Multi-nozzle jet flow video dynamic monitoring system and method based on machine vision

Abstract

The invention provides a multi-nozzle jet flow video dynamic monitoring system and method based on machine vision, and relates to the field of industrial vision detection. The system comprises an experimental hardware platform and a visual processing device, the experimental hardware platform organically combines a nozzle array and a camera in space to ensure that a shot and sampled jet flow area is complete and comprehensive, and a sprayed water mist area can be visually highlighted to facilitate identification and measurement of a machine vision algorithm; and the visual processing device is used for carrying out instantaneous state and cumulative distribution measurement on the laser jet region of the nozzle array. According to the system, the machine vision technology is applied to nozzle jet flow dynamic monitoring, the limitation of a traditional detection method can be broken through, comprehensive measurement of the instantaneous state and cumulative distribution of jet flow is achieved, and therefore a scientific basis is provided for nozzle design optimization and cooling process control.

Description

Technical Field

[0001] This application relates to the field of industrial vision inspection, and in particular to a multi-nozzle jet video dynamic monitoring system and method based on machine vision. Background Technology

[0002] Nozzles, as important fluid jetting devices, are widely used in the steel industry, agricultural irrigation, papermaking industry, fire fighting, and other fields. Especially in the cooling process of hot-rolled steel plates, nozzle arrays forming spray boxes are used to uniformly cool the steel plate surface to ensure the mechanical properties and surface quality of the produced steel plate. Specifically, the instantaneous distribution and cumulative effect of the jetting water flow can alter the metallographic structure in thick plates, leading to changes in the overall strength of the steel plate. Factors influencing this distribution include the stability of flow control, the rationality of nozzle array selection and combination parameters, and interference from environmental factors. Detecting the jet distribution of a nozzle array is complex, influenced by the interaction of multiple individual nozzle jets and the control effect of water flow, making its global dynamic model difficult to simulate and solve. Traditional measurement methods include using simple flow meters or pressure sensors for multi-point local measurements. These methods have the following problems: First, contact flow meters or pressure sensors can only perform local measurements, making it difficult to fully acquire the spray characteristics of the entire nozzle array and accurately reflect the spatial distribution and temporal variation of spray density. Second, the real-time monitoring data of water volume lacks dynamism and completeness, making it difficult to comprehensively evaluate the spray characteristics of the nozzles, which affects the optimization of nozzle combination design and the precise control of the actual cooling process.

[0003] Machine vision-based inspection technology is gradually becoming an important means of industrial inspection. By selecting appropriate visual tasks and corresponding visual algorithms, real-time acquired industrial visual data can be analyzed and calculated, exhibiting high efficiency, accuracy, and repeatability. Currently, machine vision-based jet dynamic monitoring technology still faces the following technical challenges: first, the identification of keyframes and the suppression of environmental noise; second, how to eliminate interference from surrounding background objects for jet region identification; and third, how to design reasonable visual indicators for the instantaneous and cumulative distribution of the inverse jet. Summary of the Invention

[0004] In view of the above problems, this application is made to provide a machine vision-based multi-nozzle jet dynamic monitoring system and method that overcomes or at least partially solves the above problems. The technical solution is as follows: Firstly, a machine vision-based multi-nozzle jet dynamic monitoring system is provided, the system comprising: Hardware platform and vision processing device; The main body of the hardware platform is constructed from an alloy frame. The components of the hardware platform include nozzles of various types to be cooled, lasers, and cameras. The nozzles are mounted on a support that can move up, down, left, and right to adjust the jet height and nozzle spacing. The laser is set at a fixed height and is used to emit laser light laterally and irradiate the cross-section of the jet. The camera lens has a fixed viewing angle and is used to shoot the cross-section of the laser-irradiated jet from above and to acquire multi-nozzle laser jet video in real time. The visual processing device includes an image acquisition module, an image preprocessing module, a jet region recognition module, an instantaneous index measurement module, a cumulative distribution measurement module, and a visualization module; The image acquisition module is used to sample the multi-nozzle laser jet video captured by the camera to obtain the first image set; The image preprocessing module is used to perform preprocessing operations on each single frame image in the first image set to obtain the second image set; The jet region recognition module uses threshold segmentation and morphological filtering techniques to perform jet region recognition on each single frame image in the second image set to obtain the third image set. The instantaneous index measurement module is used to perform jet instantaneous state distribution detection on each single frame image in the third image set to obtain the fourth image set; The cumulative distribution module is used to perform jet cumulative flow distribution estimation on the fourth image set; The visualization module is used to visualize the instantaneous state distribution and cumulative flow distribution of the jet.

