CN117723452B - Automatic detection method and device for injection information of fuel nozzle of engine - Google Patents

Abstract

The embodiment of the invention provides an automatic detection method and device for injection information of a fuel nozzle of an engine, and belongs to the field of atomization detection of fuel nozzles by an image processing technology. The automatic detection method of the injection information comprises the following steps: calibrating a camera and acquiring a calibration parameter M of the camera; acquiring a single-frame image I in one-time detection of the fuel nozzle operation by using the calibrated camera; according to the single-frame image I, determining an effective fog drop area S in the single-frame image I; and acquiring the droplet size R _S of the fuel nozzle according to the effective droplet area S and the calibration parameter M. According to the invention, the fog drop size and the atomization angle of the fuel nozzle are simultaneously analyzed by adopting a vision-based method, and the automatic measurement of the fog drop size and the atomization angle can be completed by constructing a detection environment once without independently implementing two detection steps or equipping two sets of detection equipment, so that the whole flow of nozzle detection is simplified, and the working efficiency of nozzle detection is improved.

Description

Method and device for automatically detecting engine fuel nozzle injection information

Technical Field

The invention relates to the application of image processing technology in the field of fuel nozzle atomization detection, and in particular to an automatic detection method and device for engine fuel nozzle injection information.

Background technique

The fuel nozzle assembly (hereinafter referred to as the nozzle) is one of the important components of the combustion chamber in the engine. The nozzle can atomize the fuel into a large number of small particle droplets, which can effectively increase the contact area between the fuel and the air, thereby achieving the purpose of saving fuel consumption and improving combustion efficiency.

Nozzle atomization quality is an important parameter to characterize nozzle performance and also determines the quality effect of the actual process. Therefore, droplet size and atomization angle are two key injection information indicators of fuel nozzle performance.

At present, the automated detection method for droplet size detection is mainly based on the detection method of the laser particle size analyzer. For example, the Chinese invention patents with application numbers CN201410426094.9, CN202010041110.8 and CN202010534013.2 disclose methods based on this idea. The detection results of this method are accurate, stable and reliable. Among them, the Chinese patent "A spray characteristic parameter detection device and method", application number CN201410426094.9, authorization announcement number CN104181083B, discloses a spray characteristic parameter detection method, which uses a high-speed camera image acquisition system to detect the spray profile modeling spray angle, spray macroscopic two-dimensional distribution, and spray particle flight speed, and can adjust the relative position of the nozzle and the laser particle size analyzer according to the spray profile and macroscopic two-dimensional distribution and the designed path, and measure the spray particle size on the basis of the above; the Chinese patent "Nozzle spray particle size detection device", application number CN202010041110.8, authorization announcement number CN111157410A, discloses a nozzle spray particle size detection device, which measures the size of droplets sprayed by the nozzle to be tested by placing the transmitting end and the receiving end of the laser particle size analyzer on both sides of the spray area of the nozzle to be tested. The Chinese patent "A two-dimensional spray field particle size distribution detection device", application number CN202010534013.2, authorization announcement number CN111678847B, discloses a two-dimensional spray field particle size distribution detection device, which measures the size distribution of all droplets in the spray field by installing the generating end and the laser receiving end of the laser particle size analyzer on two supporting platforms respectively, and rotating the laser particle size analyzer.

For the nozzle atomization angle, the commonly used automatic measurement methods mainly include probe-based detection methods and visual detection-based methods. Among them, the probe-based method, such as the Chinese patent "A fuel nozzle atomization angle automatic measurement mechanism and automatic measurement method", with application number CN202110983437.1 and authorization announcement number CN113720277B, discloses a fuel nozzle atomization angle automatic measurement mechanism and automatic measurement method. The patent arranges the probe on the nozzle spray path, and controls the probe movement through PLC and other programs until it just reaches the spray edge, so that the nozzle atomization angle can be accurately detected; based on the visual detection method, such as the Chinese invention patents with application numbers CN201910146275.9 and CN202110666998.9, this method is adopted. This method has the advantages of being simple and fast, and at the same time, it has strong adaptability to different nozzles to be tested, and is easier to batch detect and analyze. Among them, the Chinese patent " A vision-based automatic detection system and method for nozzle atomization angle", application number CN201910146275.9, authorization announcement number CN109816678B, discloses a vision-based automatic detection method for nozzle atomization angle, which obtains image data of nozzle atomization fluid injection through a camera, and obtains the nozzle atomization angle after computer analysis of these image data; Chinese patent "A nozzle atomization angle detection operation device and method", application number CN202110666998.9, authorization announcement number CN113483697B, discloses a nozzle atomization angle detection method, which uses a second shooting structure to complete the image of the atomized fluid sprayed from the nozzle to be tested, and after analysis by the console, the nozzle atomization angle data is obtained according to the image of the atomized fluid.

