the compression factor under working conditions; is the universal gas constant.

[0065] In a specific embodiment, a check valve 15 is also provided in the filling return air pipeline to adjust the amount of return air by controlling the check valve; in addition, the marine filling machine also includes an electronic control and display system 16, which is respectively connected to the liquid mass flow meter and the gas mass flow meter to display the detected flow data. At the same time, the machine vision module can also automatically identify the flow display value of the marine filling machine by photographing the flow data displayed by the electronic control and display system. The on-site calibration data obtained by the smart terminal can also be transmitted to the cloud platform in real time; the calibration history data can be queried in real time on the web and app; and the remote calibration function can be performed.

[0066] In a specific embodiment, the machine vision module identifies and obtains the data of the mass flow meter in the following manner:

[0067] Obtain a mass flow meter reading image, and use an improved threading method to identify the reading numbers in the image. The improved threading method includes:

[0068] First scan from the midpoint of the upper boundary of the number from top to bottom, then scan from the midpoint of the 3 / 4 of the left boundary of the number and the midpoint of the 1 / 4 of the left boundary from left to right, determine the type of number based on the results of the three scans, and get the cumulative flow indication of the mass flow meter, including:

[0069] The three scans respectively obtain the corresponding counting variable values ​​count1, count2 and count3. Under the initial conditions, count1, count2 and count3 are all assigned 0; in the first scan, when the pixel value of the first row is 0 or the pixel value of the previous row is 255 and the pixel value of this row is 0, the counting variable count1 is increased by 1; in the second scan, when the pixel value of the first column is 0, the counting variable count2 is increased by 1, and when the pixel value of the previous column is 255 and the pixel value of this column is 0, the counting variable count2 is increased by 10; in the third scan, when the pixel value of the first column is 0, the counting variable count3 is increased by 1, and when the pixel value of the previous column is 255 and the pixel value of this column is 0, the counting variable count3 is increased by 10; according to the counting rules of the above variables and the characteristics of 0-9 numbers, the type of number is judged to obtain the cumulative flow indication of the mass flowmeter.

[0070] In this way, the image recognition algorithm can be greatly simplified, the running time of the image recognition algorithm can be shortened, the efficiency of automatic reading can be improved, the verification efficiency can be greatly improved, and the quality and reliability of the reading can be improved.

[0071] Accordingly, this embodiment also provides a method for intelligent verification of a marine filling machine based on machine vision, based on the above-mentioned on-site verification system, mainly including the following steps:

[0072] (1) Connect the liquid filling gun and the air return gun of the marine filling machine in series with the inlet end of the calibration liquid filling pipeline and the outlet end of the calibration air return pipeline respectively; that is, connect the calibration device in series with the marine filling machine;

[0073] (2) Marine filling machine equipment should be well grounded and the power-on preheating time should be no less than 30 minutes;

[0074] (3) Start the calibration device to perform a cyclic precooling process, and the cryogenic liquid flows through the standard device and the mass flow meter for precooling;

[0075] (4) During the pre-cooling cycle, observe the cryogenic liquid flow parameters displayed by the electronic control and display system. When the test conditions are met, adjust the circulation flow of the calibration device to the set flow value according to the set test flow point, and run until the fluid state is stable;

[0076] (5) After the fluid state is stable, the flow signals of the standard device and the mass flow meter are collected synchronously, and the cumulative mass flow output by the standard device and the mass flow meter under test is recorded within one minute. The cumulative mass flow value output by the standard device is compared with the cumulative mass flow value of the mass flow meter under test. The maximum error and repeatability of the marine filling machine are obtained based on the test data analysis.

[0077] In summary, the present invention connects the calibration device in series with the marine filling machine, uses the machine vision module to automatically identify the flow display value of the marine filling machine, and automatically controls the flow of the marine filling machine; without damaging the pipeline, the ultrasonic measurement of LNG flow technology uses the ultrasonic propagation velocity difference method to indirectly measure the fluid flow in the pipeline as the standard value, and detects the accuracy and repeatability of the marine filling machine metering equipment through the filling process, so as to realize the automatic reading and automatic calibration of the flow of the marine filling machine. The purpose of the present invention is to solve the problem that there is no calibration method and device for marine filling machines to carry out calibration work. The development of the present invention is conducive to solving the problem of metrological calibration guarantee in the field of marine LNG filling, providing accurate metering services for port LNG filling companies, and maintaining the metering fairness of marine LNG filling. Through the innovation of metrological calibration technology, the efficiency of metrological calibration can be greatly improved, costs can be saved, and the safety of calibration personnel can be guaranteed.

