Method, device and equipment for detecting spraying quality of oil sprayer and medium

Abstract

The invention relates to the technical field of image processing, in particular to a method, device and equipment for detecting the spraying quality of an oil sprayer and a medium. In the application, the position of a spray root in a target spray image is determined, the distance between each target pixel point in a segmented image and the spray root is calculated according to the position of the spray root, an expansion matrix corresponding to each target pixel point is determined according to the distance, and morphological expansion processing is performed on the segmented image according to the expansion matrix. The self-adaptability of the expansion matrix is improved, so that when morphological expansion processing is carried out on the segmented image according to the expansion matrix, target pixel points far away from the spraying holes can be connected, the spraying integrity is guaranteed, and the spraying efficiency is improved. According to the spray quantity, the spray area of each target area, the spray included angle of each target area and the spray deflection angle of each target area, the spray quality is classified, the spray weight and the spray form are considered, and the spray quality detection accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular, to a method, device, equipment and medium for detecting the spray quality of an injector. Background Art

[0002] The development of industry has triggered environmental and energy crises. As an engine, the diesel engine has made great contributions to the development of industry. However, the emissions of diesel engines will also have a certain impact on the environment. Therefore, it is particularly important to improve the combustion quality of diesel fuel. As an important part of the engine fuel supply system, the spray quality of the injector greatly affects the combustion perfection and emission quality. Generally, the method for detecting the spray quality is to observe multiple injections manually, or use a collecting device such as an oil collecting pan with a certain number of collecting holes to collect the sprayed fuel spray, and obtain the spray pattern distribution by weighing the fuel mass in each hole, and then obtain the spray cone angle according to this distribution characteristic. However, manual detection is prone to misidentification, resulting in low detection accuracy. Therefore, in the process of detecting the spray quality of the injector, how to improve the detection accuracy has become an urgent problem to be solved. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a method, device, equipment and medium for detecting the spray quality of an injector to solve the problem of low detection accuracy in the process of detecting the spray quality of the injector.

[0004] In a first aspect, an embodiment of the present invention provides a method for detecting the spray quality of an injector, and the spray quality detection method includes: Obtain a target spray image for detecting the spray quality, perform segmentation processing on the target spray image to obtain a segmented image; According to the segmented image, determine the position of the spray root in the target spray image, calculate the distance between each target pixel point in the segmented image and the spray root according to the position of the spray root, determine the dilation matrix corresponding to each target pixel point according to the distance, and perform morphological dilation processing on the segmented image according to the dilation matrix to obtain the target area and the number of sprays corresponding to the spray; Calculate the spray area, spray angle and spray deflection angle of each target area; According to the number of sprays, the spray area of each target area, the spray angle of each target area and the spray deflection angle of each target area, determine the spray characteristics of the target spray image, and classify the spray quality according to the spray characteristics to obtain a classification result.

[0005] In a second aspect, an embodiment of the present invention provides a device for detecting the spray quality of an injector, and the spray quality detection device includes: An acquisition module, configured to acquire a target spray image for detecting spray quality, perform segmentation processing on the target spray image, and obtain a segmented image; A processing module, configured to determine the position of the spray root in the target spray image according to the segmented image, calculate the distance between each target pixel point in the segmented image and the spray root according to the position of the spray root, determine the dilation matrix corresponding to each target pixel point according to the distance, and perform morphological dilation processing on the segmented image according to the dilation matrix to obtain the target area and the number of sprays corresponding to the spray; A calculation module, configured to calculate the spray area, spray angle, and spray deflection angle of each target area; A classification module, configured to determine the spray characteristics of the target spray image according to the number of sprays, the spray area of each target area, the spray angle of each target area, and the spray deflection angle of each target area, classify the spray quality according to the spray characteristics, and obtain a classification result.

[0006] In a third aspect, an embodiment of the present invention provides a computer device, which includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the spray quality detection method described in the first aspect is implemented.

[0007] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the spray quality detection method described in the first aspect is implemented.

[0008] The beneficial effects of the present invention compared with the prior art are as follows: In this application, the target spray image is segmented to determine the position of the spray root in the target spray image. According to the position of the spray root, the distance between each target pixel point in the segmented image and the spray root is calculated. According to the distance, the dilation matrix corresponding to each target pixel point is determined. According to the dilation matrix, morphological dilation processing is performed on the segmented image to improve the self-adaptability of the dilation matrix. When morphological dilation processing is performed on the segmented image according to the dilation matrix, target pixel points far from the spray hole can be connected to ensure the integrity of the spray. According to the number of sprays, the spray area of each target area, the spray angle of each target area, and the spray deflection angle of each target area, the spray characteristics of the target spray image are determined. According to the spray characteristics, the spray quality is classified, considering the weight and shape of the spray, and the accuracy of spray quality detection is improved. Description of the Drawings

[0009] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for the description of the embodiments of the present invention. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0010] Figure 1 It is a schematic diagram of the application environment of a method for detecting the spray quality of an injector provided by an embodiment of the present invention; Figure 2 It is a schematic flowchart of a method for detecting the spray quality of an injector provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of a target spray image provided by an embodiment of the present invention; Figure 4 It is a schematic diagram of a denoised spray image provided by an embodiment of the present invention; Figure 5 It is a schematic structural diagram of a device for detecting the spray quality of an injector provided by an embodiment of the present invention; Figure 6 It is a schematic structural diagram of a computer device provided by an embodiment of the present invention. Detailed implementation manners

[0011] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0012] In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present invention. However, those skilled in the art should understand that the present invention can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present invention.

[0013] It should be understood that when used in the specification and claims of the present invention, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0014] It should also be understood that the term "and / or" as used in the specification of the present invention and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0015] As used in the specification of the present invention and the appended claims, the term "if" can be interpreted as "when", "once", "in response to determining", or "in response to detecting" depending on the context. Similarly, the phrase "if determined" or "if [the described condition or event] is detected" can be interpreted as meaning "once determined", "in response to determining", "once [the described condition or event] is detected", or "in response to detecting [the described condition or event]" depending on the context.