[0005] In one possible implementation, the sampling operation includes: Frames of the multi-nozzle laser jet video captured by the camera are selected at fixed intervals to obtain the first image set. , Satisfying equation (1): (1) In equation (1), For calculating time After that A set of images sampled at intervals. Controls the number of elements in the selected image set.

[0006] In one possible implementation, the preprocessing operations include: Select any one image from the first image set as the first target image. For the first target image Perform guided filtering operation: (2) In equation (2), This is the preprocessed image output after filtering. Let be any pixel coordinate value of the first target image. It is a local mean. The guiding image is used to direct the filtering direction. For local covariance, For local variance, It is a constant used to prevent numerical overflow; The local covariance calculation satisfies equation (3): (3) In equation (3), Therefore The neighborhood centered arbitrary pixel coordinates, For the neighborhood The number of pixels within the cell; The local variance calculation satisfies equation (4): (4) Substitute the calculation results of equations (3) and (4) into equation (2) to implement the guided filtering operation.

[0007] In one possible implementation, the jet region identification operation includes: Select any image from the second image set as the second target image. The green laser jet region in the second target image is identified using the following calculation formula: (5) In equation (5), To complete the third target image after jet region identification, To segment the binary result, the segmented binary result satisfies equation (6); (6) In equation (6), For pixels The value of the red channel, For pixels The value of the green channel, The critical value used to distinguish the jet from the background.

[0008] In one possible implementation, the jet instantaneous state distribution detection operation includes: Select any image from the third image set as the third target image. Using a linear model for the third target image The jet density distribution was simulated, and a linear model was established that satisfies equation (7): (7) In equation (7), This represents the amount of water passing through the nozzle per unit time. Let be the cross-sectional area of ​​the nozzle. The average pixel value of the region. and Let be the linear parameters to be fitted. and Calculate to satisfy equation (8): (8) Calculated according to equation (8) and Then, calculate the pixel reference density of the instantaneous jet region: (9) In equation (9), This is the pixel reference density for the final calculated transient jet region.

[0009] In one possible implementation, the jet cumulative flow distribution estimation operation includes: The pixel reference density of the transient jet region in consecutive frames of the fourth image set is cumulatively calculated along the bandwidth direction. First, the cumulative reference value of pixel flux is calculated: (10) In equation (10), Accumulate a reference value for the flow rate for each pixel. For the fourth image set, the first Pixel reference density of the transient jet region of the frame. The reference velocity for simulating the movement of the steel plate on the image; Pixel flow cumulative reference value Perform local window smoothing: (11) In equation (11), For Centered on, To calculate the side length of a square window, For The pixel coordinates within the center window; Calculate the cumulative flow distribution of the jet: (12) In equation (12), It is cumulatively distributed along the width direction of the steel plate; Reference velocity v and side length The calculation satisfies equation (13): (13) In equation (13), The actual speed of the steel plate as measured by physical measurement. For conversion ratios, The side length of a single test tube in a water-filled container, measured physically.

[0010] In one possible implementation, the visualization operations include: Instantaneous state distribution visualization: A heatmap is used to visualize the pixel reference density of the transient jet region in each frame of the fourth image set. Cumulative flow distribution visualization: The cumulative flow distribution of the jet is visualized using a bar chart.

