Industrial oil pollution particle identification method and device, electronic equipment and medium

Abstract

The invention discloses an industrial oil pollution particle identification method and device, electronic equipment and a medium, and relates to the technical field of image processing. The method comprises the following steps: controlling an image acquisition board to perform image acquisition to obtain a to-be-recognized image with scale marks; preprocessing the to-be-recognized image to obtain a binary image; calculating a pixel distance between any two adjacent scale lines in the binary image; further obtaining a conversion proportion; performing image segmentation processing on the binarized image to obtain a plurality of independent single particle images; determining the light transmittance, the pixel size and the surface roughness of the particles; and determining a target type of the particle based on the light transmittance, the actual size and the surface roughness of the particle. Therefore, the light transmittance, the actual size and the surface roughness of the particles are obtained through the to-be-recognized image with the scale marks, so that the target type of the particles is determined, and the number of each type of particles in the to-be-recognized image can be obtained.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and particularly relates to a method, device, electronic device and medium for identifying industrial oil pollution particles. Background Art

[0002] During the use of industrial oil, due to reasons such as oil pollution, oil temperature and pressure, friction and wear of mechanical parts, and air mixing, particles will be generated in the oil. These particles will accelerate the oxidation and deterioration rate of the oil, accelerate the consumption of additives, affect heat dissipation, and cannot form a complete oil film, resulting in abnormal damage to the equipment. Therefore, it is very necessary to timely monitor and control the types of particles in the lubricating oil and the content of each type of particle. However, currently, the monitoring of industrial oil is usually carried out through a particle counter, which can only determine the total number of particles and cannot confirm the types of particles and the number of each type of particle. Summary of the Invention

[0003] This application aims to at least solve one of the technical problems existing in the prior art. For this reason, this application proposes a method, device, electronic device and medium for identifying industrial oil pollution particles, which can accurately detect the number of each type of particle in the image to be identified.

[0004] An identification device for industrial oil pollution particles according to an embodiment of the first aspect of this application, the identification device includes a transparent pipeline, a glass lens, a backlight assembly, a macro lens and an image acquisition board. The glass lens is provided with scale lines. The transparent pipeline is used for the oil to be detected to pass through. The backlight assembly is located below the transparent pipeline. The glass lens is arranged on the outer side wall of the bottom of the transparent pipeline. The macro lens is located below the glass lens. The image acquisition board is connected to the macro lens.

[0005] The method includes:

[0006] When it is detected that there is oil in the transparent pipeline, control the backlight assembly to emit light, and control the image acquisition board to perform image acquisition to obtain an image to be identified with scale lines.

[0007] Preprocess the image to be identified to obtain a binary image.

[0008] Calculate the pixel distance between any two adjacent scale lines in the binary image.

[0009] Based on the pixel distance and the actual distance between two scale lines, obtain a conversion ratio.

[0010] Perform image segmentation processing on the binary image to obtain a plurality of independent single-particle images; each single-particle image only includes one particle.

[0011] Based on the single-particle image and the image to be recognized, determine the light transmittance, pixel size, and surface roughness of the particle;

[0012] Based on the conversion ratio and the pixel size of the particle, calculate the actual size of the particle;

[0013] Based on the light transmittance, actual size, and surface roughness of the particle, determine the target type of the particle.

[0014] According to the method for identifying industrial oil pollution particles in the embodiments of the present application, it has at least the following beneficial effects: When the recognition method detects that there is oil in the transparent pipeline, it controls the backlight component to emit light and controls the image acquisition board to perform image acquisition to obtain an image to be recognized with scale lines; preprocess the image to be recognized to obtain a binary image; calculate the pixel distance between any two adjacent scale lines in the binary image; based on the pixel distance and the actual distance between the two scale lines, obtain the conversion ratio; perform image segmentation processing on the binary image to obtain multiple independent single-particle images; each single-particle image only includes one particle; based on the single-particle image and the image to be recognized, determine the light transmittance, pixel size, and surface roughness of the particle; based on the conversion ratio and the pixel size of the particle, calculate the actual size of the particle; based on the light transmittance, actual size, and surface roughness of the particle, determine the target type of the particle. In this way, by using the image to be recognized with scale lines, the light transmittance, actual size, and surface roughness of the particle are obtained, so as to determine the target type of the particle, and the number of each type of particle in the image to be recognized can be obtained. It can accurately detect the number of each type of particle in the image to be recognized.