[0016] In addition, in the description of the specification of the present invention and the appended claims, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0017] Reference to "one embodiment" or "some embodiments" or the like described in the specification of the present invention means that a specific feature, structure, or characteristic described in connection with that embodiment is included in one or more embodiments of the present invention. Thus, statements such as "in one embodiment", "in some embodiments", "in other some embodiments", "in still other embodiments", etc. that appear in different places in this specification are not necessarily all referring to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in another way. The terms "comprising", "including", "having", and their variants all mean "including but not limited to", unless otherwise specifically emphasized in another way.

[0018] The embodiments of the present invention can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is a theory, method, technology, and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use the knowledge to obtain the best results.

[0019] Artificial intelligence basic technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. Artificial intelligence software technologies mainly include several major directions such as computer vision technology, robotics, biometric technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0020] It should be understood that the magnitudes of the sequence numbers of the steps in the following embodiments do not imply the order of execution. The order of execution of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0021] To illustrate the technical solution of the present invention, the following specific embodiments are used for illustration.

[0022] A method for detecting the spray quality of an injector provided by an embodiment of the present invention can be applied in an application environment such as Figure 1 the following. Figure 1 FIG. is a schematic diagram of the application environment of a method for detecting the spray quality of an injector provided by an embodiment of the present invention; wherein, the client communicates with the server to solve the problem of low detection accuracy during the detection of the spray quality of the injector. The client, also known as the user terminal, refers to a program that provides local services corresponding to the server. The client can be installed on, but not limited to, various personal computers, laptop computers, smart phones, tablet computers, and portable wearable devices. The server can be implemented by an independent server or a server cluster composed of multiple servers.

[0023] Referring to Figure 2 , which is a schematic flowchart of a method for detecting the spray quality of an injector provided by an embodiment of the present invention. Taking the server in Figure 1 as an example, the method includes the following steps: S201: Obtain a target spray image for detecting spray quality, and perform segmentation processing on the target spray image to obtain a segmented image.

[0024] In step S201, the target spray image is a stable image during the spray of the injector collected, that is, the number of sprays in the target spray image is equal to the number of injector nozzles. Edge detection is performed on the target spray image to detect pixel points with obvious brightness changes in the target spray image. The region of interest is the region containing the spray, and the number of sprays is the number of regions of interest.

[0025] In this embodiment, a target spray image for detecting spray quality is determined from the sequence spray images collected during the spray of the injector. Referring to Figure 3 , which is a schematic diagram of a target spray image provided by an embodiment of the present invention. The number of spray beams in the image is equal to the number of injector nozzles.

[0026] Segment the target spray image to obtain a segmented image. During the segmentation process, set the corresponding pixel threshold. When the pixel value of a pixel point in the target spray image is greater than the pixel threshold, determine the pixel point as a target pixel point. When the pixel value of a pixel point in the target spray image is not greater than the pixel threshold, set the pixel value of the pixel point to 0 to obtain the corresponding segmented image. The pixel threshold can be set according to the actual situation and is not limited in this embodiment.

[0027] In this embodiment, obtain the target spray image for detecting the spray quality, segment the target spray image to obtain a segmented image, so as to determine the spray area according to the segmented image, and then determine the corresponding spray characteristics according to the spray area.

[0028] Optionally, obtaining the target spray image for detecting the spray quality includes: Obtain the sequence spray images of the spray ejected by the fuel injector, and extract the number of peaks in the waveform diagram of the current spray image in the sequence spray images; Obtain the number of spray holes in the sprayer, and determine whether the number of spray holes is equal to the number of peaks; If the number of spray holes is equal to the number of peaks, determine the current spray image as the target spray image.

[0029] In this embodiment, obtain the sequence spray images of the spray ejected by the fuel injector. Among them, the sequence spray images include all images from the start to the end of the fuel injector spray. When obtaining the sequence spray images of the spray ejected by the fuel injector, control the fuel injector driving device through the high-pressure fuel supply system to drive the fuel injector to spray. Install a high-speed camera in front of the fuel injector injection separated by plexiglass to collect the spray images of the fuel injector. The high-speed camera is placed perpendicular to the axis of the fuel injector, and the distance and focal length are adjusted so that it just collects a 10 cm × 10 cm area on the plane where the axis of the fuel injector is located. Since the exposure time is extremely short, an area light source needs to be added to supplement the light for collecting the spray transient images.

[0030] After obtaining the sequence spray images of the spray ejected by the fuel injector, take one of the spray images as the current spray image, extract the waveform diagram of the current spray image, determine the number of peaks in the waveform diagram, obtain the number of spray holes in the sprayer, and determine whether the number of spray holes is equal to the number of peaks; if the number of spray holes is equal to the number of peaks, determine the current spray image as the target spray image. Among them, the waveform diagram is the frequency-domain waveform diagram of the current spray image.

[0031] In this embodiment, determine the target spray image by judging whether the number of spray holes is equal to the number of peaks, so as to ensure that the target spray image is the image when the fuel injector sprays stably, thereby improving the accuracy of detecting the spray quality based on the target spray image.

[0032] Optionally, perform segmentation processing on the target spray image to obtain a segmented image, including: Determine the spray images before the target spray image in the sequence of spray images as the target sequence of spray images; Obtain a background image, determine the first image in the target sequence of spray images as the target image, and calculate the pixel difference between each corresponding pixel point of the target image and the background image; If there is a pixel difference greater than a preset difference threshold, determine the background image as the target background image; If there is no pixel difference greater than the preset difference threshold, determine the target image as the background image, determine the first image in the remaining images of the target sequence of spray images as the target image, and perform the step of calculating the pixel difference between the target image and the background image until the target background image is obtained; Perform background denoising on the target spray image according to the target background image to obtain a denoised spray image; Perform segmentation processing on the denoised spray image to obtain a segmented image.

