A method and system for measuring fuel nozzle spray angle based on vision recognition

Abstract

This application relates to a method and system for measuring the spray angle of a fuel nozzle based on visual recognition, belonging to the field of nozzle testing technology. The method includes acquiring a spray image of the nozzle to be measured; acquiring the nozzle coordinates in the spray image; filtering out a region to be detected from the spray image; performing edge detection on the region to be detected in the spray image to obtain the edge region of the spray from the nozzle to be measured; and obtaining the angle information of the nozzle spray based on the nozzle coordinates and the edge region. By extracting the edges of the spray area generated by the nozzle to be measured in the spray image, the edge region is obtained. Since the spray from a nozzle is usually a cone-shaped region centered on the nozzle coordinates, the trend of the edge region is combined with the nozzle coordinate edges to determine the cone-shaped region, and the angle information is obtained through visual recognition. This application improves the accuracy of obtaining the angle information of the nozzle spray.

Description

Technical Field

[0001] This invention relates to the field of nozzle testing technology, and in particular to a method and system for measuring the spray angle of a fuel nozzle based on visual recognition. Background Technology

[0002] Nozzles are a component of aircraft jet engines, used to expel high-speed airflow to generate thrust. Adjustable nozzles can change the direction of the airflow to control the aircraft's attitude, speed, and direction, and are also crucial for maintaining the aircraft's balance and stability.

[0003] Nozzle spray angle testing is a crucial test item in nozzle testing. The accuracy of the nozzle spray angle is essential for engine performance, combustion efficiency, and emission control. Incorrect spray angles can lead to uneven combustion, high-temperature zones, pollutant generation, and reduced engine life. Currently, the method for measuring the angle of aircraft engine fuel nozzles typically involves moving a test probe to the spray point and measuring the distance between the probes to calculate the spray angle. However, errors can occur in the movement position of the test probe and the measurement of the distance, resulting in inaccurate nozzle angle measurements. Improving the accuracy of nozzle spray angle measurement is a problem that urgently needs to be solved. Summary of the Invention

[0004] To improve the accuracy of nozzle angle measurement, this application provides a method and system for measuring fuel nozzle spray angle based on vision recognition.

[0005] Firstly, this application provides a method for measuring the spray angle of a fuel nozzle based on visual recognition, which adopts the following technical solution:

[0006] A method for measuring fuel nozzle spray angle based on vision recognition, comprising:

[0007] Acquire a spray image from the nozzle to be measured;

[0008] Obtain the nozzle coordinates in the spray image;

[0009] Select the region to be detected from the spray image;

[0010] Edge detection is performed on the region to be detected in the spray image to obtain the edge region of the spray from the nozzle to be measured.

[0011] Based on the nozzle coordinates and the edge region, the nozzle spray angle information is obtained.

[0012] By adopting the above technical solution, edge detection is performed on the area to be detected in the spray image. By extracting the edge of the spray area generated by the nozzle to be measured in the spray image, the edge region is obtained. Since the spray from the nozzle is usually a cone-shaped area centered on the nozzle coordinates, the trend of the edge region is combined with the nozzle coordinate edge to determine the cone region. The visual recognition method helps to improve the accuracy of obtaining the angle information of the nozzle spray.

[0013] Optionally, the step of performing edge detection on the region to be detected in the spray image to obtain the edge region of the spray from the nozzle to be measured specifically includes:

[0014] Gradient calculation is performed on the spray image to obtain the gradient magnitude of each pixel in the spray image;

[0015] Based on the gradient magnitude, search for local maxima in the spray image to obtain the maximum pixel grid.

[0016] The spray image is binarized and denoised based on the maximum pixel grid to obtain a binarized image;

[0017] The binarized image is processed to obtain the edge region of the spray from the nozzle to be measured.