The inventors of the present application discovered in the process of realizing the present invention that the above-mentioned scheme of the prior art has the following defects: first, as a special detection equipment, the laser particle size analyzer is large in size, high in cost, and its deployment, use and maintenance processes are relatively complicated; second, the probe-based measurement method is limited by the detection accuracy of the probe, and the detection environment is complex, the detection cycle is long, and the implementation cost is high; finally, in the existing fuel nozzle detection, the detection process of droplet size and atomization angle is cumbersome and requires many types of instruments.

Summary of the invention

The purpose of an embodiment of the present invention is to provide a method that uses a vision-based method to simultaneously analyze the droplet size and atomization angle of a fuel nozzle. By building a detection environment once, the automatic measurement of the droplet size and the atomization angle can be completed without the need to implement two separate detection steps or equip two sets of detection equipment, thereby simplifying the entire process of nozzle detection and improving the work efficiency of nozzle detection.

In order to achieve the above-mentioned object, an embodiment of the present invention provides a method for automatically detecting injection information of an engine fuel nozzle, characterized in that the method comprises: calibrating a camera and obtaining calibration parameters of the camera. ; Using the calibrated camera, obtain a single-frame image I in a fuel nozzle working test; according to the single-frame image I, determine the effective droplet area in the single-frame image I/> ; and according to the effective droplet area/> and the calibration parameters/> , get the droplet size of the fuel nozzle/> .

Optionally, the camera calibration parameters are obtained. The steps include: using the camera to completely photograph a checkerboard test card placed under the fuel nozzle to obtain a calibration image; and according to the calibration image, using a calibration method: Calculate the calibration parameters/> , where /> 、/> Respectively represent the horizontal and vertical focal lengths of the camera, /> 、/> are the offsets of the camera imaging center, /> Indicates the rotation of the camera in space coordinates, /> Indicates the offset of the camera in space coordinates.

Optionally, the step of obtaining a single-frame image I in a fuel nozzle operation inspection includes: in the inspection, using a calibrated camera to capture multiple frames of images of the fuel nozzle operation; saving the multiple frames of images as a universal format file; reading each frame of the multiple frames of images; and parsing each frame of the image to obtain each single-frame image I in the multiple frames of images.

Optionally, each single-frame image I is a single-channel grayscale image.

Optionally, the effective droplet area in the single frame image I is determined according to the single frame image I. The steps include: obtaining an image B after the single frame image I is filtered out of the foreground according to a fixed threshold binarization method, wherein the foreground is the fog droplets in the image I; obtaining an image C after the single frame image I is filtered out of the background according to an adaptive threshold binarization method; and performing image morphology: /> , process and update image C, where, /> ; Detect the fog droplets in the image C, and then obtain the initial fog droplet area/> ; and according to the image B and the initial droplet area/> , analyze and filter the invalid fog droplet area, and then determine the valid fog droplet area in the single frame image I/> .

Optionally, the effective droplet area and calibration parameters/> , get the droplet size of the fuel nozzle/> The droplet size is obtained according to the following equation/> :/> , where /> Represents the effective droplet area in the single frame image I/> The sequence number in the effective droplet set represented is/> of mist droplets, The serial number is counted as/> The average pixel width of the droplet area, /> Represents the serial number as/> The size of the droplets , symbol/> Represents the multiplication operation of a matrix and a vector.

Optionally, the method for automatically detecting nozzle injection information further includes: determining the outline of the fuel nozzle injection in the single frame image I according to the single frame image I. ; and according to the profile/> , determine the atomization angle of the fuel nozzle .