[0033] In this embodiment, in order to improve the segmentation accuracy of the segmented image, background denoising processing is performed on the target spray image, and segmentation processing is performed on the denoised spray image. The spray images before the target spray image in the sequence of spray images are determined as the target sequence of spray images, a background image is obtained, the first image in the target sequence of spray images is determined as the target image, and the pixel difference between each corresponding pixel point of the target image and the background image is calculated. This is to facilitate determining whether spraying starts based on the pixel difference. Among them, the size of the original background image is equal to the size of the target spray image.

[0034] It should be noted that when calculating the pixel difference between the target image and the background image, subtract the pixel value of the corresponding pixel point in the target image and the background image to obtain the pixel difference between each pixel point. For example, calculate the pixel difference between the pixel value of the pixel point in the third row and the third column of the target image and the pixel value of the pixel point in the third row and the third column of the background image.

[0035] Determine whether the pixel difference between each pixel point is greater than a preset difference threshold. If there is a pixel difference greater than the preset difference threshold, it is considered that there is spray in the target image, and the background image is no longer updated. Instead, the background image is determined as the target background image. If there is no pixel difference greater than the preset difference threshold, it is considered that there is no spray in the target image. The pixel difference that appears may be caused by environmental noise, indicating that there is environmental noise in the target image. When performing segmentation processing on the target spray image, the corresponding background noise needs to be removed. Therefore, the target image is determined as the background image so that when performing background denoising on the target spray image based on the background image, the background noise existing in the target spray image can be removed. Among them, the preset difference threshold can be determined according to the actual situation, which is not limited in this embodiment.

[0036] After determining the target image as the background image, determine a new target image. Determine the first image in the remaining images of the target sequence spray image as the target image, and perform the step of calculating the pixel difference between the target image and the background image. Determine whether there is a pixel difference greater than the preset difference threshold. If there is a pixel difference greater than the preset difference threshold, determine the background image as the target background image; if there is no pixel difference greater than the preset difference threshold, determine the target image as the background image, determine the first image in the remaining images of the target sequence spray image as the target image, and perform the step of calculating the pixel difference between the target image and the background image until the target background image is obtained.

[0037] Perform background denoising on the target spray image according to the target background image to obtain a denoised spray image. Among them, when performing background denoising, subtract the target background image from the target spray image. Refer to Figure 4 , which is a schematic diagram of a denoised spray image provided by an embodiment of the present invention. Perform segmentation processing on the denoised spray image to obtain a segmented image. Among them, the segmentation processing process is the same as the above segmentation processing process, and will not be described in this embodiment.

[0038] In this embodiment, when performing segmentation processing on the target spray image, perform background denoising on the target spray image, use a progressive background based on the time series to determine the target background image, and when performing background denoising on the target spray image according to the target background image, to remove environmental interference during the acquisition of the spray image. To avoid when performing segmentation on the target spray image, when the pixel values of the spray area and the noise area are similar, determining the corresponding noise area as the spray area, so that the segmented image after segmentation only contains the corresponding spray area, improving the segmentation accuracy of the segmented image.

[0039] S202: Determine the position of the spray root in the target spray image according to the segmented image. Based on the position of the spray root, calculate the distance between each target pixel point in the segmented image and the spray root. According to the distance, determine the dilation matrix corresponding to each target pixel point, and perform morphological dilation processing on the segmented image according to the dilation matrix to obtain the target area and the number of sprays corresponding to the spray.

[0040] In step S202, the position of the spray root is the position of the nozzle hole. The target pixel points are the pixel points segmented in the segmented image, that is, the pixel points with pixel values greater than the pixel threshold. Calculate the distance between each target pixel point in the segmented image and the spray root, that is, calculate the distance between the spray area and the nozzle hole. According to the distance, the dilation matrix is used to perform dilation processing on the segmented image. The target area is the area where the spray diverges after spraying from the nozzle hole, and the number of sprays is the number of corresponding target areas.

[0041] In this embodiment, according to the segmented image, determine the position of the spray root in the target spray image, where the position of the spray root is the position of the nozzle hole in the fuel injector. When determining the position of the nozzle hole, since the nozzle hole of the fuel injector sprays oil outward from an extremely small hole, the spray shows a phenomenon of being dense and sparse from near to far from the nozzle hole. Therefore, the farther away from the nozzle hole, the more divergent the spray. Therefore, the position of the spray root, that is, the position of the nozzle hole, can be determined according to the aggregation degree of the target pixel points in the segmented image.

[0042] It should be noted that in this embodiment, when determining the position of the spray root in the target spray image, the density peak clustering method can be used for determination, that is, according to the aggregation degree of the target pixel points in the target spray image, determine the aggregation points, and determine the corresponding aggregation points as the spray root. It can also be determined by using other methods, which are not limited in this embodiment.

[0043] It should be noted that if there are multiple nozzle holes in the fuel injector, then there are multiple spray roots in the target spray image, that is, the positions of multiple nozzle holes.

[0044] Based on the position of the spray root, calculate the distance between each target pixel point in the segmented image and the spray root. Among them, the target pixel points are the pixel points with pixel values greater than the pixel threshold in the segmented image, that is, the pixel points including the spray area. According to the distance, determine the dilation matrix corresponding to each target pixel point. Among them, the size of the dilation matrix is related to the distance between the target pixel point and the spray root. The larger the distance, the larger the dilation matrix, and the smaller the distance, the smaller the dilation matrix.

[0045] It should be noted that when there are multiple spray roots, the divergence area corresponding to the spray roots can be determined first, and the distance between the target pixel point and the spray root in the divergence area where the target pixel point is located can be calculated. It is also possible to calculate the distance between the target pixel point and the nearest spray root. It is also possible to use other methods to calculate the distance between the target pixel point and the spray root, which is not limited in this embodiment.