[0018] By adopting the above technical solution, the gradient magnitude of each pixel in the spray image is calculated to obtain the degree of gray level change of each pixel in the spray image. The maximum pixel at the edge is selected by searching for local maxima, and the maximum pixel is binarized and denoised to obtain a binarized image that is easy to distinguish. Then, the binarized image is connected to smooth the edge of the binarized image, thereby realizing the identification of the edge region of the spray from the nozzle to be measured.

[0019] Optionally, obtaining the nozzle spray angle information based on the nozzle coordinates and the edge region specifically includes:

[0020] Based on the nozzle coordinates, the standard regression equation is modified to obtain an optimized regression equation;

[0021] Treat the center of each pixel grid in the edge region as a valid point;

[0022] Based on the optimized regression equation, linear regression is performed on all valid points in the same edge region to obtain the regression line of the corresponding edge region.

[0023] The angle information of the nozzle spray is obtained based on the slope of the regression line.

[0024] By adopting the above technical solution, since the spray generated by the nozzle starts from the nozzle coordinates, the edge region will definitely pass through the nozzle coordinates. The standard regression equation is improved so that the improved optimized regression equation passes through the nozzle coordinates. The center of each pixel in the edge region is taken as the effective point, and a linear regression operation is performed on all effective points based on the optimized regression equation. Since the obtained regression line must pass through the nozzle coordinates, the angle information of the nozzle spray can be obtained by the slope of the regression line.

[0025] Optionally, it also includes a verification step for angle information, wherein the verification step specifically includes:

[0026] Obtain the standard estimation error value for each of the regression lines;

[0027] The accuracy of the angle information is estimated based on the standard error value.

[0028] By adopting the above technical solution, the standard estimation error value is used to obtain the degree of fit between each effective point and the regression line. The greater the degree of fit, the higher the accuracy of the effective point, which makes it easier to analyze the accuracy of the final angle information based on the standard estimation error value.

[0029] Optionally, the step of filtering the area to be detected from the spray image specifically includes:

[0030] Obtain the standard spray angle of the nozzle to be tested;

[0031] The area to be detected in the spray image is obtained based on the nozzle coordinates and the standard spray angle.

[0032] By adopting the above technical solution, the area to be detected based on the standard nozzle angle and nozzle coordinates is the possible range of spray produced by the nozzle to be detected. Only the area to be detected is detected, which greatly reduces the amount of data processing.

[0033] Optionally, the step of searching for local maxima in the spray image based on the gradient magnitude to obtain the maxima pixel grid specifically includes:

[0034] Pre-set gradient magnitude threshold;

[0035] Based on the gradient magnitude, local maxima in the spray image are searched, and the local maxima are filtered based on the gradient magnitude threshold to obtain the maximum pixel grid.

[0036] By adopting the above technical solution, the gradient amplitude threshold is set to filter out pixels with large grayscale changes represented by the gradient amplitude as maxima pixels, thereby reducing the possibility of misjudging maxima pixels.

[0037] Optionally, after searching for local maxima in the spray image based on the gradient magnitude to obtain the maxima pixel grid, the method further includes:

[0038] If the number of maximum value pixels is less than the preset number, the edge region recognition is deemed to have failed, and the angle measurement of the corresponding nozzle to be measured is terminated.

[0039] By adopting the above technical solution, if the number of maximum value pixels is less than the preset number, it means that not enough maximum value pixels of the process spray edge have been detected in the area to be detected, that is, the edge area recognition has failed, and the angle of the nozzle to be measured cannot be tested. At this time, the angle test of the nozzle to be measured is immediately terminated and no subsequent operation is performed to save program running time.

[0040] Secondly, this application provides a fuel nozzle spray angle measurement system based on vision recognition, employing the following technical solution:

[0041] A fuel nozzle spray angle measurement system based on vision recognition, comprising:

[0042] The image acquisition module is used to acquire spray images from the nozzle to be measured.