[0015] According to some embodiments of the first aspect of the present application, the determining the target type of the particle based on the light transmittance, actual size, and surface roughness of the particle includes:

[0016] Based on the light transmittance, actual size, and surface roughness of the particle, construct a particle information array;

[0017] Input the particle information array into the trained particle recognition model to obtain the target type.

[0018] According to some embodiments of the first aspect of the present application, the training steps of the particle recognition model include:

[0019] Obtain a variety of training samples and the corresponding true classifications of the training samples; wherein, the true classifications include cutting abrasive particles, fatigue particles, sliding wear particles, and fiber particles; the variety of training samples include the training information array of cutting abrasive particles, the training information array of fatigue particles, the training information array of sliding wear particles, and the training information array of fiber particles; the training information array is constructed from the light transmittance, actual size, and surface roughness of the particle;

[0020] Input the training samples into an initial particle recognition model to obtain the training classification of the training samples.

[0021] Calculate a loss value based on the training classification and the true classification of the training samples.

[0022] Iteratively optimize the particle recognition model based on the loss value to obtain a trained particle recognition model.

[0023] According to some embodiments of the first aspect of the present application, the preprocessing of the image to be recognized to obtain a binary image includes:

[0024] Perform grayscale transformation on the image to be recognized to obtain a grayscale image.

[0025] Perform binary processing on the grayscale image to obtain the binary image.

[0026] According to some embodiments of the first aspect of the present application, the steps for determining the light transmittance of particles include:

[0027] Calculate the first average grayscale value of the non-particle region in the grayscale image and calculate the second average grayscale value of the single-particle image.

[0028] Calculate the light transmittance based on the first average grayscale value and the second average grayscale value.

[0029] According to some embodiments of the first aspect of the present application, the steps for determining the pixel size of particles include:

[0030] Based on the single-particle image, calculate the minimum bounding rectangle of the particle.

[0031] Take the length and width of the minimum bounding rectangle as the pixel size.

[0032] According to some embodiments of the first aspect of the present application, the steps for determining the surface roughness of particles include:

[0033] Perform noise removal processing on the single-particle image to obtain a denoised image.

[0034] Perform sharpening processing on the denoised image to obtain a sharpened image.

[0035] Convert the sharpened image into grayscale contour data, and calculate the approximate contour arithmetic mean deviation of the grayscale contour data, and the approximate contour arithmetic mean deviation is used as the surface roughness of the particle.

[0036] The second aspect of the embodiments of the present application provides a device for identifying industrial oil pollution particles, which is applied to an equipment for identifying industrial oil pollution particles. The identification equipment includes a transparent pipeline, a glass lens, a backlight assembly, a macro lens and an image acquisition board. The glass lens is provided with scale lines. The transparent pipeline is used for the oil to be detected to pass through. The backlight assembly is located below the transparent pipeline. The glass lens is arranged on the outer side wall of the bottom of the transparent pipeline. The macro lens is located below the glass lens. The image acquisition board is connected to the macro lens;

[0037] The device includes:

[0038] An image acquisition module, configured to control the backlight assembly to emit light and control the image acquisition board to perform image acquisition when it is detected that there is oil in the transparent pipeline, so as to obtain an image to be identified with scale lines;

[0039] A preprocessing module, configured to preprocess the image to be identified to obtain a binary image;

[0040] A first calculation module, configured to calculate the pixel distance between any two adjacent scale lines in the binary image;

[0041] A second calculation module, configured to obtain a conversion ratio based on the pixel distance and the actual distance between two scale lines;

[0042] An image segmentation module, configured to perform image segmentation processing on the binary image to obtain a plurality of independent single-particle images; each single-particle image only includes one particle;

[0043] A first determination module, configured to determine the light transmittance, pixel size and surface roughness of the particle based on the single-particle image and the image to be identified;

[0044] A third calculation module, configured to calculate the actual size of the particle based on the conversion ratio and the pixel size of the particle;

[0045] A second determination module, configured to determine the target type of the particle based on the light transmittance, actual size and surface roughness of the particle.

