Oil nozzle hole angle detection method based on machine vision and related device

Abstract

The invention relates to the technical field of oil nozzle detection, and discloses an oil nozzle hole angle detection method based on machine vision and a related device. The method comprises the following steps that image recognition and positioning are conducted on a standard sample piece, a datum line is obtained, and the datum line is a left-right bisector penetrating through the standard sample piece; performing image recognition on the to-be-detected oil nozzle to obtain the center of the oil nozzle and the center of the specific hole area of the oil nozzle, and connecting to obtain an oil nozzle hole angle line; and calculating an included angle between the oil nozzle hole angle line and the reference line, judging whether the included angle is within a preset qualified range, and if so, determining that the oil nozzle is qualified. Through the automatic detection technology, the efficiency, reliability and consistency of oil injection hole angle detection are improved, it is ensured that all products meet the strict quality standard, and therefore the defects of a traditional detection method are effectively avoided.

Description

Technical Field

[0001] This application relates to the technical field of fuel injector detection, and particularly to a method and related device for detecting the hole angle of a fuel injector based on machine vision. Background Art

[0002] As an important part of the engine fuel system, the main function of the fuel injector is to atomize the fuel and spray it into the combustion chamber to make it fully mixed with air, so as to achieve efficient combustion. As a key component of the electronic fuel injection engine, the hole angle of the fuel injector directly affects the precise control of the fuel injection volume, the atomization effect of the fuel injection, and the distribution of the fuel in the combustion chamber, and then determines the overall performance, fuel efficiency, and emission control level of the engine. Therefore, accurately detecting the hole angle of the fuel injector is a key link to ensure the engine performance, improve fuel economy, and optimize emission control.

[0003] Currently, the detection of the fuel injector hole angle mainly relies on manual operation, and spot checks are carried out on a two-dimensional measuring instrument using simple tooling fixtures. This method not only has low detection efficiency and high cost, but also is difficult to ensure accuracy. Even for a skilled worker, it takes at least five minutes to detect a single fuel injector. In addition, the measuring equipment is expensive and relies on customized tooling fixtures, with poor versatility. More importantly, manual detection is greatly affected by subjective factors, making it difficult to ensure consistency. At the same time, the spot check method cannot cover all products, and there is a risk of missed inspection. Summary of the Invention

[0004] In order to improve the efficiency, reliability, and consistency of the fuel injector hole angle detection and ensure that all products meet strict quality standards, this application provides a method and related device for detecting the hole angle of a fuel injector based on machine vision.

[0005] In the first aspect, this application provides a method for detecting the hole angle of a fuel injector based on machine vision, adopting the following technical solutions:

[0006] A method for detecting the hole angle of a fuel injector based on machine vision includes the following steps:

[0007] S1. Perform image recognition and positioning on the standard sample of the selected part number to obtain a reference line, where the reference line is the left-right bisector passing through the standard sample;

[0008] S2. Perform image recognition on the fuel injector to be detected corresponding to the selected part number to obtain the center of the fuel injector and the center of the specific hole area of the fuel injector, and connect them to obtain the fuel injector hole angle line;

[0009] S5. Calculate the included angle between the fuel injector hole angle line and the reference line, and determine whether the included angle is within the preset qualified range. If so, it is qualified.

[0010] Optionally, between S2 and S5, the following steps are further included:

[0011] S3. Perform static repeatability analysis on the first fuel injector to be detected of this part number;

[0012] S4. Perform dynamic repeatability analysis on the first fuel injector to be detected of this part number.

[0013] Optionally, S4 includes the following steps:

[0014] S41. Fix the fuel injector to be detected on the fixture;

[0015] S42. Repeat the measurement multiple times and record the included angle between the fuel injector hole angle line and the reference line, and calculate the static repeatability range. Wherein, the static repeatability range is the difference between the maximum value and the minimum value of the included angle between the fuel injector hole angle line and the reference line in multiple measurements;

[0016] S43. Judge whether the static repeatability range is less than the first preset threshold. If so, it is determined that the static repeatability test is passed.