[0043] The nozzle position acquisition module is used to acquire the nozzle coordinates in the spray image;

[0044] The region acquisition module is used to filter out the regions to be detected from the spray image;

[0045] The edge region recognition module is used to perform edge detection on the region to be detected in the spray image to obtain the edge region of the spray from the nozzle to be measured.

[0046] The angle information recognition module is used to obtain the angle information of the nozzle spray based on the nozzle coordinates and the edge area.

[0047] Thirdly, this application provides a computer device that adopts the following technical solution:

[0048] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the computer program as described in any of the above methods.

[0049] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution:

[0050] A computer-readable storage medium includes a computer program stored thereon that can be loaded by a processor and executed as in any of the methods described above. Attached Figure Description

[0051] Figure 1This is a schematic diagram of a nozzle angle measuring device according to one embodiment of this application.

[0052] Figure 2 This is a flowchart of a nozzle spray angle measurement method according to one embodiment of this application.

[0053] Figure 3 This is a flowchart of a method for detecting edge regions according to one embodiment of this application.

[0054] Figure 4 This is a flowchart of a method for measuring angle information according to one embodiment of this application.

[0055] Figure 5 This is a block diagram of a nozzle spray angle measurement system according to one embodiment of this application.

[0056] Explanation of reference numerals in the attached diagram: 1. Test chamber; 2. Nozzle mounting base; 3. Camera. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0058] This application discloses a method for measuring the spray angle of a fuel nozzle based on visual recognition. (Refer to...) Figure 1 , 2 A method for measuring the spray angle of a fuel nozzle based on visual recognition is applied to a nozzle angle measuring device. The nozzle angle measuring device includes a test chamber, a camera, and a nozzle mounting base. The camera is located on one side of the test chamber. The method includes:

[0059] Step S101: Obtain a spray image of the nozzle to be measured.

[0060] The spray images can be captured by image acquisition devices such as cameras. Furthermore, because the spray is cone-shaped, the results obtained from different directions are consistent. Therefore, cameras can be placed on any side of the test chamber, or multiple cameras can be placed in different directions within the test chamber to acquire spray images from multiple angles.

[0061] Step S102: Obtain the nozzle coordinates in the spray image.

[0062] It should be understood that since the nozzle's installation position is fixed, the nozzle coordinates should also be consistent. For example, the nozzle coordinates can be used as the origin of the coordinate system.

[0063] Step S103: Select the area to be detected from the spray image.

[0064] It should be understood that the main purpose of measuring the nozzle spray angle is to conduct quality inspection of the nozzle and determine whether the nozzle spray angle is qualified. Therefore, the spray range of each nozzle is fixed to screen out the area to be tested, which helps to reduce the amount of data processing.

[0065] Step S104: Perform edge detection on the area to be detected in the spray image to obtain the edge area of ​​the spray from the nozzle to be measured;

[0066] Step S105: Obtain the nozzle spray angle information based on the nozzle coordinates and the edge area.

[0067] In the above embodiments, edge detection is performed on the area to be detected in the spray image. By extracting the edge of the spray area generated by the nozzle to be measured in the spray image, the edge area is obtained. Since the spray of the nozzle is usually a cone-shaped area centered on the nozzle coordinates, the trend of the edge area is combined with the nozzle coordinate edge to determine the cone-shaped area, so as to obtain the spray angle information.

[0068] As one implementation of step S103, step S103 specifically includes:

[0069] Obtain the standard spray angle of the nozzle to be tested; based on the nozzle coordinates and the standard spray angle, obtain the area to be tested in the spray image.

[0070] As one possible approach, a polar coordinate system can be established with the nozzle coordinates as the origin. If the standard spray angle is 90 degrees, the area to be detected in the spray image can be the region from 4 / π to -4 / π in the polar coordinate system.

[0071] In the above embodiments, the area to be detected, obtained based on the standard nozzle angle and nozzle coordinates, is the possible range of spray produced by the nozzle to be detected. Only the area to be detected is detected, which greatly reduces the amount of data processing.