Optionally, the atomization angle of the fuel nozzle is determined The steps include: obtaining the image after the background of the single frame image I is removed according to the maximum between-class variance (OTSU) method/> ; According to the image/> , using the Suzuki85 contour detection algorithm, determine the contour of the fuel nozzle injection/> ; and according to the profile/> , by using the profile angle calculation method, the profile angle of the fuel nozzle injection is obtained, and then the atomization angle of the fuel nozzle is determined/> , wherein the steps of the contour angle calculation method include: obtaining the contour / The boundary point set V is randomly grouped into V ₁ , V ₂ ...V _n ; n straight lines corresponding to the random grouping are obtained by straight line fitting; two straight lines representing the boundary are determined; and the angle between the two straight lines representing the boundary is calculated, which is the atomization angle of the fuel nozzle. .

Optionally, the method for automatically detecting nozzle injection information further includes: obtaining the droplet size in each single frame image I of the multiple frame images in a detection of the fuel nozzle operation. and atomization angle/> ; and synthesizing and outputting the multiple droplet sizes in the one detection/> and atomization angle/> .

In addition, an embodiment of the present invention further provides an automatic detection device for engine fuel nozzle injection information, the device comprising: a memory; and a processor, configured to execute the automatic detection method for nozzle injection information.

Through the above technical solution, the embodiment of the present invention uses a calibrated camera to shoot the spray area of the fuel nozzle when it is working, obtains a single-frame instantaneous clear image of the spray form, and then obtains the detection data of the atomization angle through image processing algorithm analysis and measurement. At the same time, the camera calibration parameters and image processing algorithm analysis and measurement are used to obtain the detection data of the droplet size, thereby simplifying the steps of fuel nozzle detection, improving the efficiency of detection, and providing technical support for large-scale nozzle detection and testing in a short period of time. At the same time, the detection implementation process does not involve special detection instruments such as laser particle size analyzers, saving the detection cost of fuel nozzles, thereby improving the research and development and production efficiency of fuel nozzles.

Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

FIG1 is a flow chart of a method for automatically detecting injection information of an engine fuel nozzle provided by an embodiment of the present invention;

FIG. 2 is a diagram of obtaining the calibration parameters of the camera provided by an embodiment of the present invention. Flow chart of the steps;

FIG3 is a flowchart of the steps of obtaining a single frame image I in a fuel nozzle operation detection according to an embodiment of the present invention;

FIG. 4 is a diagram of determining the effective droplet area in a single frame image I provided by an embodiment of the present invention. Flow chart of the steps;

FIG. 5 is a diagram of a method for determining the droplet size of a fuel nozzle according to an embodiment of the present invention. Flow chart of the steps;

FIG. 6 is a diagram of a method for determining the atomization angle of the fuel nozzle provided by an embodiment of the present invention. Flow chart of the steps;

7 is a flowchart of the steps of the contour angle calculation method provided by an embodiment of the present invention;

8 is a flow chart of the detection data integration step in the method for automatically detecting the injection information of the engine fuel nozzle provided by an embodiment of the present invention;

9 is a structural diagram of an automatic detection device for engine fuel nozzle injection information provided by an embodiment of the present invention;

FIG. 10 is a module diagram of an automatic detection device for engine fuel nozzle injection information provided by an embodiment of the present invention.

Detailed ways

The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

Referring to FIG. 1 , an embodiment of the present invention discloses a method for automatically detecting injection information of an engine fuel nozzle, the method comprising:

Step S100: calibrate the camera and obtain the calibration parameters of the camera ;

In some embodiments, the embodiments of the present invention perform a camera calibration operation and obtain the camera calibration parameters when the camera has not been calibrated or there are no camera calibration parameters. As shown in FIG. 2 , an embodiment of the present invention provides a method for obtaining calibration parameters of the camera. The process includes:

Step S110, using the camera to completely photograph a checkerboard test card placed below the fuel nozzle to obtain a calibration image;

Step S120, according to the calibration image, by a calibration method: Calculate the calibration parameters/> , where /> 、/> Respectively represent the horizontal and vertical focal lengths of the camera, /> 、/> are the offsets of the camera imaging center, /> Indicates the rotation of the camera in space coordinates, /> Indicates the offset of the camera in the space coordinate. Optionally, the calibration method is the Zhang Zhengyou calibration method preferred in the embodiment of the present invention.