[0046] The third aspect of the embodiments of the present application provides an electronic device. The electronic device includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, it implements the method for identifying industrial oil pollution particles according to any one of the first aspect of the embodiments.

[0047] In a fourth aspect embodiment of the present application, a computer-readable storage medium is provided. The storage medium stores a computer program, and is characterized in that when the computer program is executed by a processor, it implements the method for identifying industrial oil contamination particles according to any one of the embodiments of the first aspect.

[0048] Additional aspects and advantages of the present application will be given in part in the following description, become apparent in part from the following description, or be understood through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The following further describes the present application with reference to the drawings and embodiments, where:

[0050] Figure 1 is a schematic structural diagram of an identification device for industrial oil contamination particles according to an embodiment of the present application;

[0051] Figure 2 is a schematic flowchart of the steps of a method for identifying industrial oil contamination particles according to an embodiment of the present application;

[0052] Figure 3 is a schematic structural diagram of an identification device for industrial oil contamination particles according to an embodiment of the present application;

[0053] Figure 4 is a schematic structural diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] The embodiments of the present application are described in detail below. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions throughout. The embodiments described below with reference to the drawings are exemplary only for explaining the present application and should not be construed as limiting the present application.

[0055] In the description of the present application, it should be understood that the orientation descriptions, such as up, down, front, back, left, right, etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. This is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present application.

[0056] In the description of the present application, the meaning of several is more than one, the meaning of multiple is more than two, greater than, less than, exceeding, etc. are understood as not including the number itself, and above, below, within, etc. are understood as including the number itself. If there is a description of first and second, it is only for the purpose of distinguishing technical features and should not be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features or the sequence relationship of the indicated technical features.

[0057] In the description of this application, unless otherwise clearly defined, terms such as "setting", "installation", "connection", etc. should be understood in a broad sense, and those skilled in the art can reasonably determine the specific meanings of the above terms in this application in combination with the specific content of the technical solution.

[0058] In the description of this application, the description with reference to terms such as "one embodiment", "some embodiments", "schematic embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of this application. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0059] First, an identification device for industrial oil contamination particles in the embodiments of this application will be introduced. Refer to Figure 1 , Figure 1 is a schematic structural diagram of the identification device for industrial oil contamination particles in the embodiments of this application. The identification device includes a transparent pipeline, a glass lens, a backlight assembly, a macro lens, and an image acquisition board. The glass lens is provided with scale lines. The transparent pipeline is used for the oil to be detected to pass through. The backlight assembly is located below the transparent pipeline. The glass lens is arranged on the outer side wall of the bottom of the transparent pipeline; the macro lens is located below the glass lens, and the image acquisition board is connected to the macro lens. The backlight assembly is located directly above the glass lens, and the backlight assembly is used to emit light and send uniform light to the transparent pipeline. The macro lens is located directly below the glass lens, and the image acquisition board takes pictures of the glass lens and the transparent pipeline through the macro lens, so as to obtain an image of the oil in the transparent pipeline. Since the glass lens is provided with scale lines, the image of the oil also has scale lines, so as to obtain an image to be identified with scale lines.

[0060] In some embodiments, the identification device for industrial oil contamination particles further includes a one-way valve (not shown in the figure). One end of the one-way valve is connected to the oil passage of the industrial equipment, and the other end is connected to the transparent pipeline. Through the one-way valve, the industrial oil in the industrial equipment is transported to the transparent pipeline. It should be noted that the pressure of the one-way valve is small, so that the oil can flow slowly in the transparent pipeline, which is convenient for the image acquisition board to take pictures through the macro lens.

[0061] Based on Figure 1 , an identification method for industrial oil contamination particles in the first aspect embodiments of this application is proposed. The identification method for industrial oil contamination particles in the embodiments of this application is applied to Figure 1Schematic recognition device. The recognition method for industrial oil contamination particles can be executed on a terminal, on a server side, or be software running on a terminal or server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the recognition method for industrial oil contamination particles, etc., but is not limited to the above forms.

[0062] This application can be used in numerous general-purpose or special-purpose computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet-type devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and so on. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0063] Referring to Figure 2 , Figure 2 is a schematic flowchart of the steps of the recognition method for industrial oil contamination particles according to an embodiment of this application. The recognition method for industrial oil contamination particles includes but is not limited to steps S210 to S280.