[0072] As one implementation of step S104, step S104 specifically includes:

[0073] Step S1041: Perform gradient calculation on the spray image to obtain the gradient magnitude of each pixel in the spray image.

[0074] Specifically, the gradient calculation method can be to process the spray image using the Sobel operator to calculate the gradient magnitude of each pixel in the image. The Sobel operator uses convolution kernels in the horizontal and vertical directions to measure the change in gray value of the pixel grid, and the gradient magnitude represents the rate of change of gray value.

[0075] Furthermore, the gradient magnitude can be calculated using the following formula: Gradient magnitude = sqrt((horizontal gradient)) 2+(gradient in the vertical direction) 2 ).

[0076] Step S1042: Based on the gradient magnitude, search for local maxima in the spray image to obtain the maximum pixel grid;

[0077] Specifically, local maximum detection is based on the neighborhood of a pixel. If the gradient magnitude of a pixel exceeds the gradient magnitudes of all pixels in its neighborhood, it is considered a local maximum. Furthermore, different window sizes and shapes can be set when searching for local maxima. A larger window results in thicker detected edges, while a smaller window results in thinner detected edges. These settings can be adjusted according to the specific needs of the user.

[0078] Step S1043: Perform binarization and denoising processing on the spray image based on the maximum value pixel grid to obtain a binarized image;

[0079] The denoising process can be dilation and erosion in morphological operations. Erosion can eliminate small white noise, while dilation can fill small black holes, which helps to extract isolated maxima pixels and makes connected maxima pixels smoother.

[0080] Step S1044: Perform connectivity processing on the binarized image to obtain the edge region of the spray from the nozzle to be measured.

[0081] Connectivity processing refers to extracting connected pixels from a binarized image, analyzing these connected pixels, and filtering out connected pixels with too small an area to remove small noise regions, thereby removing noise regions and retaining edge regions with larger connected areas.

[0082] In the above embodiments, the gradient magnitude of each pixel in the spray image is calculated to obtain the degree of gray level change of each pixel in the spray image. The maximum pixel at the edge is selected by searching for local maxima, and the maximum pixel is binarized and denoised to obtain a binarized image that is easy to distinguish. Then, the binarized image is connected to smooth the edge of the binarized image, thereby realizing the identification of the edge region of the spray from the nozzle to be measured.

[0083] As one implementation of step S1042, step S1042 specifically includes:

[0084] A gradient magnitude threshold is preset; based on the gradient magnitude, local maxima in the spray image are searched, and the local maxima are filtered based on the gradient magnitude threshold to obtain the maximum value pixel grid.

[0085] In the above embodiments, by setting a gradient amplitude threshold, pixels with large grayscale changes represented by the gradient amplitude are selected as maxima pixels, thereby reducing the possibility of misjudging maxima pixels.

[0086] As one implementation of step S105, step S105 specifically includes:

[0087] Step S1051: Based on the nozzle coordinates, modify the standard regression equation to obtain the optimized regression equation.

[0088] Specifically, the standard regression equation can be expressed as: Y = aX + b, where Y is the dependent variable, X is the independent variable, a is the regression coefficient (slope), and b is the intercept. By using the nozzle coordinates as the forced passpoint (X0, Y0) of the standard regression equation, and then substituting the forced passpoint (X0, Y0) into the standard regression equation, the values ​​or relationship between a and b are calculated. Substituting the values ​​or relationship between a and b back into the standard regression equation yields the optimized regression equation.

[0089] As an example, if we set the nozzle coordinates to the origin (0,0) and substitute them into the standard regression equation, we get b=0. At this point, the optimized regression equation is Y=aX.

[0090] Step S1052: Take the center of each pixel grid in the edge region as a valid point;

[0091] Step S1053: Based on the optimized regression equation, perform linear regression on all valid points in the same edge region to obtain the regression line of the corresponding edge region.