Step S200, using the calibrated camera to obtain a single frame image I of a fuel nozzle operation detection;

In some embodiments, as shown in FIG. 3 , an embodiment of the present invention provides a process flow for obtaining a single frame image I in a fuel nozzle operation detection, the process comprising:

Step S210, in the one detection, using a calibrated camera to capture multiple frames of images of the fuel nozzle in operation. Optionally, in the embodiment of the present invention, a calibrated high-speed industrial camera is preferably used to complete the shooting. The high-speed industrial camera can capture more than 25 multiple frames of images per second in one detection, and can capture many instantaneous information that is invisible to the human body, and can clearly capture a clear image of the spray instant in the spray area when the fuel nozzle is working;

Step S220, saving the multiple frames of images as a universal format file. Optionally, converting the specific image format of the high-speed industrial camera into a universal image format and saving it for subsequent processing. The universal format for saving the multiple frames of image data includes but is not limited to a tiff image file or an avi video file.

Step S230, reading each frame of the multiple frames of images;

Step S240, parsing each frame of the image to obtain each single frame of the multi-frame image I. In some embodiments, the embodiment of the present invention obtains the original single frame of the image by parsing the read image data of each frame into a pixel space. Optionally, each single-frame image I is a single-channel grayscale image. If the analyzed single-frame image is If the image is a color image, grayscale conversion of the image is performed. In the embodiment of the present invention, the grayscale conversion of the image is preferably performed using the following formula:

in, are the red, green and blue channels of the color image respectively,/> is the converted single-channel grayscale image.

Step S310, determining the effective droplet area in the single frame image I according to the single frame image I ;

In some embodiments, as shown in FIG. 4 , an embodiment of the present invention provides a method for determining a valid droplet area in a single frame image I. The process includes:

Step S311, according to the fixed threshold binarization method, obtain the image B after the single frame image I is filtered out of the foreground, wherein the foreground is the fog droplets in the image I; Optionally, the step of obtaining the image B after the single frame image I is filtered out of the foreground according to the fixed threshold binarization method provided by the embodiment of the present invention includes: Perform Gaussian blur processing: /> , get the image/> , where the image /> is the converted single-channel grayscale image/> ,/> represents convolution, /> The preferred size of the Gaussian kernel in the embodiment of the present invention is 5×5. The Gaussian kernel may be other sizes of N×N, where N is an odd number. The image is then filtered using a preset threshold. A binary image B is obtained by performing a binarization process, wherein the size of the preset threshold is preferably 30% of the maximum brightness value in the single-frame image I.

Step S312, according to the adaptive threshold binarization method, obtain the image C after the single frame image I is filtered out of the background; Optionally, the step of obtaining the image C after the single frame image I is filtered out of the background according to the adaptive threshold binarization method provided by the embodiment of the present invention includes: using a Gaussian adaptive threshold to binarize the image Perform binarization processing to obtain a binary image C, wherein the image is the converted single-channel grayscale image/> .

Step S313, through image morphology: , process and update image C, where ; Optionally, the image morphology provided by the embodiment of the present invention:/> The step of processing and updating the image C includes: performing two operations of dilation and erosion on the image C in sequence, wherein: For corrosion operation,/> For expansion operation.

Step S314, detecting the fog droplets in the image C, and then obtaining the initial fog droplet area ; Optionally, the embodiment of the present invention provides/> To implement the fog droplet detection operation in the image C that has been processed by image morphology, and obtain the initial fog droplet area/> ;

Step S315: based on the image B and the initial droplet area , analyze and filter the invalid fog droplet area, and then determine the valid fog droplet area in the single frame image I/> ; Optionally, the embodiment of the present invention uses the image B after the fog droplets are screened out and the initial fog droplet area/> , through/> , filter the initial droplet area/> The invalid fog droplet area in the single frame image I is determined, and then the valid fog droplet area in the single frame image I is determined. .

Step S410, according to the effective droplet area and the calibration parameters/> , get the droplet size of the fuel nozzle/> ; In some embodiments, the present invention provides a method according to the effective droplet area / and calibration parameters Get the droplet size of the fuel nozzle/> The method mainly obtains the droplet size according to the following equation/> : To achieve this, refer to FIG5 , the specific steps of the method include:

Step S411, by , count the effective droplet area/> The serial number is/> The average pixel width of the droplet area; where, /> Represents the effective droplet area in the single frame image I/> The sequence number in the effective droplet set represented is/> of droplets, algorithm/> It is used to count the average pixel width of an area in an image, using the statistical sequence number as/> The number of pixels in the effective droplet area is used to represent the sequence number / > The average width of the droplets.