[0011] Secondly, a machine vision-based method for dynamic monitoring of multi-nozzle jets is provided, the method comprising: A hardware platform is pre-built, the main body of which is constructed from an alloy frame. The components of the hardware platform include nozzles of various types to be tested for cooling, lasers, and cameras. The nozzles are mounted on supports that can move up, down, left, and right to adjust the jet height and nozzle spacing. The laser is set at a fixed height to emit laser light laterally and irradiate the cross-section of the jet. The camera lens has a fixed viewing angle to capture the cross-section of the laser-irradiated jet from above and to acquire real-time video of the multi-nozzle laser jet. The first image set is obtained by sampling the multi-nozzle laser jet video captured by the camera; The second image set is obtained by performing preprocessing operations on each single frame image in the first image set; The third image set is obtained by performing jet region identification on each single frame image in the second image set using threshold segmentation and morphological filtering techniques. A fourth image set is obtained by performing jet instantaneous state distribution detection on each single frame image in the third image set. Perform jet cumulative flux distribution estimation on the fourth image set; Visualize the instantaneous state distribution and cumulative flow distribution of the jet.

[0012] Thirdly, an electronic device is provided, comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform a machine vision-based multi-nozzle jet dynamic monitoring method.

[0013] Fourthly, a storage medium is provided that stores a computer program, wherein the computer program is configured to execute a machine vision-based multi-nozzle jet dynamic monitoring method at runtime.

[0014] Based on the above technical solutions, this application provides a machine vision-based multi-nozzle jet video dynamic monitoring system and method. The system includes an experimental hardware platform and a vision processing device. The experimental hardware platform organically combines the nozzle array and camera in space to ensure the completeness and comprehensiveness of the sampled jet area, visually highlighting the sprayed water mist area for recognition and measurement by machine vision algorithms. The vision processing device is used to measure the instantaneous state and cumulative distribution of the laser jet area of ​​the nozzle array. First, key frames of the video are detected and extracted. Then, preprocessing algorithms are used to suppress noise and exposure interference, and image segmentation technology is used to dynamically identify the jet area. Finally, transient density distribution and cumulative distribution indices along the strip width are calculated. This system applies machine vision technology to the dynamic monitoring of nozzle jets, overcoming the limitations of traditional detection methods and achieving comprehensive measurement of the instantaneous state and cumulative distribution of the jet, thereby providing a scientific basis for nozzle design optimization and cooling process control. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below.

[0016] Figure 1 This paper shows a schematic diagram of the hardware platform of the machine vision-based multi-nozzle jet dynamic monitoring system provided in an embodiment of this application. Figure 2 The visual processing device interface of the machine vision-based multi-nozzle jet dynamic monitoring system provided in this application embodiment is shown. Figure 3a and Figure 3b The images before and after preprocessing operations provided in a specific embodiment of this application are shown; Figure 4a , Figure 4b , Figure 4c and Figure 4d This illustration shows a visualization diagram of the instantaneous state distribution provided in a specific embodiment of this application; Figure 5a and Figure 5b This illustration shows a visualization of the cumulative traffic distribution provided in a specific embodiment of this application; Figure 6 A flowchart illustrating the steps of a machine vision-based multi-nozzle jet dynamic monitoring method provided in a specific embodiment of this application is shown. Detailed Implementation

[0017] Exemplary embodiments of the present application will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the scope of the present application to those skilled in the art.

[0018] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such use can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the term "comprising" and its variations should be interpreted as open-ended terms meaning "including but not limited to."

[0019] Currently, machine vision-based jet dynamic monitoring technology still faces the following technical challenges: first, the judgment of key frames and the suppression of environmental noise; second, how to eliminate the interference of surrounding background objects for jet region identification; and third, how to design reasonable visual indicators for the instantaneous distribution state and cumulative distribution of the reverse-mapping flow.

[0020] To address the aforementioned technical problems, this application provides a machine vision-based multi-nozzle jet dynamic monitoring system, which includes a hardware platform and a vision processing device.