[0046] It should be noted that when determining the dilation matrix corresponding to each target pixel point according to the distance, the distance is rounded up to obtain the rounded distance, and the size of the dilation matrix can be , where D is the rounded distance. The size of the dilation matrix can also be determined according to the actual situation, which is not limited in this embodiment.

[0047] The segmented image is subjected to morphological dilation processing according to the dilation matrix to obtain the target area and the number of sprays corresponding to the spray. Among them, the morphological dilation processing of the segmented image is to connect the divergent spray areas to form a closed area, obtain the target area corresponding to the spray, and determine the number of sprays according to the number of target areas. The number of sprays is equal to the number of target areas.

[0048] In this embodiment, according to the segmented image, the position of the spray root in the target spray image is determined, that is, the position of the spray hole is determined, so as to calculate the distance between the target pixel point and the spray hole, determine the divergence degree of the spray ejected from the spray hole, and determine the dilation matrix corresponding to each target pixel point according to the distance, so that the size of the dilation matrix is related to the size of the pixel point distance from the spray root, avoiding the error caused by the dilation of the spray root. According to the dilation matrix, the segmented image is subjected to morphological dilation processing, and the discontinuous target pixel points can be connected to form a corresponding closed target area.

[0049] Optionally, determining the dilation matrix corresponding to each target pixel point according to the distance includes: Determining the edge target pixel points and non-edge target pixel points in the segmented image; Obtaining the first target distance coefficient of the non-edge target pixel points and the second target distance coefficient of the edge target pixel points; Calculating the dilation matrix of each target pixel point according to the first target distance coefficient, the second target distance coefficient, and the distance.

[0050] In this embodiment, it is judged whether the target pixel point is located in the edge area of the segmented image, and the edge target pixel points and non-edge target pixel points in the segmented image are determined. Among them, the edge target pixel points are the target pixel points at the edge in the segmented image, and the non-edge target pixel points are the target pixel points not at the edge in the segmented image.

[0051] Obtain the first target distance coefficient of non-edge target pixels and the second target distance coefficient of edge target pixels, that is, determine the first target distance coefficient of non-edge target pixels and the second target distance coefficient of edge target pixels according to whether the corresponding target pixels are edge points. In this embodiment, the first target distance coefficient of non-edge target pixels is K, and the second target distance coefficient of edge target pixels is 0. The value of K is obtained based on experimental experience and is determined by the pressure of the injector.

[0052] According to the first target distance coefficient and the second target distance coefficient, and the distance, calculate the dilation matrix corresponding to the target pixels. Among them, the size of the dilation matrix corresponding to non-edge target pixels is the ceiling value of the product of the first target distance coefficient and the distance. For example, the size of the dilation matrix corresponding to non-edge target pixels is [Y×Y], where Y is the ceiling value of the product of the first target distance coefficient and the distance between the corresponding target pixel and the spray root, that is, the ceiling value of K×D, where K is the first target distance coefficient and D is the distance between the target pixel and the corresponding spray root, and this distance is the Euclidean distance. The size of the dilation matrix corresponding to edge target pixels is 0, that is, morphological dilation is not performed on edge target pixels.

[0053] In this embodiment, the first target distance coefficient of non-edge target pixels and the second target distance coefficient of edge target pixels are obtained respectively, and the corresponding target distance coefficient is multiplied by the distance of the corresponding target pixel to determine the dilation matrix of the corresponding target pixel, so that the dilation matrix of the corresponding non-edge target pixel increases with the increase of the distance, which can better connect discontinuous target pixels, and no dilation is performed on edge target pixels, avoiding the error caused by the dilation of edge target pixels.

[0054] S203: Calculate the spray area, spray angle and spray deflection angle of each target area.

[0055] In step S203, the spray area is the size of each target area, the spray angle is the angle between the spray edge lines of the corresponding target area, and the spray deflection angle is the deflection angle between adjacent target areas.

[0056] In this embodiment, the region after morphological dilation processing is a continuous region. When calculating the area of the target region, the number of target pixels included in the target region can be calculated. When calculating the spray angle and spray deflection angle of the target region, edge detection needs to be performed on the target region, and the edge lines of the target region are extracted. The angle between the edge lines of the target region is determined as the spray angle, that is, the expansion angle of the spray after passing through the spray hole. The angle between the centerlines of each spray angle is used as the spray deflection angle. Among them, if there is only one target region, that is, when the fuel injector only includes one spray hole, the spray deflection angle is 0. If there are multiple target regions, the spray area, spray angle, and spray deflection angle of each target region are calculated. That is, when the fuel injector includes multiple spray holes, the number of target pixels included in each target region needs to be calculated, that is, the spray area of each target region, the angle between the edge lines of each target region is calculated, that is, the spray angle of each target region, and the angle between the centerlines of the spray angles of adjacent target regions is calculated, that is, the deflection angle of each target region. Among them, the spray deflection angle is the angle between the centerline of the spray angle in the target region adjacent to this target region after counterclockwise rotation and the centerline of the spray angle of this target region.

[0057] It should be noted that when performing edge detection on each target region, the Canny operator edge detection method can be used for edge detection to obtain the edge of the target region. According to the edge of the target region, the edge lines of the target region are extracted. Among them, when extracting the edge lines of the target region, the Hough transform algorithm can be used for line extraction. Other algorithms can also be used for edge detection and line extraction, which are not limited in this embodiment.

[0058] In this embodiment, the spray area, spray angle, and spray deflection angle of each target region are calculated to obtain the quantization values during the fuel injection process of the fuel injector, so as to perform quality inspection based on multiple quantization values during the fuel injection process of the fuel injector.

[0059] Optionally, calculating the spray area, spray angle, and spray deflection angle of each target region includes: Calculating the gradient magnitude and gradient direction of each target pixel in the target region; Extracting the spray edge of the target region according to the gradient magnitude and gradient direction of each target pixel; According to the spray edge, extracting the edge lines of the target region, and according to the edge lines, calculating the spray area, spray angle, and spray deflection angle of each target region.