Step S412, by , get the serial number as/> The actual droplet size of the droplets/> ; Among them, /> Represents the serial number as/> The actual droplet size of the droplets/> , symbol/> Represents the multiplication operation of a matrix and a vector, by representing the single frame image I in which the sequence number is / > The vector of the effective droplet area size is multiplied by the calibration parameter M matrix to obtain the sequence number / > The true size of the droplets/> .

Step S320, based on the single frame image I, determining the outline of the fuel nozzle injection in the single frame image I ;

Step S420, according to the outline , determine the atomization angle of the fuel nozzle/> ;

In some embodiments, as shown in FIG. 6 , an embodiment of the present invention provides a method for determining a fogging angle according to a single frame image I. A method comprising:

Step S321, according to the maximum between-class variance (OTSU) method, obtain the image after the single frame image I is filtered out of the background ;Optional, for the image/> Perform Gaussian blur processing: /> , get the image/> , where/> is the converted single-channel grayscale image/> ,/> The preferred Gaussian kernel size of the embodiment of the present invention is 11×11. The Gaussian kernel size can be other N×N, where N is an odd number. The maximum between-class variance (OTSU) algorithm is then used to classify the image. Perform binarization processing to obtain the image of the single frame image I after removing the background/> .

Step S322: based on the image , using the Suzuki85 contour detection algorithm, determine the contour of the fuel nozzle injection/> ; Optionally, the embodiment of the present invention preferably uses the Suzuki85 contour detection algorithm:/> For the image/> Performing a profile detection operation to determine the profile of the fuel nozzle injection .

Step S421, according to the outline , by using the profile angle calculation method, the profile angle of the fuel nozzle injection is obtained, and then the atomization angle of the fuel nozzle is determined/> .

Optionally, an embodiment of the present invention provides a method for calculating a contour angle: , the steps of the contour angle calculation method are shown in Figure 7:

Step S422, obtaining the contour The random grouping V ₁ , V ₂ ...V _n of the boundary point set V; optionally, n represents the number of groups, and the value of n is: Let the number of points in the boundary point set V be m, then/> .

Step S423, obtaining n straight lines corresponding to the random grouping by straight line fitting; wherein the n straight lines F ₁ , F ₂ . . . F _n correspond to the random groups V ₁ , V _{2 .} . . V _n , respectively.

Step S424, determine two straight lines representing the boundary; optionally, for each of the n straight lines F ₁ , F ₂ ..F _n , respectively calculate and accumulate the distances to all points in the boundary point set V to obtain comprehensive distances S ₁ , S ₂ ..S _n corresponding to F ₁ , F ₂ ..F _n , respectively, find the two smallest comprehensive distances _Sk , S _l among the comprehensive distances S ₁ , S ₂ ..S _n , and the corresponding two straight lines F _k , F _l are the straight lines representing the boundary.

Step S425, calculating the angle between the two straight lines representing the boundary, which is the atomization angle of the fuel nozzle. Calculate the angle between the two straight lines F _k and F _l representing the boundary, that is, obtain the profile angle of the fuel nozzle injection, that is, the atomization angle of the fuel nozzle/> .

In some embodiments, as shown in FIG. 8 , the method for automatically detecting the injection information of the fuel nozzle of the engine provided by the embodiment of the present invention further includes a detection data integration step process, which includes:

Step S510, obtaining the droplet size in each single frame image I of the multiple frames of images in a detection of the fuel nozzle operation and atomization angle/> ; In each single frame image I, multiple droplet sizes can be detected / > The measurement data of a multi-frame image sequence of a detection test experiment can detect the fog angle/> Multiple measurement data.

Step S520, synthesizing and outputting the multiple droplet sizes in the one detection and atomization angle/> ; Optionally, the detection data integration module provided in the embodiment of the present invention can process the droplet size obtained from a detection experiment/> and atomization angle/> Data on droplet size /> and atomization angle/> The maximum, minimum, mean and variance of the data are calculated, the corresponding statistical information is calculated and the raw data is exported in the form of a table.

As shown in FIG9 , an embodiment of the present invention further provides an automatic detection device for fuel nozzle injection information of an engine, the device comprising: a memory and a processor, wherein the processor is configured to implement the following steps when executing a program: calibrating a camera and obtaining calibration parameters of the camera; ; Using the calibrated camera, obtain a single-frame image I in a fuel nozzle working test; according to the single-frame image I, determine the effective droplet area in the single-frame image I/> ; and according to the effective droplet area/> and the calibration parameters/> , get the droplet size of the fuel nozzle/> .