[0021] like Figure 1 As shown, the main body of the hardware platform is constructed from an alloy frame, and the components of the hardware platform include various types of cooling nozzles to be tested (i.e., Figure 1 The cooling device shown in the diagram), the laser (i.e. Figure 1 The system includes a laser source (illustrated in the diagram) and a camera; the nozzles are mounted on a support that can move up, down, left, and right to adjust the jet height and nozzle spacing; the laser is set at a fixed height to emit laser light laterally and irradiate the cross-section of the jet; the camera lens has a fixed viewing angle to capture the laser-irradiated jet cross-section from above and to acquire multi-nozzle laser jet video in real time.

[0022] The height of the laser needs to be precisely matched and set according to the actual jet height to ensure that the laser beam emitted laterally can accurately illuminate the preset monitoring section of the jet. This avoids the laser beam not covering the jet area due to the laser being too high, and also prevents the laser beam from penetrating the space below the jet due to the laser being too low, thus failing to effectively capture the characteristics of the jet section. Ultimately, this achieves precise alignment between the laser and the jet section.

[0023] like Figure 2As shown, the vision processing device includes an image set acquisition module, an image preprocessing module, a jet region recognition module, an instantaneous index measurement module, a cumulative distribution measurement module, and a visualization module.

[0024] The image acquisition module is used to sample the multi-nozzle laser jet video captured by the camera to obtain the first image set; The image preprocessing module is used to perform preprocessing operations on each single frame image in the first image set to obtain the second image set; The jet region recognition module uses threshold segmentation and morphological filtering techniques to perform jet region recognition on each single frame image in the second image set to obtain the third image set. The instantaneous index measurement module is used to perform jet instantaneous state distribution detection on each single frame image in the third image set to obtain the fourth image set; The cumulative distribution module is used to perform jet cumulative flow distribution estimation on the fourth image set; The visualization module is used to visualize the instantaneous state distribution and cumulative flow distribution of the jet.

[0025] This embodiment combines a hardware platform with a vision processing device to achieve coverage of multi-nozzle jets from physical process capture to data quantification analysis, ultimately accurately acquiring jet visual data that can be used for quantification analysis.

[0026] This application embodiment provides a possible implementation method. The sampling operation implemented by the image preprocessing module above may specifically include the following steps: Frames of the multi-nozzle laser jet video captured by the camera are selected at fixed intervals to obtain the first image set. , Satisfying equation (1): (1) In equation (1), For calculating time After that A set of images sampled at intervals. Controls the number of elements in the selected image set.

[0027] This embodiment uses a fixed time interval as the sampling granularity to obtain jet dynamic data that is adapted for subsequent algorithm processing.

[0028] This application embodiment provides a possible implementation method. The preprocessing operation implemented by the image preprocessing module above may specifically include the following steps: Select any one image from the first image set as the first target image. For the first target image Perform guided filtering operation: (2) In equation (2), This is the preprocessed image output after filtering. Let be any pixel coordinate value of the first target image. It is a local mean. The guiding image is used to direct the filtering direction. For local covariance, For local variance, It is a constant used to prevent numerical overflow; The local covariance calculation satisfies equation (3): (3) In equation (3), Therefore The neighborhood centered arbitrary pixel coordinates, For the neighborhood The number of pixels within the cell; The local variance calculation satisfies equation (4): (4) Substitute the calculation results of equations (3) and (4) into equation (2) to implement the guided filtering operation.

[0029] This embodiment can transform the original jet image containing noise interference into a clear and edge-preserving preprocessed image, with pixel-level precision as the processing granularity, providing high-quality image data for subsequent jet region identification.