[0017] Optionally, S5 includes the following steps:

[0018] S51. Remove the fixture or the fuel injector to be detected and reinstall it;

[0019] S52. Repeat the measurement multiple times and record the included angle between the fuel injector hole angle line and the reference line, and calculate the dynamic repeatability range. Wherein, the dynamic repeatability range is the difference between the maximum value and the minimum value of the included angle between the fuel injector hole angle line and the reference line in multiple measurements;

[0020] S53. Judge whether the dynamic repeatability range is less than the second preset threshold. If so, it is determined that the dynamic repeatability test is passed.

[0021] Optionally, S1 includes the following steps:

[0022] S11. Install the standard sample corresponding to the current fuel injector part number to be detected;

[0023] S12. Use an industrial camera to obtain an image of the standard sample;

[0024] S13. Process the image of the standard sample using the Canny edge extraction algorithm to identify the left edge and the right edge of the standard sample;

[0025] S14. Obtain the starting point of the left edge and the starting point of the right edge, and calculate the midpoint of the connection line between the starting point of the left edge and the starting point of the right edge, and use it as the first midpoint;

[0026] S15. Obtain the left edge end point and the right edge end point, calculate the midpoint of the line connecting the left edge end point and the right edge end point, and use it as the second midpoint;

[0027] S16. Connect the first midpoint and the second midpoint to obtain a straight line passing through the standard sample, which is used as the reference line for subsequent hole angle measurement.

[0028] Optionally, the steps of S2 are as follows:

[0029] S21. Process the image of the fuel injector using the Canny edge extraction algorithm to obtain the edge features of the fuel injector; among them, the edge features include the edge contour;

[0030] S22. Use the Hough transform to detect the circular area in the image, and filter out the circle representing the main structure of the fuel injector according to the radius size of the circle;

[0031] S23. Calculate the center coordinates of the filtered circle and use them as the center coordinates of the fuel injector;

[0032] S24. According to the currently detected fuel injector part number and the position of the reference line, set a rectangular area above the reference line and centered on the center of the fuel injector as the prior area where fuel injection holes may exist;

[0033] S25. Perform threshold segmentation on the prior area to convert it into a black and white binary image to highlight the possible hole areas;

[0034] S26. Perform connected component segmentation on the binary image to divide the set of pixel points connected together in the image into different areas;

[0035] S27. Calculate the center point coordinates of each connected component, and filter out the connected component closest to the reference line as the target fuel injector specific hole area.

[0036] In the second aspect, the present application provides a method for detecting the hole angle of a fuel injector based on machine vision, adopting the following technical solution:

[0037] A method for detecting the hole angle of a fuel injector based on machine vision, including a processor, and a program of the method for detecting the hole angle of a fuel injector based on machine vision described in any one of the above is running in the processor.

[0038] In the third aspect, the present application provides a storage medium, adopting the following technical solution:

[0039] A storage medium stores a program of the method for detecting the hole angle of a fuel injector based on machine vision described in any one of the above.

[0040] In summary, the present application includes at least one of the following beneficial technical effects: the automatic detection of the angle of the nozzle hole is realized. At the same time, combined with the reliability analysis method (static repeatability analysis and dynamic repeatability analysis), the influence of pixel jitter caused by slight changes in the light source and camera accuracy problems, and image pixel changes caused by workers changing materials on the detection accuracy are controlled within an acceptable range, ensuring the stability and consistency of the detection accuracy during the operation, and providing a solid guarantee for the quality control of the nozzle products. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a flow chart of a method for detecting an injector nozzle angle based on machine vision in a certain embodiment of the present application.

[0042] Figure 2 It is the standard sample image collected in this application.

[0043] Figure 3 is the fuel injector image collected in this application.

[0044] Figure 4 It is a static repeatability analysis flow chart in a certain embodiment of the present application.

[0045] Figure 5 This is a static repeatability result diagram in a certain embodiment of the present application.

[0046] Figure 6 It is a dynamic repeatability analysis flow chart in a certain embodiment of the present application.

[0047] Figure 7 This is a dynamic repeatability result diagram in a certain embodiment of the present application.

[0048] Figure 8 This is a diagram of hole angle detection results in a certain embodiment of the present application. DETAILED DESCRIPTION

[0049] Embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings.

[0050] In the description of this specification, the description with reference to the terms "certain embodiments", "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with the embodiments or examples are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0051] This embodiment of the present application discloses a method for detecting the angle of fuel injection nozzle holes based on machine vision. Refer to Figure 1 , which includes the following steps S1 - S5.