[0060] In this embodiment, when performing edge detection on the target region, the gradient magnitude and gradient direction of each target pixel in the target region are calculated. Among them, the calculation formulas for the gradient magnitude and gradient direction of each target pixel are as follows: ; Among them, I(i, j) is the pixel value corresponding to the target pixel point (i, j), I(i, j + 1) is the pixel value of the pixel point (i, j + 1), I(i + 1, j + 1) is the pixel value of the pixel point (i + 1, j + 1), I(i + 1, j) is the pixel value of the pixel point (i + 1, j), fx(i, j) is the partial derivative of the pixel value of the target pixel point (i, j) in the x direction, fy(i, j) is the partial derivative of the pixel value of the target pixel point (i, j) in the y direction, M(i, j) is the gradient magnitude of the target pixel point (i, j), and H(i, j) is the gradient direction of the target pixel point (i, j).

[0061] In this embodiment, in order to balance the requirements of accurate edge positioning and noise suppression in gradient magnitude calculation, it is selected to determine the gradient magnitude and gradient direction of the target pixel point by calculating the finite differences of the first-order partial derivatives in the x direction, y direction, 45° direction, and 135° direction within the 8-neighborhood of the target pixel point.

[0062] According to the gradient direction of each target pixel point, calculate two adjacent pixel points in the opposite direction of the straight line where the gradient direction is located for the target pixel point. According to the pixel values corresponding to the two adjacent pixel points, calculate the gradient magnitudes of the two adjacent pixel points to obtain the adjacent gradient magnitudes, and judge the magnitude relationship between the gradient magnitude of the target pixel point and the corresponding adjacent gradient magnitudes. According to the judgment result, judge whether the target pixel point is the edge of the spray.

[0063] It should be noted that when calculating two adjacent pixel points in the opposite direction of the straight line where the gradient direction is located for each target pixel point according to the gradient direction of the target pixel point, the calculation formula is as follows: ; Among them, i is the abscissa of the target pixel point, and j is the ordinate of the target pixel point. is the gradient direction of the target pixel point (i, j). and are two pixel points adjacent to the corresponding target pixel point. Among them, The coordinates of are , The coordinates of are .

[0064] When calculating the gradient magnitudes of two adjacent pixel points according to the pixel values corresponding to the two adjacent pixel points, the interpolation method can be used to calculate the gradient magnitudes of the two adjacent pixel points. The calculation formula is as follows: ; Among them, M(i, j) is the gradient magnitude of the target pixel point (i, j), M(i + 1, j) is the gradient magnitude of the pixel point (i + 1, j), M(i, j + 1) is the gradient magnitude of the pixel point (i, j + 1), and M(i + 1, j + 1) is the gradient magnitude of the pixel point (i + 1, j + 1). is the gradient direction of the target pixel point (i, j). is the adjacent pixel point 's gradient magnitude. is the adjacent pixel point 's gradient magnitude.

[0065] Judge the magnitude of the gradient magnitude of the target pixel point and the corresponding adjacent gradient magnitudes. If the gradient magnitude of the target pixel point is less than the gradient magnitude of the adjacent pixel point and less than the gradient magnitude of the adjacent pixel point , it is determined that the target pixel point is not the spray edge.

[0066] After determining the target pixel points that are not the spray edge, perform thresholding on the remaining target pixel points to obtain a preset high threshold and a preset low threshold. Among them, the preset high threshold is the pixel value that can clearly distinguish strong edges, and the preset low threshold is slightly lower than the preset high threshold.

[0067] Retain the pixel points greater than the preset high threshold, assign the pixel values of other pixel points to 0 to obtain a high-threshold image. Retain the pixel points less than or equal to the preset low threshold, assign the pixel values of other pixel points to 0 to obtain a low-threshold image. In the high-threshold image, connect the edge contours, find all non-zero pixels as starting points, track and connect adjacent edge pixels along the edge direction, and use the low-threshold image to fill the edge gaps. For each endpoint, search for weak edge points in the neighborhood of the low-threshold image and connect the weak edge points to form a complete edge contour to obtain the final spray edge.

[0068] Perform line detection on the spray edge to obtain the edge line of the spray edge. Among them, in this embodiment, the Hough transform algorithm is used for line detection. Other algorithms can also be used for line detection, which is not limited in this embodiment.

[0069] After extracting the edge lines of the target area, determine the endpoint coordinates of the line intersections in the target area, calculate the area of the target area according to each endpoint coordinate, that is, the spray area, calculate the angle between the edge lines at the spray root, that is, the spray angle, and calculate the angle between the centerlines of the spray angles of adjacent target areas, that is, the deflection angle of each target area. Among them, the spray deflection angle is the angle between the centerline of the spray angle in the target area adjacent to this target area after counterclockwise rotation and the centerline of the spray angle in this target area.

[0070] In this embodiment, according to the gradient magnitude and gradient direction of each target pixel, the spray edge of the target area is extracted to improve the accuracy of extracting the spray edge.

[0071] S204: Determine the spray characteristics of the target spray image according to the number of sprays, the spray area of each target area, the spray angle of each target area, and the spray deflection angle of each target area. Classify the spray quality according to the spray characteristics to obtain a classification result.

[0072] In step S204, the number of sprays, the spray area of each target area, the spray angle of each target area, and the spray deflection angle of each target area are used as the spray characteristics of the target spray image. The spray quality is classified according to the spray characteristics to obtain a classification result, where the classification result includes qualified quality and unqualified quality.

[0073] In this embodiment, when determining the spray characteristics of the target spray image according to the number of sprays, the spray area of each target area, the spray angle of each target area, and the spray deflection angle of each target area, calculate the average value of the spray areas of all target areas, the average value of the spray angles of all target areas, and the average value of the spray deflection angles of all target areas. The number of sprays, the average value of the spray area, the average value of the spray angle, and the average value of the spray deflection angle are determined as the spray characteristics of the target spray image.