[0030] This application embodiment provides a possible implementation method. The jet region identification operation implemented by the above jet region identification module may specifically include the following steps: Select any image from the second image set as the second target image. The green laser jet region in the second target image is identified using the following calculation formula: (5) In equation (5), To complete the third target image after jet region identification, To segment the binary result, the segmented binary result satisfies equation (6); (6) In equation (6), For pixels The value of the red channel, For pixels The value of the green channel, The critical value used to distinguish the jet from the background.

[0031] This embodiment can transform a preprocessed jet image containing background interference into a clean jet image that retains only the target area, providing accurate target data for subsequent instantaneous index measurements.

[0032] This application embodiment provides a possible implementation method. The instantaneous state distribution detection operation of the jet implemented by the instantaneous index measurement module above may specifically include the following steps: Select any image from the third image set as the third target image. Using a linear model for the third target image The jet density distribution was simulated, and a linear model was established that satisfies equation (7): (7) In equation (7), This represents the amount of water passing through the nozzle per unit time. Let be the cross-sectional area of ​​the nozzle. The average pixel value of the region. and Let be the linear parameters to be fitted. and Calculate to satisfy equation (8): (8) Calculated according to equation (8) and Then, calculate the pixel reference density of the instantaneous jet region: (9) In equation (9), This is the pixel reference density for the final calculated transient jet region.

[0033] This embodiment can convert the visual features of the jet region into quantifiable physical density indicators, thereby enabling the acquisition of instantaneous state distribution data of multi-nozzle jets.

[0034] This application embodiment provides a possible implementation method. The jet cumulative flow distribution estimation operation implemented by the above cumulative distribution module specifically includes the following steps: The pixel reference density of the transient jet region in consecutive frames of the fourth image set is cumulatively calculated along the bandwidth direction. First, the cumulative reference value of pixel flux is calculated: (10) In equation (10), Accumulate a reference value for the flow rate for each pixel. For the fourth image set, the first Pixel reference density of the transient jet region of the frame. The reference velocity for simulating the movement of the steel plate on the image; Pixel flow cumulative reference value Perform local window smoothing: (11) In equation (11), For Centered on, To calculate the side length of a square window, For The pixel coordinates within the center window; Calculate the cumulative flow distribution of the jet: (12) In equation (12), It is cumulatively distributed along the width direction of the steel plate; Reference velocity v and side length The calculation satisfies equation (13): (13) In equation (13), The actual speed of the steel plate as measured by physical measurement. For conversion ratios, The side length of a single test tube in a water-filled container, measured physically.

[0035] This embodiment can transform discrete multi-frame jet instantaneous density data into cumulative flow distribution results that reflect actual industrial scenarios.

[0036] This application embodiment provides a possible implementation method. The visualization operation implemented by the visualization module above specifically includes the following steps: Instantaneous state distribution visualization: A heatmap is used to visualize the pixel reference density of the transient jet region in each frame of the fourth image set. Cumulative flow distribution visualization: The cumulative flow distribution of the jet is visualized using a bar chart.

[0037] The above describes various implementation methods of the machine vision-based multi-nozzle jet dynamic monitoring system provided in the embodiments of this application. The following will further illustrate the machine vision-based multi-nozzle jet dynamic monitoring system in the embodiments of this application through specific examples.

[0038] This specific embodiment applies machine vision technology to the dynamic monitoring of nozzle jets, which can overcome the limitations of traditional detection methods and achieve comprehensive measurement of the instantaneous state and cumulative distribution of the jet, thereby providing a scientific basis for nozzle design optimization and cooling process control. Figure 6 The implementation process of this specific embodiment is as follows: Step 1: Hardware Platform Setup. The main components include various types of cooling nozzles, a camera, and a laser device. The nozzles are mounted on a flexible, adjustable support frame, allowing for vertical and horizontal movement to adjust key parameters such as the jet height and the distance between nozzles. The laser emits horizontally at a fixed height, illuminating a specific cross-section of the jet. Simultaneously, the camera maintains a fixed viewing angle, capturing the laser jet cross-section from above and recording video data in real time. Furthermore, the camera is powered by a PoE industrial switch and simultaneously transmits the video data stream to a computer for software functionality.