[0092] The vertical line where the ordinate of the nozzle coordinate is located is the dividing line. The same edge region refers to the region located on the same side of the dividing line. That is, the two sides of the dividing line are two edge regions. When performing linear regression calculation, the effective points of the same edge region are calculated separately.

[0093] It should be understood that performing linear regression on the effective points is equivalent to performing linear fitting on all effective points. The regression line obtained by the linear regression operation is the regression line that best reflects the distribution trend of all effective points, and the regression line obtained based on the optimized regression equation must pass through the nozzle coordinates.

[0094] Step S1054: Obtain the nozzle spray angle information based on the slope of the regression line.

[0095] In the above implementation, since the spray generated by the nozzle starts from the nozzle coordinates, the edge region will definitely pass through the nozzle coordinates. The standard regression equation is improved so that the improved optimized regression equation passes through the nozzle coordinates. The center of each pixel in the edge region is taken as the effective point, and a linear regression operation is performed on all effective points based on the optimized regression equation. Since the obtained regression line must pass through the nozzle coordinates, the angle information of the nozzle spray can be obtained by the slope of the regression line.

[0096] As a further embodiment of the nozzle spray angle measurement method, the nozzle spray angle measurement method also includes a verification step for the angle information, the verification step specifically including:

[0097] Obtain the standard estimation error value for each regression line; based on the standard estimation error value, estimate the accuracy of the angle information.

[0098] It should be understood that the standard estimation error is an indicator of the relative deviation between the actual value and its estimated value. In this embodiment, the standard estimation error is used to measure the representativeness of the regression line to all valid points in the marginal region. The smaller the standard estimation error, the better the regression line fits, that is, the more it can represent all valid points in the marginal region, and the higher the accuracy of the regression line.

[0099] Furthermore, if the standard estimation error value is greater than the preset value, it indicates that the edge area of ​​the nozzle is not a regular straight line. Even if the nozzle angle information is qualified, the nozzle spray shape is not a regular cone. At this time, it is determined that there is an uneven spray in the corresponding nozzle, so that the staff can make adjustments to the nozzle based on the prompt.

[0100] In the above implementation method, the standard estimation error value is used to obtain the degree of fit between each effective point and the regression line. The greater the degree of fit, the higher the accuracy of the effective point, which makes it easier to analyze the accuracy of the final angle information based on the standard estimation error value.

[0101] As a further embodiment of the nozzle spray angle measurement method, after step S1042, the method further includes:

[0102] If the number of maximum value pixels is less than the preset number, the edge region recognition is deemed to have failed, and the angle measurement of the corresponding nozzle to be measured is terminated.

[0103] In the above implementation, if the number of maximum value pixels is less than the preset number, it means that not enough maximum value pixels of the process spray edge have been detected in the area to be detected, that is, the edge area recognition has failed, and the angle of the nozzle to be measured cannot be tested. At this time, the angle test of the nozzle to be measured is immediately terminated and no subsequent operation is performed to save program running time.

[0104] Furthermore, this application discloses a nozzle spray angle measurement system. The nozzle spray angle measurement system can be applied to a computer device, and is a schematic diagram of the architecture of a computer device for implementing the above-described method provided in this embodiment of the invention. In this embodiment, the computer device may include the nozzle spray angle measurement system, a machine-readable storage medium, and a processor.

[0105] In this embodiment, the machine-readable storage medium and the processor can be located in a computer device and are separately configured. Alternatively, the machine-readable storage medium can be independent of the computer device and accessed by the processor. The nozzle spray angle measurement system may include multiple functional modules stored in the machine-readable storage medium, such as the various software functional modules included in the nozzle spray angle measurement system. When the processor executes the computer program corresponding to the software functional module in the nozzle spray angle measurement system, the nozzle spray angle measurement system provided in the aforementioned method embodiment is implemented.