[0064] Step S210, when it is detected that there is oil in the transparent pipeline, control the backlight component to emit light and control the image acquisition board to perform image acquisition to obtain a to-be-recognized image with scale lines;

[0065] Specifically, emitting uniform light to the transparent pipeline through the backlight component can make the acquired to-be-recognized image clearer and help improve the recognition accuracy in the subsequent process.

[0066] In some embodiments, the image acquisition board uses a CMOS image acquisition board.

[0067] In some embodiments, the recognition device further includes a light shield, which is installed on the outer sidewall of the bottom of the transparent pipeline, and both the glass lens and the macro lens are located inside the light shield.

[0068] In one embodiment, the recognition device further includes a frosted glass lens, which is provided on the outer sidewall of the top of the transparent pipeline, and the frosted glass lens is located directly below the backlight assembly.

[0069] Step S220: Preprocess the image to be recognized to obtain a binary image.

[0070] Step S230: Calculate the pixel distance between any two adjacent scale lines in the binary image.

[0071] Step S240: Obtain a conversion ratio based on the pixel distance and the actual distance between the two scale lines.

[0072] In step S220 of some embodiments, the Hough transform algorithm is used to detect the straight lines in the binary image. The straight lines in the binary image are scale lines, so that adjacent scale lines can be determined, and the vertical distance between two adjacent scale lines is calculated as the pixel distance. Thus, based on the pixel distance and the actual distance between the two scale lines, the conversion ratio is obtained. Exemplarily, if the pixel distance is X and the actual distance between the two scale lines is Y millimeters, then the conversion ratio A = Y / X.

[0073] It should be noted that the Hough Transform algorithm is an important feature extraction technology in the field of image processing. It can be used to detect straight lines in a binary image. The core idea of the Hough transform is to transform the edge points in the image into the parameter space and use the voting mechanism in the parameter space to detect the geometric shapes in the image.

[0074] Step S250: Perform image segmentation on the binary image to obtain multiple independent single-particle images; each single-particle image includes only one particle.

[0075] In some embodiments, step S250 may include steps S251 to S254.

[0076] Step S251: Invert the binary image to obtain an inverted image.

[0077] Step S252: Perform dilation processing on the inverted image to obtain a dilated image.

[0078] In some embodiments, the dilation processing is a 2D magnification technique that can gather scattered pixels together to form denser pixel clusters.

[0079] Step S253: Detect the objects in the dilated image, identify the pixel clusters with holes as bubbles, and count the pixel clusters with holes to obtain the number of bubbles.

[0080] Specifically, in the dilated image, the pixel clusters with holes are bubbles, while the pixel clusters without holes are particles.

[0081] Step S254: Remove the regions corresponding to the bubbles in the binary image, segment the particles in the binary image to obtain multiple independent single-particle images.

[0082] In one embodiment, the watershed algorithm is used to perform segmentation processing on the binary image, so that each particle in the binary image serves as a single-particle image. The watershed algorithm is a mathematical morphology segmentation method based on topological theory and has wide applications in the field of image segmentation.

[0083] It should be noted that bubbles are usually included in the oil. Through steps S251 to S254, the identification of bubbles is achieved, and the number of bubbles is obtained. And in step S254, after removing the regions corresponding to the bubbles in the binary image, the particles in the binary image are segmented to obtain multiple independent single-particle images. In this way, the influence of bubbles on the segmentation can be avoided, and the segmentation accuracy can be improved.

[0084] Step S260: Based on the single-particle image and the image to be recognized, determine the light transmittance, pixel size, and surface roughness of the particles;

[0085] Step S270: Calculate the actual size of the particles based on the conversion ratio and the pixel size of the particles;

[0086] Exemplarily, the conversion ratio is A, the pixel size of the particle includes the pixel length N and the pixel width M, and the actual size of the particle includes the length D and the width W. Then D = A * N, and W = A * M.

[0087] Step S280: Determine the target type of the particles based on the light transmittance, actual size, and surface roughness of the particles.