[0039] Step 2: Sample multi-nozzle laser jet video frames at fixed intervals. Save and process keyframes of the continuously acquired video stream using the following formula. Selected in The time is 1 second, and each collection is approximately 1 second. The images are analyzed. This sampling method ensures that the periodic dynamic changes of the jet in each video segment are captured, guaranteeing the richness and accuracy of the data.

[0040] Step 3: Perform image preprocessing. When calculating the guided filter, the guided image can be selected from the original image. To avoid numerical issues, choose After preprocessing, transmission noise and highlights in the video frames are suppressed while preserving subtle water droplet texture details. The images before and after preprocessing are as follows: Figure 3a and Figure 3b As shown, Figure 3a The image before preprocessing. Figure 3b This is the preprocessed image.

[0041] Step 4: Identify the jet region. This involves a combination of image thresholding and morphological processing, fully utilizing prior knowledge from industrial vision, where thresholding... The reference range is 0.6-0.8, and the structural elements are treated morphologically. and Typically, circular elements are selected, and the core size reference value for controlling the size is 5.

[0042] Step 5: Calculate the instantaneous density distribution. Data acquisition and experiments are performed on a single nozzle, and density reference values ​​are simulated and calculated. The parameters are calculated using the least squares algorithm. and Then, the spatial distribution of image pixels can be transformed into the density spatial distribution. .

[0043] Step 6: Calculate the cumulative distribution based on the reference density calculated in Step 5. Simulate the movement of the steel plate by installing the water-holding device on the steel strip conveyor and using a motion servo controller to control the transmission speed, while simultaneously storing the experimental video data. Once the water-holding device is completely removed from the jet region, the actual cumulative water distribution is obtained. After analysis using software, the cumulative algorithm measurement results are obtained.

[0044] Step 7: Visualize the instantaneous jet state distribution and cumulative jet flow distribution data obtained in the preceding steps to obtain, as shown below. Figure 4a , Figure 4b , Figure 4c and Figure 4d The instantaneous state distribution visualization diagram shown and as follows Figure 5a and Figure 5b The diagram shows a visualization of the cumulative traffic distribution.

[0045] The vision processing device in this embodiment is developed using the Python language and packaged as an .exe executable file for rapid deployment on-site. During development, all data measurement, visualization, and storage functions feature user-friendly visual interfaces and human-computer interaction capabilities for ease of operation. Simultaneously, it provides formulas for cumulative distribution. actual speed and the side length of the test tube After multiple experimental tests, the average Pearson correlation coefficient between the designed cumulative distribution index and the actual distribution is greater than 90%, which meets the needs of actual industrial testing.

[0046] It should be noted that the sequence numbers of the steps in the above embodiments do not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. In practical applications, all the above possible implementation methods can be arbitrarily combined in a combined manner to form possible embodiments of this application, which will not be described in detail here.

[0047] Based on the machine vision-based multi-nozzle jet dynamic monitoring system provided above, and based on the same inventive concept, this application also provides a machine vision-based multi-nozzle jet dynamic monitoring method, which includes: A hardware platform is pre-built, the main body of which is constructed from an alloy frame. The components of the hardware platform include nozzles of various types to be tested for cooling, lasers, and cameras. The nozzles are mounted on supports that can move up, down, left, and right to adjust the jet height and nozzle spacing. The laser is set at a fixed height to emit laser light laterally and irradiate the cross-section of the jet. The camera lens has a fixed viewing angle to capture the cross-section of the laser-irradiated jet from above and to acquire real-time video of the multi-nozzle laser jet. The first image set is obtained by sampling the multi-nozzle laser jet video captured by the camera; The second image set is obtained by performing preprocessing operations on each single frame image in the first image set; The third image set is obtained by performing jet region identification on each single frame image in the second image set using threshold segmentation and morphological filtering techniques. A fourth image set is obtained by performing jet instantaneous state distribution detection on each single frame image in the third image set. Perform jet cumulative flux distribution estimation on the fourth image set; Visualize the instantaneous state distribution and cumulative flow distribution of the jet.