[0106] In this embodiment, the computer device may include one or more processors. The processor can process information and / or data related to a service request to perform one or more functions described in this invention. In some embodiments, the processor may include one or more processing engines (e.g., a single-core processor or a multi-core processor). As merely an example, the processor may include one or more hardware processors, such as a central processing unit (CPU), an application-specific integrated circuit (ASIC), a special-purpose instruction set processor (ASIP), a graphics processing unit (GPU), a physical arithmetic unit (PPU), a digital signal processor (DSP), a field-programmable gate array (FPGA), a programmable logic device (PLD), a controller, a microcontroller unit, a reduced instruction set computer (RISC), a microprocessor, or any combination thereof.

[0107] Machine-readable storage media can store data and / or instructions. In some embodiments, machine-readable storage media can store acquired data or information. In some embodiments, machine-readable storage media can store data and / or instructions for execution or use by the computer device, which can implement the exemplary methods described in this application by executing or using the data and / or instructions. In some embodiments, machine-readable storage media may include mass storage, removable storage, volatile read-write memory, read-only memory (ROM), or similar or any combination of the examples above. Exemplary mass storage may include disks, optical disks, solid-state drives, etc. Exemplary removable storage may include flash drives, floppy disks, optical disks, memory cards, compressed disks, magnetic tapes, etc. Exemplary volatile read-write memory may include random access memory (RAM). Exemplary random access memory may include dynamic RAM, double-rate synchronous dynamic RAM, static RAM, thyristor RAM, and zero-capacitance RAM, etc. Exemplary ROM may include masked ROM, programmable ROM, erasable programmable ROM, electronically erasable programmable ROM, compressed disk ROM, and digital universal disk ROM, etc.

[0108] The nozzle spray angle measurement system included in the computer device may include one or more software functional modules. These software functional modules may be programs or instructions stored in the machine-readable storage medium. When executed by a corresponding processor, these software functional modules are used to implement the aforementioned methods. For example, when executed by the processor of a UAV, they are used to implement the method steps performed by the UAV; or when executed by the computer device, they are used to implement the method steps performed by the computer device.

[0109] For details, please refer to Figure 5 The fuel nozzle spray angle measurement system based on vision recognition provided in this embodiment may include:

[0110] The image acquisition module is used to acquire spray images from the nozzle to be measured.

[0111] The nozzle position acquisition module is used to acquire the nozzle coordinates in the spray image;

[0112] The region acquisition module is used to filter out the regions to be detected from the spray image;

[0113] The edge region recognition module is used to perform edge detection on the region to be detected in the spray image to obtain the edge region of the spray from the nozzle to be measured.

[0114] The angle information recognition module is used to obtain the angle information of the nozzle spray based on the nozzle coordinates and the edge area.

[0115] The fuel nozzle spray angle measurement system based on vision recognition provided in this application can realize the above-mentioned fuel nozzle spray angle measurement method based on vision recognition, and the specific working process of the fuel nozzle spray angle measurement system based on vision recognition can be referred to the corresponding process in the above method embodiment.

[0116] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0117] The present invention also discloses a computer-readable storage medium, comprising storing a computer program that can be loaded by a processor and executed as in any of the methods described above.

[0118] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0119] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0120] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A method for measuring the spray angle of a fuel nozzle based on visual recognition, characterized in that, include: Acquire a spray image from the nozzle to be measured; Obtain the nozzle coordinates in the spray image; Select the region to be detected from the spray image; The step of filtering out the region to be detected from the spray image specifically includes: obtaining the standard spray angle of the nozzle to be detected; and obtaining the region to be detected in the spray image based on the nozzle coordinates and the standard spray angle. Edge detection is performed on the region to be detected in the spray image to obtain the edge region of the spray from the nozzle to be measured. Based on the nozzle coordinates and the edge region, the nozzle spray angle information is obtained to determine whether the nozzle spray angle is qualified; the nozzle spray angle information is represented as a conical region centered on the nozzle coordinates, and the angle of the conical region is determined by the trend of the edge region and combined with the edge of the nozzle coordinates. The step of obtaining the nozzle spray angle information based on the nozzle coordinates and edge region specifically includes: Based on the nozzle coordinates, the standard regression equation is modified to obtain an optimized regression equation; Treat the center of each pixel grid in the edge region as a valid point; Based on the optimized regression equation, linear regression is performed on all valid points in the same edge region to obtain the regression line of the corresponding edge region. The angle information of the nozzle spray is obtained based on the slope of the regression line; It also includes a verification step for angle information, the verification step specifically including: Obtain the standard estimation error value for each regression line, which is used to measure the representativeness of the regression line to all valid points in the marginal region; The accuracy of the angle information is estimated based on the standard estimated error value. Specifically, if the standard estimated error value is greater than the preset value, it means that the edge area of ​​the nozzle is not a regular straight line. When the angle information of the nozzle is qualified, it is determined that there is uneven spraying of the corresponding nozzle.