[0052] S1. Perform image recognition and positioning on the standard sample of the selected part number to obtain a reference line, where the reference line is the left - right bisector passing through the standard sample.

[0053] Specifically, in a certain embodiment, S1 includes the following sub - steps S11 - S17.

[0054] S11. Install the standard sample corresponding to the fuel injection nozzle part number to be detected currently.

[0055] To ensure the accuracy and consistency of measurement, first, it is necessary to perform image recognition and positioning on the standard sample and obtain a reference line for subsequent measurement. This reference line, which is the left - right bisector passing through the standard sample, can be used as a reference for subsequent hole angle measurement. To obtain this reference line, first install the standard sample corresponding to the fuel injection nozzle part number to be detected currently, and ensure that the sample is in a fixed position to reduce the interference of external factors on the measurement.

[0056] S12. Use an industrial camera to obtain an image of the standard sample.

[0057] This system uses an industrial camera with a resolution of 2000W and combines it with a ring light source to ensure the clarity and consistency of imaging. The camera is fixed at the top of the product and takes a panoramic view of the fuel injection nozzle sample from a top - down perspective, focusing on obtaining accurate information in the fuel injection hole area. The system's field of view is 24mm×18mm, the working distance is 155 ± 10mm, and the height of the light source can be adjusted within the range of 0 ± 30mm to adapt to the lighting requirements under different working conditions. All relevant calculations are performed in the digital image, and the units are all pixels to ensure the high - precision and standardization of the detection.

[0058] S13. Use the Canny edge extraction algorithm to process the image of the standard sample to identify the left edge and right edge of the standard sample.

[0059] Refer to Figure 2 , here the Canny edge extraction algorithm is used to pre - process the standard sample image to accurately identify its left edge and right edge. The Canny edge detection algorithm is a common edge detection method that can effectively remove noise and enhance the target contour by calculating gradients and non - maximum suppression.

[0060] S14. Obtain the starting point A1 of the left edge and the starting point A2 of the right edge, and calculate the mid - point of the line connecting the starting point A1 of the left edge and the starting point A2 of the right edge, and use it as the first mid - point A.

[0061] S15. Obtain the left edge end point B1 and the right edge end point B2, calculate the midpoint of the line connecting the left edge end point B_1 and the right edge end point B2, and use it as the second midpoint B.

[0062] S16. Connect the first midpoint A and the second midpoint B to obtain a straight line AB passing through the standard part, which is used as the reference line for subsequent hole angle measurement.

[0063] Finally, with reference to Figure 2 and Figure 3 , perform linear fitting on these two midpoints to obtain a straight line passing through the standard part, which is the reference line. This reference line is used as a reference for subsequent fuel injector hole angle measurement, enabling the detection system to perform measurements based on a stable alignment method without relying on manual adjustment or additional mechanical jigs. Traditional detection methods often require manual operation of a two-dimensional measuring instrument, which takes a long time and is easily affected by subjective factors. In contrast, this method automatically generates a reference line through image processing technology, achieving the efficiency and consistency of fuel injector angle detection.

[0064] S2. Perform image recognition on the fuel injector to be detected corresponding to the selected part number, obtain the center of the fuel injector and the center of the specific hole area of the fuel injector, and connect them to obtain the fuel injector hole angle line.

[0065] Specifically, in one embodiment, S2 includes the following sub-steps S21 - S27.

[0066] S21. Process the image of the fuel injector using the Canny edge extraction algorithm to obtain the edge features of the fuel injector; among them, the edge features include the edge contour.

[0067] First, it is necessary to perform image recognition on the fuel injector to be detected and extract its edge features. Use the Canny edge extraction algorithm to process the image of the fuel injector to obtain a clear edge contour.

[0068] S22. Use the Hough transform to detect circular regions in the image and filter out the circles representing the main structure of the fuel injector according to the radius size of the circles.

[0069] S23. Calculate the center coordinates of the filtered circles and use them as the center coordinates of the fuel injector.

[0070] After obtaining the edge information of the fuel injector, use the Hough transform to further detect circular regions in the image and filter out the circles corresponding to the main structure of the fuel injector according to the radius size of the circles. The Hough transform is a mathematical method for detecting specific shapes (such as straight lines, circles, etc.) and is applicable to objects with clear edges in the image. In this detection method, the main body of the fuel injector is usually cylindrical, so the circular cross-section of the fuel injector can be accurately identified through the Hough transform, and further calculate the center coordinates of the circle, which are used as the center coordinates of the fuel injector.