[0074] When classifying the spray quality according to the spray characteristics, boolean statistics are performed on the number of sprays, the average value of the spray area, the average value of the spray angle, and the average value of the spray deflection angle respectively. For example, obtain the number of spray holes, the standard spray area, the standard spray angle, and the standard spray deflection angle. If the number of sprays is equal to the number of spray holes, the number of sprays is determined to be true; otherwise, it is determined to be false. If the difference between the average value of the spray area and the average value of the standard spray area is less than the preset area threshold, the average value of the spray area is determined to be true; otherwise, it is determined to be false. If the difference between the average value of the spray angle and the average value of the standard spray angle is less than the preset angle threshold, the average value of the spray angle is determined to be true; otherwise, it is determined to be false. If the difference between the average value of the spray deflection angle and the average value of the standard spray deflection angle is less than the preset deflection angle threshold, the average value of the spray deflection angle is determined to be true; otherwise, it is determined to be false. Among them, the preset area threshold, the preset angle threshold, and the preset deflection angle threshold are set according to the actual situation, and are not limited in this embodiment.

[0075] Classify the spray quality according to the statistical results to obtain a classification result. For example, if all the statistical results are true values, the classification result is determined to be qualified quality; if there are false values in the statistical results, the classification result is determined to be unqualified quality.

[0076] To verify the accuracy and robustness of spray quality detection using corresponding spray characteristics, spray quality detection was performed on the target spray images of 100 actually collected fuel injectors, and the quality detection results are shown in the following table:

[0077] Among them, the spray area is the average value of the spray areas of each target spray image, the spray angle is the average value of the spray angles of each target spray image, and the spray deflection angle is the average value of the spray deflection angles of each target spray image.

[0078] In this embodiment, various types of characteristics are used to classify the spray quality, improving the classification accuracy.

[0079] In another embodiment, when determining the spray characteristics of the target spray area, the characteristics of the corresponding target area can also be determined according to the spray area of the corresponding target area, the spray angle of the corresponding target area, and the spray deflection angle of the corresponding target area. That is, for any target area, the spray area, spray angle, and spray deflection angle of the target area are spliced to obtain the characteristics of the corresponding target area, and then the spray quantity is spliced with the characteristics of each target area to obtain the spray characteristics. The truth or falsehood of the spray quantity and the characteristics of each target area are statistically analyzed. For example, the number of spray holes and the standard characteristics are obtained; if the spray quantity is equal to the number of spray holes, the spray quantity is determined to be true, otherwise, it is determined to be false; for any target area, if the similarity between the characteristics of the target area and the standard characteristics is greater than the preset similarity threshold, the characteristics of the target area are determined to be true, otherwise, it is determined to be false. Among them, for different numbers of spray holes, the corresponding standard characteristics are different, and the similarity threshold is set according to the actual situation, which is not limited in this embodiment.

[0080] The spray quality is classified according to the statistical results to obtain the classification result. For example, if all the statistical results are true values, the classification result is determined to be qualified in quality; if there are false values in the statistical results, the classification result is determined to be unqualified in quality.

[0081] Optionally, determining the spray characteristics of the target spray image includes: Determining the spray area characteristics according to the spray area of each target area; Determining the spray angle characteristics according to the spray angle of each target area; Determining the spray deflection angle characteristics according to the spray deflection angle of each target area; Splicing and fusing the spray quantity, spray area characteristics, spray angle characteristics, and spray deflection angle characteristics to obtain the spray characteristics of the target spray image.

[0082] In this embodiment, the splicing order is marked for each target area, the spray areas of each target area are spliced according to the splicing order to obtain the spray area feature, the spray angles of each target area are spliced according to the splicing order to obtain the spray angle feature; the spray deflection angles of each target area are spliced according to the splicing order to obtain the spray deflection angle feature; the spray quantity, the spray area feature, the spray angle feature and the spray deflection angle feature are spliced and fused to obtain the spray feature of the target spray image.

[0083] In another embodiment, after determining the spray quantity, the spray area feature, the spray angle feature and the spray deflection angle feature, when determining the spray feature of the target spray image, the structural feature and the image feature of the target spray image can also be fused, wherein the structural feature of the target spray image includes the color feature, the texture feature and the shape feature, and the image feature of the target spray image is the feature extracted based on the convolutional neural network.

[0084] It should be noted that the color feature can include the frequency of each color appearance, statistical quantities such as the mean, variance and skewness of the color. The texture feature can be the spatial relationship of pixel gray values. The shape feature can be the edge coordinate sequence of the target area.

[0085] It should be noted that the convolutional neural network structure can include: a convolutional layer, a pooling layer connected to the convolutional layer, and a global average pooling layer connected to the pooling layer; wherein, the convolutional layer is used to extract the features of the target spray image; the pooling layer is used to reduce the dimension of the features extracted by the convolutional layer to keep the targets in the target spray image translation invariant; the global average pooling is used to take the average value of all pixels in the image after dimension reduction as the feature value, and can extract the highly abstract features of the feature map.

[0086] The spray quantity, the spray area feature, the spray angle feature, the spray deflection angle feature, the structural feature and the image feature are fused to obtain the spray feature of the target spray image. Among them, when performing feature fusion, splicing fusion can be performed, or other methods can be used for fusion, which is not limited in this embodiment.

[0087] In this embodiment, by fusing each feature, the information from different dimensions can be integrated together, so that the spray feature can contain various types of data to provide rich image information, thereby improving the accuracy of the spray feature.

[0088] Optionally, the spray quality is classified according to the spray feature to obtain a classification result, including: Obtain a trained spray quality classification model; Input the spray feature into the trained spray quality classification model, and output the classification result.

[0089] In this embodiment, a trained spray quality classification model is obtained. The spray quality classification model is a deep learning model, such as a logistic regression model, or other types of classification models can also be used. This embodiment does not make any limitations.