[0088] It's worth noting that particles in industrial oils are categorized into cutting abrasive particles, fatigue particles, sliding wear particles, and fiber particles. The number of each particle type in the image to be identified can be used to deduce the concentration of each particle type in the industrial oil, allowing the source of contamination to be determined based on the concentration of each particle type. For example, cutting abrasive particles are hard particles with long, sharp edges. They may be caused by friction between the machining mechanism and the hard workpiece, which easily scratches the oil film. Fatigue particles are typically flakes or blocks with irregular edges. They are often shed from metal surfaces such as bearings and gears. Excessive levels of fatigue particles may indicate a risk of component fatigue failure. Sliding wear particles are relatively small, smooth, and nearly spherical. They are formed by the shedding of surface material due to sliding friction. Fiber particles are long, thin, and filamentous. They may be shed from sealing materials or filter elements. Excessive levels of fiber particles can cause filter clogging.

[0089] The method for identifying industrial oil contamination particles of an embodiment of the present application, through the above-mentioned steps S210 to S280, controls the backlight assembly to emit light when the presence of oil in the transparent pipe is detected, and controls the image acquisition board to perform image acquisition to obtain an image to be identified with scale lines; pre-processes the image to be identified to obtain a binary image; calculates the pixel distance between any two adjacent scale lines in the binary image; obtains a conversion ratio based on the pixel distance and the actual distance between the two scale lines; performs image segmentation processing on the binary image to obtain multiple independent single particle images, each single particle image including only one particle; determines the transmittance, pixel size, and surface roughness of the particle based on the single particle image and the image to be identified; calculates the actual size of the particle based on the conversion ratio and the pixel size of the particle; and determines the target type of the particle based on the transmittance, actual size, and surface roughness of the particle. In this way, the transmittance, actual size, and surface roughness of the particle are obtained from the image to be identified with scale lines, thereby determining the target type of the particle and obtaining the number of particles of each type in the image to be identified. Accurately detect the number of each particle in the image to be identified.

[0090] In some embodiments, step S280 , determining the target type of the particle based on the light transmittance, actual size, and surface roughness of the particle includes but is not limited to steps S281 and S282 ;

[0091] Step S281, constructing a particle information array based on the light transmittance, actual size and surface roughness of the particles;

[0092] In one embodiment, since the input of the particle recognition model is a one-dimensional vector, based on the light transmittance, actual size, and surface roughness of the particle, a particle information array is constructed and used as the one-dimensional vector as the input of the particle recognition model. Exemplarily, if the light transmittance of the particle is K, the actual size is [D, W], where D is the length and W is the width, and the surface roughness is L, then the obtained particle information array is [K, D, W, L].

[0093] Step S282: Input the particle information array into the trained particle recognition model to obtain the target type.

[0094] It should be noted that the main differences between the types of each particle lie in the light transmittance, actual size, and surface roughness. In this application, the light transmittance, actual size, and surface roughness of the particle are used as the particle recognition model, enabling the particle recognition model to obtain the characteristic information of the particle and resulting in a relatively high classification accuracy of the particle recognition model.

[0095] It should be noted that in the traditional method, the image is generally directly input into the neural network model. Due to the presence of a lot of interference information in the image, the processing speed of the neural network model is slow, and the samples that can be used as training images in this field are few, which easily leads to overfitting of the model, resulting in a very low classification accuracy. In this application, the key information of the particle is directly used as the input of the model. And in the case where the samples used as training images in this field are few, the light transmittance, actual size, and surface roughness of the particle can provide prior knowledge, which can reduce the risk of overfitting of the recognition model, thereby improving the accuracy of the particle recognition model.

[0096] In some embodiments, the particle recognition model adopts a Multilayer Perceptron (MLP) model. The MLP model is composed of multiple neurons (or called neural nodes) arranged in a hierarchical structure. These layers include an input layer, one or more hidden layers, and an output layer. The neurons between layers are connected by weights, and the information propagates forward from the input layer to the output layer in sequence without feedback connections. Through the non-linear transformation of the hidden layer, the MLP can approximate any continuous function (universal approximation theorem), which makes it have powerful capabilities in dealing with complex tasks.

[0097] In some embodiments, the training steps of the particle recognition model include Step S310 to Step S360;

[0098] Step S310: Obtain multiple training samples and the corresponding true classifications; wherein, the true classifications include cutting abrasive particles, fatigue particles, sliding wear particles, and fiber particles; the multiple training samples include the training information arrays of cutting abrasive particles, fatigue particles, sliding wear particles, and fiber particles; the training information array is constructed from the light transmittance, actual size, and surface roughness of the particles.