Optionally, the camera calibration parameters are obtained. The steps include: using the camera to completely photograph a checkerboard test card placed under the fuel nozzle to obtain a calibration image; and according to the calibration image, using a calibration method: Calculate the calibration parameters/> , where /> 、/> Respectively represent the horizontal and vertical focal lengths of the camera, /> 、/> are the offsets of the camera imaging center, /> Indicates the rotation of the camera in space coordinates, /> Indicates the offset of the camera in space coordinates.

Optionally, the step of obtaining a single-frame image I in a fuel nozzle operation inspection includes: in the inspection, using a calibrated camera to capture multiple frames of images of the fuel nozzle operation; saving the multiple frames of images as a universal format file; reading each frame of the multiple frames of images; and parsing each frame of the image to obtain each single-frame image I in the multiple frames of images.

Optionally, each single-frame image I is a single-channel grayscale image.

Optionally, the effective droplet area in the single frame image I is determined according to the single frame image I. The steps include: obtaining an image B after the single frame image I is filtered out of the foreground according to a fixed threshold binarization method, wherein the foreground is the fog droplets in the image I; obtaining an image C after the single frame image I is filtered out of the background according to an adaptive threshold binarization method; and performing image morphology: /> , process and update image C, where, /> ; Detect the fog droplets in the image C, and then obtain the initial fog droplet area/> ; and according to the image B and the initial droplet area/> , analyze and filter the invalid fog droplet area, and then determine the valid fog droplet area in the single frame image I/> .

Optionally, the effective droplet area and calibration parameters/> , get the droplet size of the fuel nozzle/> The droplet size is obtained according to the following equation/> :/> , where /> Represents the effective droplet area in the single frame image I/> The sequence number in the effective droplet set represented is/> of mist droplets, The serial number is counted as/> The average pixel width of the droplet area, /> Represents the serial number as/> The size of the droplets , symbol/> Represents the multiplication operation of a matrix and a vector.

Optionally, the method for automatically detecting nozzle injection information further includes: determining the outline of the fuel nozzle injection in the single-frame image I according to the single-frame image I; and determining the atomization angle of the fuel nozzle according to the outline. .

Optionally, the atomization angle of the fuel nozzle is determined The steps include: obtaining the image after the background of the single frame image I is removed according to the maximum between-class variance (OTSU) method/> ; According to the image/> , using the Suzuki85 contour detection algorithm to determine the contour of the fuel nozzle injection; and according to the contour, using the contour angle calculation method to obtain the contour angle of the fuel nozzle injection, and then determine the atomization angle of the fuel nozzle/> , wherein the steps of the contour angle calculation method include: obtaining the contour / The boundary point set V is randomly grouped into V ₁ , V ₂ ...V _n ; n straight lines corresponding to the random grouping are obtained by straight line fitting; two straight lines representing the boundary are determined; and the angle between the two straight lines representing the boundary is calculated, which is the atomization angle of the fuel nozzle. .

Optionally, the method for automatically detecting the nozzle spray information further includes: the droplet size in each single frame image I of the multiple frame images in one detection of the fuel nozzle operation and atomization angle/> ; and synthesizing and outputting the multiple droplet sizes in the one detection/> and atomization angle/> .

FIG10 is a module diagram of an automatic detection device for the injection information of an engine fuel nozzle provided by an embodiment of the present invention. Referring to FIG10 , the automatic detection device for the injection information of an engine fuel nozzle provided by an embodiment of the present invention includes 6 processing modules, namely: a camera calibration module, an image acquisition module, an image analysis module, a droplet size detection module, an atomization angle detection module, and a detection data integration module. Among them, the camera calibration module uses a checkerboard test card to calibrate a high-speed industrial camera and obtain camera calibration parameters; the image acquisition module is responsible for receiving and saving multiple frames of image data of the working screen of the fuel nozzle taken by the camera; the image analysis module is responsible for reading each frame of image data stored in the image acquisition module, and analyzing to obtain the original single-frame images. ; The droplet size detection module is responsible for detecting the droplet size by taking a single frame image of the fuel nozzle working/> Analyze and determine the single frame image/> The effective droplet area in/> , combined with the camera calibration parameters/> , and finally get the droplet size/> The atomization angle detection module is responsible for analyzing the working picture of the fuel nozzle to obtain the atomization angle / > The detection data integration module processes a detection experiment to obtain the data of droplet size and atomization angle, calculates the corresponding statistical information and exports the original data in the form of a table. Optionally, the droplet size detection module and the atomization angle detection module can be performed simultaneously.