[0048] Based on the same inventive concept, embodiments of this application also provide an electronic device, including a processor and a memory, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the above-described machine vision-based multi-nozzle jet dynamic monitoring method.

[0049] In an exemplary embodiment, an electronic device is provided, comprising a processor and a memory. The processor and memory are connected, for example, via a bus. Optionally, the electronic device may further include a transceiver. It should be noted that in practical applications, the transceiver is not limited to one unit, and the structure of this electronic device does not constitute a limitation on the embodiments of this application.

[0050] The processor can be a CPU (Central Processing Unit), GPU (Graphics Processing Unit), DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), FPGA, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0051] A bus can include a pathway for transmitting information between the aforementioned components. The bus can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc.

[0052] The memory may be ROM (Read Only Memory) or other types of static storage devices capable of storing static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices capable of storing information and instructions, or EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited to these.

[0053] The memory stores computer program code that executes the scheme of this application, and its execution is controlled by a processor. The processor executes the computer program code stored in the memory to implement the content shown in the foregoing method embodiments.

[0054] Among them, electronic devices include, but are not limited to: mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and in-vehicle terminals (such as in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers.

[0055] Based on the same inventive concept, this application also provides a storage medium storing a computer program, wherein the computer program is configured to execute the above-described machine vision-based multi-nozzle jet dynamic monitoring method at runtime.

[0056] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that within the spirit and principles of this application, modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein; and these modifications or substitutions do not cause the corresponding technical solutions to leave the protection scope of this application.

Claims

1. A multi-nozzle jet dynamic monitoring system based on machine vision, characterized in that, This includes the hardware platform and the vision processing device; The main body of the hardware platform is constructed from an alloy frame. The components of the hardware platform include nozzles of various types to be cooled, lasers, and cameras. The nozzles are mounted on a support that can move up, down, left, and right to adjust the jet height and nozzle spacing. The laser is set at a fixed height and is used to emit laser light laterally and irradiate the cross-section of the jet. The camera lens has a fixed viewing angle and is used to shoot the cross-section of the laser-irradiated jet from above and to acquire multi-nozzle laser jet video in real time. The visual processing device includes an image acquisition module, an image preprocessing module, a jet region recognition module, an instantaneous index measurement module, a cumulative distribution measurement module, and a visualization module; The image acquisition module is used to sample the multi-nozzle laser jet video captured by the camera to obtain the first image set; The image preprocessing module is used to perform preprocessing operations on each single frame image in the first image set to obtain the second image set; The jet region recognition module uses threshold segmentation and morphological filtering techniques to perform jet region recognition on each single frame image in the second image set to obtain the third image set. The instantaneous index measurement module is used to perform jet instantaneous state distribution detection on each single frame image in the third image set to obtain the fourth image set; The cumulative distribution module is used to perform jet cumulative flow distribution estimation on the fourth image set; The visualization module is used to visualize the instantaneous state distribution and cumulative flow distribution of the jet.

2. The system according to claim 1, characterized in that, Sampling operations include: Frames of the multi-nozzle laser jet video captured by the camera are selected at fixed intervals to obtain the first image set. , Satisfying equation (1): (1) In equation (1), For calculating time After that A set of images sampled at intervals. Controls the number of elements in the selected image set.

3. The system according to claim 1, characterized in that, Preprocessing operations include: Select any one image from the first image set as the first target image. For the first target image Perform guided filtering operation: (2) In equation (2), This is the preprocessed image output after filtering. Let be any pixel coordinate value of the first target image. It is a local mean. The guiding image is used to direct the filtering direction. For local covariance, For local variance, It is a constant used to prevent numerical overflow; The local covariance calculation satisfies equation (3): (3) In equation (3), Therefore The neighborhood centered arbitrary pixel coordinates, For the neighborhood The number of pixels within the cell; The local variance calculation satisfies equation (4): (4) Substitute the calculation results of equations (3) and (4) into equation (2) to implement the guided filtering operation.