[0099] Step S320: Input the training samples into the initial particle recognition model to obtain the training classifications of the training samples.

[0100] Step S330: Calculate the loss value based on the training classifications and true classifications of the training samples.

[0101] Step S340: Iteratively optimize the particle recognition model based on the loss value to obtain the trained particle recognition model.

[0102] It should be noted that the loss value is calculated using a preset loss function. After obtaining the loss value, the initial particle recognition model is updated using the gradient descent algorithm based on the loss value, and this process is repeated cyclically until the number of cycles reaches the preset number, thereby obtaining the trained particle recognition model. The loss function can be a cross-entropy loss function or other loss functions, and the present application does not make specific limitations in this regard.

[0103] In some embodiments, step S220 includes step S221 and step S222;

[0104] Step S221: Perform grayscale transformation on the image to be recognized to obtain a grayscale image.

[0105] Step S222: Perform binarization processing on the grayscale image to obtain a binary image.

[0106] It is worth noting that the image to be recognized is an RGB image. Converting the RGB image to a grayscale image, the grayscale image only contains luminance information. Compared with the color information of the red, green, and blue channels of the color image, the grayscale image is more simplified, reducing the amount of data to be processed. Moreover, the classification of particles has nothing to do with the color of the particles. Grayscaling can remove the interference of color information and highlight the contours and details of the image. Then, binarization processing is performed on the grayscale image to obtain a binary image. Binarization processing realizes dividing the pixel points in the grayscale image into two categories: black and white, which can significantly enhance the contrast of the image, making the targets in the image more clearly presented. And binarization processing can remove the noise points in the grayscale image, thereby improving the clarity and accuracy of the image.

[0107] In some implementations, the steps for determining the light transmittance of the particles include step S410 and step S420;

[0108] Step S410, calculate the first average gray value of the non-granular area in the grayscale image, and calculate the second average gray value of the single-particle image;

[0109] Step S420, calculate the light transmittance based on the first average gray value and the second average gray value.

[0110] In one embodiment, determine the corresponding area of the single-particle image in the grayscale image, and calculate the average gray value of the corresponding area as the second average gray value of the single-particle image. Specifically, the light transmittance = the second average gray value / the first average gray value.

[0111] In some embodiments, the determination steps of the pixel size of the particle include step S510 to step S520.

[0112] Step S510, calculate the minimum bounding rectangle of the particle based on the single-particle image;

[0113] Step S520, take the length and width of the minimum bounding rectangle as the pixel size.

[0114] Specifically, calculate the extreme coordinate values of the contour of the particle in the horizontal and vertical directions in the particle image, generate a rectangle parallel to the coordinate axes as the minimum bounding rectangle. Take the length and width of the minimum bounding rectangle as the pixel size.

[0115] In some embodiments, the determination steps of the surface roughness of the particle include step S610 to step S630;

[0116] Step S610, perform noise removal processing on the single-particle image to obtain a denoised image;

[0117] Step S620, perform sharpening processing on the denoised image to obtain a sharpened image;

[0118] Step S630, convert the sharpened image into gray contour data, and calculate the approximate contour arithmetic mean deviation of the gray contour data. The approximate contour arithmetic mean deviation is used as the surface roughness of the particle.

[0119] Specifically, first perform Gaussian filtering denoising on the single-particle image to further remove the noise in the single-particle image and obtain a denoised image, which can effectively suppress Gaussian noise and retain the edge structure. However, Gaussian filtering denoising may cause the edges of the particles in the single-particle image to become blurred. Therefore, it is necessary to perform sharpening processing on the denoised image. Specifically, use the Laplacian operator sharpening algorithm to perform sharpening processing on the denoised image to obtain a sharpened image, which can enhance the blurred edges and details of the particles, improve the local gradient change, and improve the accuracy of the approximate contour arithmetic mean deviation calculated subsequently. Convert the sharpened image into gray contour data and calculate the approximate contour arithmetic mean deviation of the gray contour data.