[0071] S24. Select a rectangular area above the reference line as the prior area where specific injection holes may exist according to the currently detected fuel injector part number and the reference line position.

[0072] After obtaining the center coordinates of the fuel injector, it is necessary to determine the center position of the specific hole area of the fuel injector. Based on the known reference line position, a rectangular area is set as the prior area to ensure that the target specific injection hole is within this area. The setting of this prior area helps to reduce the amount of calculation and improve the accuracy of recognition.

[0073] S25. Perform threshold segmentation on this prior area to convert it into a black-and-white binary image to highlight the possible hole areas.

[0074] Subsequently, perform threshold segmentation on this prior area to convert it into a black-and-white binary image to highlight the possible injection hole areas. Threshold segmentation is a common image processing method. By setting a gray threshold, the pixel points in the image are divided into foreground and background to enhance the contrast of the target area and make the injection hole area more obvious.

[0075] S26. Perform connected component segmentation on the binary image to divide the set of pixel points connected together in the image into different areas.

[0076] S27. Calculate the center point coordinates of each connected component, and screen out the connected component closest to the reference line as the target specific hole area of the fuel injector.

[0077] After obtaining the binary image, further perform connected component segmentation to divide the set of connected pixel points into different independent areas. Connected component segmentation can effectively identify multiple independent targets in the image and classify them according to the connectivity of the areas. In this detection scheme, this method is used to distinguish the possible injection hole areas and screen out the area closest to the reference line as the final specific hole area of the fuel injector. During the screening process, the center coordinates of each connected area are calculated, and the area closest to the reference line is selected as the final target injection hole area to ensure the precise positioning of the injection hole.

[0078] As an example, in a certain embodiment, first obtain the center coordinates of the fuel injector: Use the Canny edge extraction algorithm to obtain the edge features of the fuel injector to get a clear edge contour. Then, use the Hough transform to detect the circular areas in the image, and screen out the circle representing the main structure of the fuel injector according to the radius size of the circle. Finally, calculate the center coordinates of this circle and use it as the center coordinates of the fuel injector (2050.32, 1216.75).

[0079] Then, obtain the central coordinates of the specific hole area of a specific fuel injector: According to the fuel injector part number detected currently and the known position of the reference line, set a rectangular area above the reference line and centered on the center of the fuel injector as the prior area where fuel injection holes may exist. Perform threshold segmentation on this prior area to convert it into a black-and-white binary image, highlighting the possible hole areas. Perform connected component segmentation on the binary image to divide the set of pixel points connected together in the image into different areas. Screen out the connected component closest to the reference line and use it as the specific hole area of the target fuel injector, and calculate its center point coordinates (1933.3, 848.081). Finally, connect the center of the fuel injector and the center of the specific hole area of the fuel injector to obtain the fuel injector hole angle line.

[0080] S3. Perform static repeatability analysis on the first fuel injector to be detected of this part number.

[0081] Specifically, referring to Figure 4 , in a certain embodiment, S3 includes the following sub-steps S31 - S33.

[0082] S31. Fix the fuel injector to be detected on the fixture.

[0083] S32. Repeat the measurement multiple times and record the included angle between the fuel injector hole angle line and the reference line, and calculate the static repeatability range. Among them, the static repeatability range is the difference between the maximum value and the minimum value of the included angle between the fuel injector hole angle line and the reference line in multiple measurements.

[0084] S33. Determine whether the static repeatability range is less than the first preset threshold. If so, it is determined that the static repeatability test is passed.

[0085] Taking the fuel injector to be detected in the part number XED95 as an example, the qualified fuel injector hole angle range is 175.20 ± 1°, and the first preset threshold for static repeatability analysis is set to 0.03°.

[0086] For example, keep the product and the fixture stationary, measure and record the included angle between the hole angle line and the reference line, and repeat 15 times. Calculate the range of all measurement results, as shown in Figure 5 . The range is 0.026° which is less than the set threshold 0.03°, that is, the static repeatability test is passed.

[0087] S4. Perform dynamic repeatability analysis on the first fuel injector to be detected of this part number.

[0088] Specifically, referring to Figure 6 , in a certain embodiment, S4 includes the following sub-steps S41 - S43.