[0078] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable ordinary technicians in the field to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made based on the essence of the content of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent verification system for marine filling machines based on machine vision, characterized in that: Including marine filling machines, calibration devices, machine vision modules and intelligent terminals; The marine filling machine comprises a filling pipeline and a filling return pipeline; a liquid mass flowmeter is arranged in the filling pipeline, and a liquid filling gun is installed at the outlet end of the filling pipeline; a pneumatic valve is arranged between the liquid filling gun and the liquid mass flowmeter in the filling pipeline; a gas mass flowmeter is arranged in the return pipeline, and a return gun is installed at the inlet end of the return pipe; the liquid mass flowmeter and the gas mass flowmeter are collectively referred to as mass flowmeters; The calibration device comprises a calibration liquid adding pipeline and a calibration air return pipeline; the inlet end of the calibration liquid adding pipeline serves as a liquid adding port to be connected to a liquid adding gun; a first ultrasonic flow measuring device is clamped in the calibration liquid adding pipeline; a second ultrasonic flow measuring device is clamped in the calibration air return pipeline, and the outlet end of the calibration air return pipeline serves as an air return port to be connected to an air return gun; the first ultrasonic flow measuring device and the second ultrasonic flow measuring device are collectively referred to as a standard device; The machine vision module is used to identify and obtain data from the mass flow meter, and transmit the obtained data to the intelligent terminal, which controls and adjusts the pneumatic valve to keep the flow stable; The data detected by the standard device is transmitted to the intelligent terminal through the signal receiving and processing module, and the intelligent terminal compares and calculates the data detected by the mass flow meter and the data detected by the standard device to automatically verify the accuracy and repeatability of the filling machine; The first ultrasonic flow measurement device and the second ultrasonic flow measurement device have the same structure, and both include two transducers, which are respectively installed on opposite sides of the pipeline; The first ultrasonic flow measurement device and the second ultrasonic flow measurement device both use ultrasonic LNG flow measurement technology to detect and obtain flow data in the pipeline: The ultrasonic signal is emitted by a transducer installed on one side of the pipeline, and is received by another transducer after passing through the opposite side of the pipeline wall. The ultrasonic signal is emitted alternately in the downstream and upstream directions. The ultrasonic signal propagates in the flowing medium, and the propagation time in the downstream direction is shorter than the propagation time in the upstream direction. The average flow velocity of the fluid on the ultrasonic signal propagation path is determined by measuring the propagation time difference. The measured path average flow velocity is corrected according to the cross-sectional distribution state of the fluid flow velocity in the pipeline, and the cross-sectional average flow velocity proportional to the volume flow rate, that is, the flow data in the pipeline, can be obtained. The propagation time of the ultrasonic signal in the downstream direction (1-1) The propagation time of ultrasonic signal in the upstream direction (1-2) Propagation time difference for: (1-3) Ultrasound mean propagation time (1-4) Speed ​​of sound (1-5) Flow rate (1-6) flow (1-7) In formulas (1-1) to (1-7), P is the ultrasonic wave propagation path length; L is the distance between the two ultrasonic flow measurement devices; C is the speed of sound under the measurement conditions; It is the angle between the pipe and the vertical pipe section.

2. The machine vision-based intelligent verification system for marine filling machines according to claim 1, characterized in that: It also includes temperature sensors and pressure sensors to detect the temperature and pressure of LNG in the filling pipeline to compensate for the temperature and pressure parameters in time; under standard working conditions, the flow rate of the marine filling machine is compared with the flow rate of the standard device. Under standard conditions, the mass flow rate is: (1-8) Where: is the mass flow rate; is the molar mass; It is the stress of working conditions; is the thermodynamic temperature under operating conditions; is the compression factor under working conditions; is the universal gas constant.

3. The machine vision-based intelligent verification system for marine filling machines according to claim 1, characterized in that: A check valve is also provided in the filling return air pipeline.

4. The machine vision-based intelligent verification system for marine filling machines according to claim 1, characterized in that: The marine filling machine also includes an electronic control and display system, which is connected to the liquid mass flow meter and the gas mass flow meter respectively to display the detected flow data.

5. The machine vision-based intelligent verification system for marine filling machines according to claim 1, characterized in that: The machine vision module identifies and obtains the data of the mass flow meter in the following manner: Obtain a mass flow meter reading image, and use an improved threading method to identify the reading numbers in the image. The improved threading method includes: First scan from the midpoint of the upper boundary of the number from top to bottom, then scan from the midpoint of 3 / 4 of the left boundary of the number and the midpoint of 1 / 4 of the left boundary from left to right respectively. Determine the type of number based on the results of the three scans to obtain the cumulative flow indication of the mass flow meter.

6. The machine vision-based intelligent verification system for marine filling machines according to claim 5, characterized in that: Determining the digital type according to the results of the three scans to obtain the cumulative flow indication of the mass flow meter includes: The three scans respectively obtain the corresponding counting variable values ​​count1, count2 and count3. Under the initial conditions, count1, count2 and count3 are all assigned 0; in the first scan, when the pixel value of the first row is 0 or the pixel value of the previous row is 255 and the pixel value of this row is 0, the counting variable count1 is increased by 1; in the second scan, when the pixel value of the first column is 0, the counting variable count2 is increased by 1, and when the pixel value of the previous column is 255 and the pixel value of this column is 0, the counting variable count2 is increased by 10; in the third scan, when the pixel value of the first column is 0, the counting variable count3 is increased by 1, and when the pixel value of the previous column is 255 and the pixel value of this column is 0, the counting variable count3 is increased by 10; according to the counting rules of the above variables and the characteristics of 0-9 numbers, the type of number is judged to obtain the cumulative flow indication of the mass flowmeter.

7. A method for intelligent verification of a marine filling machine based on machine vision, the method being based on the intelligent verification system of a marine filling machine according to claim 1, characterized in that: The method comprises: (1) Connect the liquid filling gun and air return gun of the marine filling machine in series with the inlet end of the calibration liquid filling pipeline and the outlet end of the calibration air return pipeline respectively; (2) Power on and preheat the marine filling machine (3) Start the calibration device to perform a cyclic precooling process, and the cryogenic liquid flows through the standard device and the mass flow meter for precooling; (4) During the pre-cooling cycle, observe the cryogenic liquid flow parameters displayed by the electronic control and display system. When the test conditions are met, adjust the circulation flow of the calibration device to the set flow value according to the set test flow point, and run until the fluid state is stable; (5) After the fluid state is stable, the flow signals of the standard device and the mass flow meter are collected synchronously, and the cumulative mass flow output by the standard device and the mass flow meter under test is recorded within one minute. The cumulative mass flow value output by the standard device is compared with the cumulative mass flow value of the mass flow meter under test.

Images