The devices in this article can be servers, PCs, PADs, mobile phones, etc.

The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program that initializes the following method steps: calibrating a camera and obtaining calibration parameters of the camera. ; Obtain a single-frame image I in a fuel nozzle operation detection; Determine the effective droplet area in the single-frame image I according to the single-frame image I/> ; and according to the effective droplet area/> and calibration parameters/> , get the droplet size of the fuel nozzle/> .

Optionally, the camera calibration parameters are obtained. The steps include: using the camera to completely photograph a checkerboard test card placed under the fuel nozzle to obtain a calibration image; and according to the calibration image, using a calibration method: Calculate the calibration parameters/> , where /> 、/> Respectively represent the horizontal and vertical focal lengths of the camera, /> 、/> are the offsets of the camera imaging center, /> Indicates the rotation of the camera in space coordinates, /> Indicates the offset of the camera in space coordinates.

Optionally, the step of obtaining a single-frame image I in a fuel nozzle operation inspection includes: in the inspection, using a calibrated camera to capture multiple frames of images of the fuel nozzle operation; saving the multiple frames of images as a universal format file; reading each frame of the multiple frames of images; and parsing each frame of the image to obtain each single-frame image I in the multiple frames of images.

Optionally, each single-frame image I is a single-channel grayscale image.

Optionally, the effective droplet area in the single frame image I is determined according to the single frame image I. The steps include: obtaining an image B after the single frame image I is filtered out of the foreground according to a fixed threshold binarization method, wherein the foreground is the fog droplets in the image I; obtaining an image C after the single frame image I is filtered out of the background according to an adaptive threshold binarization method; and performing image morphology: /> , process and update image C, where, /> ; Detect the fog droplets in the image C, and then obtain the initial fog droplet area/> ; and according to the image B and the initial droplet area/> , analyze and filter the invalid fog droplet area, and then determine the valid fog droplet area in the single frame image I/> .

Optionally, the effective droplet area and calibration parameters/> , get the droplet size of the fuel nozzle/> The droplet size is obtained according to the following equation/> :/> , where /> Represents the effective droplet area in the single frame image I/> The sequence number in the effective droplet set represented is/> of mist droplets, The serial number is counted as/> The average pixel width of the droplet area, /> Represents the serial number as/> The size of the droplets , symbol/> Represents the multiplication operation of a matrix and a vector.

Optionally, the method for automatically detecting nozzle injection information further includes: determining the outline of the fuel nozzle injection in the single-frame image I according to the single-frame image I; and determining the atomization angle of the fuel nozzle according to the outline. .

Optionally, the atomization angle of the fuel nozzle is determined The steps include: obtaining the image after the background of the single frame image I is removed according to the maximum between-class variance (OTSU) method/> ; According to the image/> , using the Suzuki85 contour detection algorithm to determine the contour of the fuel nozzle injection; and according to the contour, using the contour angle calculation method to obtain the contour angle of the fuel nozzle injection, and then determine the atomization angle of the fuel nozzle/> , wherein the steps of the contour angle calculation method include: obtaining the contour / The boundary point set V is randomly grouped into V ₁ , V ₂ ...V _n ; n straight lines corresponding to the random grouping are obtained by straight line fitting; two straight lines representing the boundary are determined; and the angle between the two straight lines representing the boundary is calculated, which is the atomization angle of the fuel nozzle. .

Optionally, the method for automatically detecting the nozzle spray information further includes: the droplet size in each single frame image I of the multiple frame images in one detection of the fuel nozzle operation and atomization angle/> ; and synthesizing and outputting the multiple droplet sizes in the one detection/> and atomization angle/> .

Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

The present application is described with reference to the flowchart and/or block diagram of the method, device (system), and computer program product according to the embodiment of the present application. It should be understood that each process and/or box in the flowchart and/or block diagram, as well as the combination of the process and/or box in the flowchart and/or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate a device for implementing the functions specified in one process or multiple processes in the flowchart and/or one box or multiple boxes in the block diagram.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and/or one or more boxes in the block diagram.