[0089] S41. Remove the fixture or the fuel injector to be detected and reinstall it.

[0090] S42. Repeat the measurement multiple times and record the included angle between the nozzle hole angle line and the reference line, and calculate the dynamic repeatability range. The dynamic repeatability range is the difference between the maximum and minimum values of the included angle between the nozzle hole angle line and the reference line in multiple measurements.

[0091] S43. Determine whether the dynamic repeatability range is less than the second preset threshold. If so, it is determined that the dynamic repeatability test is passed.

[0092] Still taking the nozzle under test in the part number XED95 as an example, the second preset threshold for dynamic repeatability analysis is set to 0.06°.

[0093] For example, remove the fixture or product and reinstall it, measure and record the included angle between the hole angle line and the reference line, and repeat 15 times. Calculate the range of all measurement results, calculate the range of all included angles measured, and the result is as shown in 7. The range is 0.052° which is less than the set threshold of 0.06°, that is, the dynamic repeatability test is passed.

[0094] S5. Calculate the included angle between the nozzle hole angle line and the reference line, and determine whether this included angle is within the preset qualified range. If so, it is qualified.

[0095] In the detection process of the nozzle hole angle, the final measurement result needs to determine whether the nozzle is qualified by comparing the included angle between the nozzle hole angle line and the reference line. The reference line is the left - right bisector passing through the standard sample, and the nozzle hole angle line is the connection line between the center of the nozzle and the center of the specific hole area of the nozzle extracted by the image - processing method. By calculating the included angle between these two lines, it is possible to objectively evaluate whether the actual angle of the nozzle hole is within the preset qualified range, so as to achieve accurate determination of product quality.

[0096] In the specific calculation process, the acquisition of the nozzle hole angle line depends on the previous steps, including using the Canny edge extraction algorithm to identify the edge of the nozzle, using the Hough transform to extract the circular structure of the nozzle body, and calculating the center coordinates of the nozzle. At the same time, based on the connected - component segmentation, the center coordinates of the specific hole area of the nozzle are identified. By connecting the two key points, the nozzle hole angle line can be obtained. Next, calculate the included angle between this angle line and the reference line, and this included angle reflects the offset degree of the fuel injection hole relative to the standard sample.

[0097] To ensure the accuracy of the determination, the qualified standard for the injector nozzle hole angle is preset based on experimental data and process requirements. For example, the good product standard for a certain injector nozzle part number stipulates that the included angle should be between 175.20 ± 1°. If the calculated included angle value falls within this range, the detection result of the injector nozzle hole angle is determined to be qualified (OK); if it exceeds this range, it is determined to be unqualified (NG). This judgment method can strictly control the injection direction of the injector nozzle, ensure the fuel atomization effect and injection accuracy, and thus optimize the combustion efficiency, emission performance, and overall power output of the engine.

[0098] Compared with the traditional manual measurement method, this step adopts a fully automated calculation method, avoiding the errors that may be brought by manual angle reading, and at the same time improving the detection speed and consistency. Manual measurement usually relies on a two-dimensional measuring instrument, which requires skilled workers to operate, and data fluctuations may occur during the measurement process due to human factors. However, this method combines a high-precision industrial camera with an image processing algorithm to ensure that the measurement results of each injector nozzle can stably meet the quality requirements, thereby greatly improving the reliability and production efficiency of the detection. In addition, the automated angle calculation can also be applied to large-scale detection scenarios, realizing the efficient screening of the injector nozzle hole angle and reducing the quality risk caused by unqualified products entering the market.

[0099] For example, the included angle between the injector nozzle hole angle line and the reference line is measured to be 175.181°. Compare this included angle with the preset qualified range. If it falls within the qualified range of the included angle, record the detection result as "OK".

[0100] Optionally, when performing batch detection subsequently, replace the injector nozzle product to be tested, and repeat S2 to S5 until the detection of all injector nozzle products of this part number is completed.

[0101] The embodiment of the present application also discloses a method for detecting the injector nozzle hole angle based on machine vision, including a processor, and a program of the method for detecting the injector nozzle hole angle based on machine vision as described in any one of the above is run in the processor.

[0102] The embodiment of the present application also discloses a storage medium storing a program of the method for detecting the injector nozzle hole angle based on machine vision as described in any one of the above.