[0090] Among them, the training process of the trained spray quality classification model includes the following steps: Obtain an initial spray quality classification model and training data. The training data includes sample spray features in the sample spray images and classification labels of the sample spray images. Use the training data to perform supervised training on the initial spray quality classification model to obtain a trained spray quality classification model. Among them, the sample spray features are obtained by using the method for determining the spray features of the target spray image described above for feature extraction.

[0091] Input the spray features into the trained spray quality classification model, and output a classification result. The classification result includes qualified quality and unqualified quality.

[0092] In this embodiment, the trained spray quality classification model is used to classify the spray quality. The spray quality classification model automatically processes the spray features, improving the processing efficiency. Moreover, the spray quality classification model can eliminate subjective biases and improve the classification accuracy.

[0093] In this application, the target spray image is segmented to determine the position of the spray root in the target spray image. According to the position of the spray root, the distance between each target pixel point in the segmented image and the spray root is calculated. According to the distance, the dilation matrix corresponding to each target pixel point is determined. The segmented image is subjected to morphological dilation processing according to the dilation matrix to improve the self-adaptability of the dilation matrix. So that when the segmented image is subjected to morphological dilation processing according to the dilation matrix, the target pixel points far from the nozzle can be connected to ensure the integrity of the spray. According to the number of sprays, the spray area of each target region, the spray angle of each target region, and the spray deflection angle of each target region, the spray features of the target spray image are determined. The spray quality is classified according to the spray features, taking into account the weight and shape of the spray, and improving the accuracy of spray quality detection.

[0094] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0095] Please refer to Figure 5 , Figure 5 which is a schematic structural diagram of a spray quality detection device for an injector provided by an embodiment of the present invention. The spray quality detection device for the injector corresponds one-to-one with the system fault prediction method in the above embodiment. Specifically, please refer toFigure 2 the relevant descriptions in the corresponding embodiments. For ease of explanation, only the parts related to this embodiment are shown. Refer to Figure 5 , the spray quality detection device 50 includes: an acquisition module 51, a processing module 52, a calculation module 53, and a classification module 54.

[0096] The acquisition module 51 is configured to acquire a target spray image for detecting spray quality, perform segmentation processing on the target spray image, and obtain a segmented image.

[0097] The processing module 52 is configured to determine the position of the spray root in the target spray image according to the segmented image, calculate the distance between each target pixel point in the segmented image and the spray root according to the position of the spray root, determine the dilation matrix corresponding to each target pixel point according to the distance, and perform morphological dilation processing on the segmented image according to the dilation matrix to obtain the target area and the number of sprays corresponding to the spray.

[0098] The calculation module 53 is configured to calculate the spray area, spray angle, and spray deflection angle of each target area.

[0099] The classification module 54 is configured to determine the spray characteristics of the target spray image according to the number of sprays, the spray area of each target area, the spray angle of each target area, and the spray deflection angle of each target area, classify the spray quality according to the spray characteristics, and obtain a classification result.

[0100] Optionally, the above acquisition module 51 includes: An acquisition unit configured to acquire a sequence of spray images of the injector spraying, and extract the number of wave peaks in the waveform diagram of the current spray image in the sequence of spray images.

[0101] A judgment unit configured to acquire the number of spray holes in the sprayer and judge whether the number of spray holes is equal to the number of wave peaks.

[0102] A first determination unit configured to, if the number of spray holes is equal to the number of wave peaks, determine the current spray image as the target spray image.

[0103] Optionally, the above acquisition module 51 further includes: A second determination unit configured to determine the spray images before the target spray image in the sequence of spray images as the target sequence spray images.

[0104] A first calculation unit configured to acquire a background image, determine the first image in the target sequence spray images as the target image, and calculate the pixel differences between each pixel point of the target image and the background image.

[0105] A third determination unit, configured to determine the background image as the target background image if there are pixel point differences greater than a preset difference threshold.

[0106] An execution unit, configured to determine the target image as the background image if there are no pixel point differences greater than the preset difference threshold, determine the first image in the remaining images of the target sequence spray image as the target image, and execute the step of calculating the pixel point difference between the target image and the background image until the target background image is obtained.

[0107] A denoising unit, configured to perform background denoising on the target spray image according to the target background image to obtain a denoised spray image.

[0108] A segmentation unit, configured to perform segmentation processing on the denoised spray image to obtain a segmented image.

[0109] Optionally, the above processing module 52 includes: A fourth determination unit, configured to determine edge target pixel points and non-edge target pixel points in the segmented image.

[0110] A second acquisition unit, configured to acquire a first target distance coefficient of the non-edge target pixel points and a second target distance coefficient of the edge target pixel points.

[0111] A second calculation unit, configured to calculate a dilation matrix of each target pixel point according to the first target distance coefficient, the second target distance coefficient, and the distance.

[0112] Optionally, the above calculation module 53 includes: A third calculation unit, configured to calculate the gradient magnitude and gradient direction of each target pixel point in the target area.

[0113] An extraction unit, configured to extract the spray edge of the target area according to the gradient magnitude and gradient direction of each target pixel point.

[0114] A fourth calculation unit, configured to extract the edge straight line of the target area according to the spray edge, and calculate the spray area, spray angle, and spray deflection angle of each target area according to the edge straight line.

[0115] Optionally, the above classification module 54 includes: A fifth determination unit, configured to determine the spray area feature according to the spray area of each target area.

[0116] A sixth determination unit, configured to determine the spray angle feature according to the spray angle of each target area.

[0117] A seventh determination unit, configured to determine the spray deflection angle feature according to the spray deflection angle of each target area.

[0118] An obtaining unit is configured to splice and fuse the number of sprays, spray area characteristics, spray angle characteristics, and spray deflection angle characteristics to obtain the spray characteristics of the target spray image.

[0119] Optionally, the above classification module 54 further includes: A second obtaining unit is configured to obtain a trained spray quality classification model.

[0120] An output unit is configured to input the spray characteristics into the trained spray quality classification model and output a classification result.