In a typical configuration, a computing device includes one or more processors (CPU), input/output interfaces, network interfaces, and memory.

The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and/or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims (8)

1. A method for automatically detecting injection information of an engine fuel nozzle, characterized in that the method comprises: Calibrate the camera and get the calibration parameters of the camera ; Using the calibrated camera, a single frame image I is acquired during a fuel nozzle operation inspection; According to the single frame image I, determine the effective droplet area in the single frame image I ;as well as According to the effective droplet area and the calibration parameters/> , get the droplet size of the fuel nozzle/> , Wherein, the effective droplet area in the single frame image I is determined according to the single frame image I The steps include: According to a fixed threshold binarization method, an image B is obtained after the single frame image I is screened out of the foreground, wherein the foreground is the fog droplets in the image I; According to the adaptive threshold binarization method, an image C is obtained after the background of the single frame image I is screened out; Through image morphology: , process and update image C, where, /> ,/> For corrosion operation,/> For expansion operation; Detect the fog droplets in the image C, and then obtain the initial fog droplet area ;as well as According to the image B and the initial droplet area , analyze and filter the invalid fog droplet area, and then determine the valid fog droplet area in the single frame image I/> ;as well as The effective droplet area and calibration parameters/> , get the droplet size of the fuel nozzle/> The droplet size is obtained according to the following equation/> :/> , where /> Represents the effective droplet area in the single frame image I/> The sequence number in the effective droplet set represented is/> of mist droplets, /> The serial number is counted as/> The average pixel width of the droplet area, /> Represents the serial number as/> The size of the droplets , symbol/> Represents the multiplication operation of a matrix and a vector. 2. The method for automatically detecting nozzle spray information according to claim 1, characterized in that the calibration parameters of the camera are obtained. The steps include: The camera completely photographs a checkerboard test card placed under the fuel nozzle to obtain a calibration image; and According to the calibration image, through the calibration method: Calculate the calibration parameters , where /> 、/> Respectively represent the horizontal and vertical focal lengths of the camera, /> 、/> are the offsets of the camera imaging center, /> Indicates the rotation of the camera in space coordinates, /> Indicates the offset of the camera in space coordinates. 3. The method for automatically detecting nozzle injection information according to claim 1, characterized in that the step of obtaining a single frame image I in a single detection of the fuel nozzle operation comprises: In the one detection, a calibrated camera is used to capture multiple frames of images of the fuel nozzle in operation; Saving the multiple frames of images as a file in a universal format; Read each frame of the multiple frames of images; and Each frame of the image is analyzed to obtain each single frame image I in the multiple frames of the image. 4 . The method for automatically detecting nozzle spray information according to claim 3 , wherein each single-frame image I is a single-channel grayscale image. 5. The method for automatically detecting nozzle spraying information according to claim 3, characterized in that the method further comprises: According to the single frame image I, the profile of the fuel nozzle injection in the single frame image I is determined ;as well as According to the outline , determine the atomization angle of the fuel nozzle/> . 6. The method for automatically detecting nozzle injection information according to claim 5, characterized in that the atomization angle of the fuel nozzle is determined The steps include: According to the maximum between-class variance (OTSU) method, the image after the single frame image I is filtered out of the background is obtained. ; According to the image , using the Suzuki85 contour detection algorithm, determine the contour of the fuel nozzle injection/> ;as well as According to the outline , through the contour angle calculation method, the contour angle of the fuel nozzle injection is obtained, that is, the atomization angle of the fuel nozzle/> , wherein the steps of the contour angle calculation method include: Get the outline A random grouping V ₁ , V ₂ , …V _n of the boundary point set V ; Obtain n straight lines corresponding to the random grouping by straight line fitting; Identify two straight lines representing the boundary; and Calculate the angle between the two straight lines representing the boundary, which is the atomization angle of the fuel nozzle. . 7. The method for automatically detecting nozzle spraying information according to claim 6, characterized in that the method further comprises: Obtain the droplet size in each single frame image I of the multiple frame images in a detection of the fuel nozzle operation and atomization angle/> ;as well as Integrate and output the multiple droplet sizes in the one detection and atomization angle/> . 8. An automatic detection device for engine fuel nozzle injection information, the device comprising: Memory; and A processor configured to execute the nozzle ejection information automatic detection method according to any one of claims 1 to 7.