[0103] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A method for detecting the angle of fuel injection nozzle holes based on machine vision, characterized in that, It includes the following steps: S1. Perform image recognition and positioning on the standard sample of the selected part number to obtain a reference line, where the reference line is the left - right bisector passing through the standard sample; S2. Perform image recognition on the fuel injector to be detected corresponding to the selected part number to obtain the center of the fuel injector and the center of the specific hole area of the fuel injector, and connect them to obtain the fuel injector hole angle line; S5. Calculate the included angle between the fuel injector hole angle line and the reference line, and determine whether the included angle is within the preset qualified range. If so, it is qualified.

2. The method for detecting the angle of the fuel injection nozzle hole based on machine vision according to claim 1, characterized in that, Between S2 and S5, it also includes the following steps: S3. Perform static repeatability analysis on the first fuel injector to be detected of this part number; S4. Perform dynamic repeatability analysis on the first fuel injector to be detected of this part number.

3. The method for detecting the angle of the fuel injection nozzle hole based on machine vision according to claim 1, wherein, The said S4 includes the following steps: S41. Fix the fuel injector to be detected on the fixture; S42. Repeat measurement multiple times and record the included angle between the fuel injector hole angle line and the reference line, and calculate the static repeatability range. Among them, the static repeatability range is the difference between the maximum value and the minimum value of the included angle between the fuel injector hole angle line and the reference line in multiple measurements; S43. Determine whether the static repeatability range is less than the first preset threshold. If so, it is determined to pass the static repeatability test.

4. The method for detecting the angle of the fuel injection nozzle hole based on machine vision according to claim 1, characterized in that The said S5 includes the following steps: S51. Remove the fixture or the fuel injector to be detected and reinstall it; S52. Repeat measurement multiple times and record the included angle between the fuel injector hole angle line and the reference line, and calculate the dynamic repeatability range. Among them, the dynamic repeatability range is the difference between the maximum value and the minimum value of the included angle between the fuel injector hole angle line and the reference line in multiple measurements; S53. Determine whether the dynamic repeatability range is less than the second preset threshold. If so, it is determined to pass the dynamic repeatability test.

5. The method for detecting the angle of the fuel injection nozzle hole based on machine vision according to claim 1, wherein, The said S1 includes the following steps: S11. Install the standard sample corresponding to the current fuel injector part number to be detected; S12. Use an industrial camera to obtain an image of the standard sample; S13. Process the image of the standard sample using the Canny edge extraction algorithm to identify the left edge and the right edge of the standard sample; S14. Obtain the starting point of the left edge and the starting point of the right edge, calculate the mid - point of the connection line between the starting point of the left edge and the starting point of the right edge, and use it as the first mid - point; S15. Obtain the ending point of the left edge and the ending point of the right edge, calculate the mid - point of the connection line between the ending point of the left edge and the ending point of the right edge, and use it as the second mid - point; S16. Connect the first mid - point and the second mid - point to obtain a straight line passing through the standard sample, which is used as the reference line for subsequent hole angle measurement.

6. The method for detecting the angle of the fuel injection nozzle hole based on machine vision according to claim 1, wherein, The steps of the said S2: S21. Process the image of the fuel injector using the Canny edge extraction algorithm to obtain the edge features of the fuel injector; among them, the edge features include the edge contour; S22. Use the Hough transform to detect the circular area in the image, and screen out the circle representing the main structure of the fuel injector according to the radius size of the circle; S23. Calculate the center coordinates of the screened - out circle and use them as the center coordinates of the fuel injector; S24. According to the current detected fuel injector part number and the position of the reference line, set a rectangular area centered above the reference line and at the center of the fuel injector as the prior area where fuel injection holes may exist. S25. Threshold segmentation is performed on the prior region to convert it into a black-and-white binary image to highlight the possible hole regions; S26. Connected component segmentation is performed on the binary image to divide the set of pixel points connected together in the image into different regions; S27. Calculate the center point coordinates of each connected component, and screen out the connected component closest to the reference line as the specific hole region of the target fuel injector.

7. An injection nozzle hole angle detection system based on machine vision, characterized in that, It includes a processor, and a program of the fuel injector hole angle detection method based on machine vision as described in any one of claims 1-6 runs in the processor.

8. A storage medium, characterized in that, Store a program of the fuel injector hole angle detection method based on machine vision as described in any one of claims 1-6.

Images