[0121] For the specific limitations of the spray quality detection device for the fuel injector, reference can be made to the limitations of the spray quality detection method for the fuel injector in the above text, which will not be elaborated here. Each module in the above spray quality detection device for the fuel injector can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0122] In one embodiment, a computer device is provided. The computer device can be a server, and its internal structure diagram can be as Figure 6 shown. The computer device includes a processor, a memory, a network interface, and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to provide the target spray image. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a spray quality detection method for a fuel injector.

[0123] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the spray quality detection method for the fuel injector in the above embodiment, such as Figure 2 S201 - S204 shown. To avoid repetition, it will not be elaborated here. Or, when the processor executes the computer program, it implements the functions of each module / unit in this embodiment of the spray quality detection device for the fuel injector, such as Figure 5 the functions of the obtaining module, the processing module, the calculation module, and the classification module shown. To avoid repetition, it will not be elaborated here.

[0124] In one embodiment, a computer-readable storage medium is provided. A computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, it implements the spray quality detection method of the injector in the above embodiment. For example Figure 2 S201 - S204 shown. To avoid repetition, it will not be elaborated here. Alternatively, when the computer program is executed by a processor, it implements the functions of each module / unit in the above embodiment of the spray quality detection device of the injector. For example Figure 5 the functions of the acquisition module, processing module, calculation module, and classification module shown. To avoid repetition, it will not be elaborated here.

[0125] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0126] Those skilled in the art can clearly understand that for the convenience and brevity of description, only the above division of each functional unit and module is used as an example. In practical applications, the above functions can be allocated to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0127] The above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for detecting the spray quality of an injector, characterized in that, The spray quality detection method includes: Obtaining a target spray image for detecting spray quality, performing segmentation processing on the target spray image to obtain a segmented image; According to the segmented image, determining the position of the spray root in the target spray image, calculating the distance between each target pixel point in the segmented image and the spray root according to the position of the spray root, determining the dilation matrix corresponding to each target pixel point according to the distance, and performing morphological dilation processing on the segmented image according to the dilation matrix to obtain the target area and the number of sprays corresponding to the spray; Calculating the spray area, spray angle and spray deflection angle of each target area; Determining the spray characteristics of the target spray image according to the number of sprays, the spray area of each target area, the spray angle of each target area and the spray deflection angle of each target area, and classifying the spray quality according to the spray characteristics to obtain a classification result.

2. The spray quality detection method according to claim 1, wherein The obtaining of the target spray image for detecting spray quality includes: Obtaining a sequence of spray images of the spray ejected by the injector, and extracting the number of wave peaks in the waveform diagram of the current spray image in the sequence of spray images; Obtaining the number of spray holes in the sprayer, and determining whether the number of spray holes is equal to the number of wave peaks; If the number of spray holes is equal to the number of wave peaks, determining the current spray image as the target spray image.

3. The spray quality detection method according to claim 2, wherein The performing of the segmentation processing on the target spray image to obtain a segmented image includes: Determining the spray images before the target spray image in the sequence of spray images as the target sequence of spray images; Obtaining a background image, determining the first image in the target sequence of spray images as the target image, and calculating the pixel difference between each pixel point of the target image and the background image; If there is a pixel difference greater than a preset difference threshold, determining the background image as the target background image; If there is no pixel difference greater than the preset difference threshold, determining the target image as the background image, determining the first image in the remaining images in the target sequence of spray images as the target image, and performing the step of calculating the pixel difference between the target image and the background image until the target background image is obtained; Performing background denoising on the target spray image according to the target background image to obtain a denoised spray image; Performing segmentation processing on the denoised spray image to obtain a segmented image.

4. The spray quality detection method according to claim 1, characterized in that, The determining of the dilation matrix corresponding to each target pixel point according to the distance includes: Determining the edge target pixel points and non-edge target pixel points in the segmented image; Obtaining the first target distance coefficient of the non-edge target pixel points and the second target distance coefficient of the edge target pixel points; Calculating the dilation matrix of each target pixel point according to the first target distance coefficient, the second target distance coefficient and the distance.

5. The spray quality detection method according to claim 1, characterized in that The calculating of the spray area, spray angle and spray deflection angle of the target area includes: Calculating the gradient amplitude and gradient direction of each target pixel point in the target area; Extract the spray edge of the target area according to the gradient magnitude and gradient direction of each target pixel point. Extract the edge straight line of the target area according to the spray edge, and calculate the spray area, spray angle and spray deflection angle of each target area according to the edge straight line.

6. The spray quality detection method according to claim 1, wherein, The determination of the spray characteristics of the target spray image includes: Determine the spray area characteristic according to the spray area of each target area. Determine the spray angle characteristic according to the spray angle of each target area. Determine the spray deflection angle characteristic according to the spray deflection angle of each target area. Splice and fuse the spray quantity, the spray area characteristic, the spray angle characteristic and the spray deflection angle characteristic to obtain the spray characteristic of the target spray image.

7. The spray quality detection method according to claim 1, wherein The classification of the spray quality according to the spray characteristic to obtain the classification result includes: Obtain a trained spray quality classification model. Input the spray characteristic into the trained spray quality classification model and output the classification result.

8. A spray quality detection device for an injector, characterized in that The spray quality detection device includes: An acquisition module for acquiring a target spray image for detecting spray quality, performing segmentation processing on the target spray image to obtain a segmented image. A processing module for determining the position of the spray root in the target spray image according to the segmented image, calculating the distance between each target pixel point in the segmented image and the spray root according to the position of the spray root, determining the dilation matrix corresponding to each target pixel point according to the distance, and performing morphological dilation processing on the segmented image according to the dilation matrix to obtain the target area and spray quantity of the corresponding spray. A calculation module for calculating the spray area, spray angle and spray deflection angle of each target area. A classification module for determining the spray characteristic of the target spray image according to the spray quantity, the spray area of each target area, the spray angle of each target area and the spray deflection angle of each target area, and classifying the spray quality according to the spray characteristic to obtain the classification result.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the spray quality detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the spray quality detection method according to any one of claims 1 to 7.

Images