4. The system according to claim 3, characterized in that, The jet region identification operation includes: Select any image from the second image set as the second target image. The green laser jet region in the second target image is identified using the following calculation formula: (5) In equation (5), To complete the third target image after jet region identification, To segment the binary result, the segmented binary result satisfies equation (6); (6) In equation (6), For pixels The value of the red channel, For pixels The value of the green channel, The critical value used to distinguish the jet from the background.

5. The system according to claim 4, characterized in that, The instantaneous state distribution detection operation of the jet includes: Select any image from the third image set as the third target image. Using a linear model for the third target image The jet density distribution was simulated, and a linear model was established that satisfies equation (7): (7) In equation (7), This represents the amount of water passing through the nozzle per unit time. Let be the cross-sectional area of ​​the nozzle. The average pixel value of the region. and Let be the linear parameters to be fitted. and Calculate to satisfy equation (8): (8) Calculated according to equation (8) and Then, calculate the pixel reference density of the instantaneous jet region: (9) In equation (9), This is the pixel reference density for the final calculated transient jet region.

6. The system according to claim 5, characterized in that, The jet cumulative flow distribution estimation operation includes: The pixel reference density of the transient jet region in consecutive frames of the fourth image set is cumulatively calculated along the bandwidth direction. First, the cumulative reference value of pixel flux is calculated: (10) In equation (10), Accumulate a reference value for the flow rate for each pixel. For the fourth image set, the first Pixel reference density of the transient jet region of the frame. The reference velocity for simulating the movement of the steel plate on the image; Pixel flow cumulative reference value Perform local window smoothing: (11) In equation (11), For Centered on, To calculate the side length of a square window, For The pixel coordinates within the center window; Calculate the cumulative flow distribution of the jet: (12) In equation (12), It is cumulatively distributed along the width direction of the steel plate; Reference velocity v and side length The calculation satisfies equation (13): (13) In equation (13), The actual speed of the steel plate as measured by physical measurement. For conversion ratios, The side length of a single test tube in a water-filled container, measured physically.

7. The system according to any one of claims 1 to 6, characterized in that, Visual operations include: Instantaneous state distribution visualization: A heatmap is used to visualize the pixel reference density of the transient jet region in each frame of the fourth image set. Cumulative flow distribution visualization: The cumulative flow distribution of the jet is visualized using a bar chart.

8. A method for dynamic monitoring of multi-nozzle jets based on machine vision, characterized in that, The method includes: A hardware platform is pre-built, the main body of which is constructed from an alloy frame. The components of the hardware platform include nozzles of various types to be tested for cooling, lasers, and cameras. The nozzles are mounted on supports that can move up, down, left, and right to adjust the jet height and nozzle spacing. The laser is set at a fixed height to emit laser light laterally and irradiate the cross-section of the jet. The camera lens has a fixed viewing angle to capture the cross-section of the laser-irradiated jet from above and to acquire real-time video of the multi-nozzle laser jet. The first image set is obtained by sampling the multi-nozzle laser jet video captured by the camera; The second image set is obtained by performing preprocessing operations on each single frame image in the first image set; The third image set is obtained by performing jet region identification on each single frame image in the second image set using threshold segmentation and morphological filtering techniques. A fourth image set is obtained by performing jet instantaneous state distribution detection on each single frame image in the third image set. Perform jet cumulative flux distribution estimation on the fourth image set; Visualize the instantaneous state distribution and cumulative flow distribution of the jet.

9. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the machine vision-based multi-nozzle jet dynamic monitoring method of claim 8.

10. A storage medium, characterized in that, The storage medium stores a computer program, wherein the computer program is configured to execute the machine vision-based multi-nozzle jet dynamic monitoring method of claim 8 at runtime.

Images