2. The method according to claim 1, characterized in that, The step of performing edge detection on the region to be detected in the spray image to obtain the edge region of the spray from the nozzle to be measured specifically includes: Gradient calculation is performed on the spray image to obtain the gradient magnitude of each pixel in the spray image; Based on the gradient magnitude, search for local maxima in the spray image to obtain the maximum pixel grid. The spray image is binarized and denoised based on the maximum pixel grid to obtain a binarized image; The binarized image is processed to obtain the edge region of the spray from the nozzle to be measured.

3. The method according to claim 2, characterized in that, The step of searching for local maxima in the spray image based on the gradient magnitude to obtain the maxima pixel grid specifically includes: Pre-set gradient magnitude threshold; Based on the gradient magnitude, local maxima in the spray image are searched, and the local maxima are filtered based on the gradient magnitude threshold to obtain the maximum pixel grid.

4. The method according to claim 2, characterized in that: After searching for local maxima in the spray image based on the gradient magnitude and obtaining the maxima pixel grid, the process further includes: If the number of maximum value pixels is less than the preset number, the edge region recognition is deemed to have failed, and the angle measurement of the corresponding nozzle to be measured is terminated.

5. A fuel nozzle spray angle measurement system based on vision recognition, characterized in that, include: The image acquisition module is used to acquire spray images from the nozzle to be measured. The nozzle position acquisition module is used to acquire the nozzle coordinates in the spray image; The region acquisition module is used to filter out the region to be detected from the spray image. The specific steps of filtering out the region to be detected from the spray image include: obtaining the standard spray angle of the nozzle to be detected; and obtaining the region to be detected from the spray image based on the nozzle coordinates and the standard spray angle. An edge region recognition module is used to perform edge detection on the area to be detected in the spray image to obtain the edge region of the spray from the nozzle to be measured. An angle information recognition module is used to obtain the angle information of the nozzle spray based on the nozzle coordinates and the edge region to determine whether the nozzle spray angle is qualified. The angle information of the nozzle spray is represented as a cone-shaped area centered on the nozzle coordinates. The angle of the cone-shaped area is determined by the trend of the edge region and the nozzle coordinate edge. Specifically, it includes: modifying the standard regression equation based on the nozzle coordinates to obtain an optimized regression equation; taking the center of each pixel grid in the edge region as a valid point; performing linear regression on all valid points in the same edge region based on the optimized regression equation to obtain the regression line of the corresponding edge region; obtaining the angle information of the nozzle spray based on the slope of the regression line; and a verification step for the angle information, which specifically includes: obtaining the standard estimation error value of each regression line, which is used to measure the representativeness of the regression line to all valid points in the edge region; estimating the accuracy of the angle information based on the standard estimation error value; specifically, if the standard estimation error value is greater than a preset value, it indicates that the edge region of the nozzle is not a regular straight line. When the angle information of the nozzle is qualified, it is determined that the corresponding nozzle has uneven spray.

6. A computer device, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the computer program as described in any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, This includes storing a computer program that can be loaded by a processor and executed as described in any of claims 1-4.

Images