[0120] In some embodiments, step S630 specifically includes the following steps:

[0121] Step S631, extracting contour data of the particle contour from the sharpened image through an edge detection algorithm; wherein, the contour data includes the coordinates of the points in the contour;

[0122] Step S632, determining the center line of the particle contour;

[0123] Step S633, calculating the vertical distance between the points in the contour and the center line as the contour vertical distance;

[0124] Step S634, obtaining the approximate contour arithmetic mean deviation based on the contour vertical distance.

[0125] Specifically, the calculation formula of the approximate contour arithmetic mean deviation is:

[0126]

[0127] where Ra is the approximate contour arithmetic mean deviation, n is the number of sampling points on the contour line, and xi is the contour vertical distance between the i-th sampling point and the center line; the center line is the reference line of the contour, making the algebraic sum of the deviations of each point on the contour line to the center line zero.

[0128] It should be noted that the calculation process of the true contour arithmetic mean deviation requires three-dimensional data of the particles for calculation. For example, it is necessary to measure the three-dimensional data of the particles through relevant instruments, and the process is very complicated. In this application, based on a single-particle image, calculating the approximate contour arithmetic mean deviation as the surface roughness can improve the processing efficiency, and the approximate contour arithmetic mean deviation has the ability to characterize the surface roughness and can be used as the characteristic information of the particles.

[0129] The second aspect of the embodiments of the present application provides an identification device for industrial oil contamination particles, which is applied to Figure 1 the schematic identification device. Refer to Figure 3 , Figure 3 which is the structural schematic diagram of the identification device for industrial oil contamination particles in the embodiments of the present application. The device includes:

[0130] An image acquisition module 310, configured to control the backlight assembly to emit light and control the image acquisition board to perform image acquisition to obtain an image to be identified with scale lines when it is detected that there is oil in the transparent pipeline;

[0131] A preprocessing module 320, configured to preprocess the image to be identified to obtain a binary image;

[0132] A first calculation module 330, configured to calculate the pixel distance between any two adjacent scale lines in the binary image;

[0133] A second calculation module 340, configured to obtain a conversion ratio based on the pixel distance and the actual distance between two scale lines;

[0134] An image segmentation module 350, configured to perform image segmentation processing on the binary image to obtain a plurality of independent single-particle images; each single-particle image includes only one particle;

[0135] A first determination module 360, configured to determine the light transmittance, pixel size, and surface roughness of the particle based on the single-particle image and the image to be recognized;

[0136] A third calculation module 370, configured to calculate the actual size of the particle based on the conversion ratio and the pixel size of the particle;

[0137] A second determination module 380, configured to determine the target type of the particle based on the light transmittance, actual size, and surface roughness of the particle.

[0138] The industrial oil contamination particle recognition device according to the embodiment of the present application is used to execute the industrial oil contamination particle recognition method according to the embodiment of the first aspect. When executing the method, in the case of detecting that there is oil in the transparent pipeline, the backlight assembly is controlled to emit light, and the image acquisition board is controlled to perform image acquisition to obtain an image to be recognized with scale lines; the image to be recognized is preprocessed to obtain a binary image; the pixel distance between any two adjacent scale lines in the binary image is calculated; a conversion ratio is obtained based on the pixel distance and the actual distance between the two scale lines; the binary image is subjected to image segmentation processing to obtain a plurality of independent single-particle images; each single-particle image includes only one particle; the light transmittance, pixel size, and surface roughness of the particle are determined based on the single-particle image and the image to be recognized; the actual size of the particle is calculated based on the conversion ratio and the pixel size of the particle; the target type of the particle is determined based on the light transmittance, actual size, and surface roughness of the particle. In this way, the light transmittance, actual size, and surface roughness of the particle are obtained through the image to be recognized with scale lines, so as to determine the target type of the particle, and the number of each type of particle in the image to be recognized can be obtained. Accurately detecting the number of each type of particle in the image to be recognized is achieved.

[0139] It should be noted that the specific implementation manner of the industrial oil contamination particle recognition device is basically the same as the specific embodiment of the above-mentioned industrial oil contamination particle recognition method, and will not be repeated here. On the premise of meeting the requirements of the embodiment of the present application, other functional units may be provided in the industrial oil contamination particle recognition device to implement the industrial oil contamination particle recognition method in the above embodiment.

[0140] In the third aspect of the embodiments of the present application, an electronic device is provided. The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the method for identifying industrial oil contamination particles according to any one of the embodiments of the first aspect. The electronic device can be any intelligent terminal including a tablet computer, an in-vehicle computer, etc.

[0141] Referring to Figure 4 , Figure 4 FIG. is a schematic structural diagram of an electronic device according to an embodiment. The electronic device includes:

[0142] A processor 401, which can be implemented in ways such as a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided by the embodiments of the present application;

[0143] A memory 402, which can be implemented in forms such as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 402 can store an operating system and other application programs. When implementing the technical solutions provided by the embodiments of this specification through software or firmware, the relevant program codes are stored in the memory 402 and are called by the processor 401 to execute the method for identifying industrial oil contamination particles of the embodiments of the present application;

[0144] An input / output interface 403, which is used to implement information input and output;

[0145] A communication interface 404, which is used to implement communication and interaction between this device and other devices. Communication can be achieved through a wired method (such as USB, network cable, etc.) or through a wireless method (such as a mobile network, WIFI, Bluetooth, etc.);

[0146] A bus 405, which transmits information between various components of the device (such as the processor 401, the memory 402, the input / output interface 403, and the communication interface 404);

[0147] Among them, the processor 401, the memory 402, the input / output interface 403, and the communication interface 404 are communicatively connected to each other inside the device through the bus 405.

[0148] In a fourth aspect embodiment of the present application, a computer-readable storage medium stores a computer program which, when executed by a processor, implements the method for identifying industrial oil contamination particles according to any one of the first aspect embodiments.

[0149] As a non-transitory computer-readable storage medium, a memory can be used to store non-transitory software programs and non-transitory computer-executable programs. In addition, the memory can include high-speed random access memory, and can also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0150] The embodiments described in the embodiments of the present application are for more clearly illustrating the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present application are equally applicable to similar technical problems.

[0151] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than those shown in the figures, or combine certain steps, or different steps.

[0152] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0153] Those of ordinary skill in the art can understand that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, and appropriate combinations thereof.

[0154] In the description of this application and the above-mentioned accompanying drawings, terms such as "first", "second", "third", "fourth", etc. (if any) are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of this application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0155] It should be understood that in this application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the mapping relationship of mapped objects and indicates that there can be three relationships. For example, "A and / or B" can represent: only A exists, only B exists, and both A and B exist at the same time. Among them, A and B can be singular or plural. The character " / " generally indicates that the mapped objects before and after are in an "or" relationship. "At least one (one) of the following" or its similar expressions refer to any combination of these items, including any combination of single items (ones) or plural items (ones). For example, at least one (one) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0156] In several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the above-mentioned division of units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling, direct coupling, or communication connection to each other can be an indirect coupling or communication connection through some interfaces, devices, or units, and can be in electrical, mechanical, or other forms.

[0157] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0158] In addition, in each embodiment of the present application, the functional units can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0159] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in each embodiment of the present application. The aforementioned storage medium includes: various media that can store programs such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0160] The preferred embodiments of the embodiments of the present application have been described above with reference to the accompanying drawings. However, this does not limit the scope of the rights of the embodiments of the present application. Any modification, equivalent replacement, and improvement made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall fall within the scope of the rights of the embodiments of the present application.

Claims

1. A method for identifying industrial oil pollution particles, characterized in that, An identification device applied to industrial oil pollution particles, the identification device includes a transparent pipeline, a glass lens, a backlight assembly, a macro lens and an image acquisition board, the glass lens is provided with scale lines, the transparent pipeline is used for the oil to be detected to pass through, the backlight assembly is located below the transparent pipeline, and the glass lens is arranged on the outer side wall of the bottom of the transparent pipeline; The macro lens is located below the glass lens, and the image acquisition board is connected to the macro lens; The method includes: When it is detected that there is oil in the transparent pipeline, control the backlight assembly to emit light, and control the image acquisition board to perform image acquisition to obtain an image to be recognized with scale lines; Preprocess the image to be recognized to obtain a binary image; Calculate the pixel distance between any two adjacent scale lines in the binary image; Based on the pixel distance and the actual distance between two scale lines, obtain a conversion ratio; Perform image segmentation processing on the binary image to obtain a plurality of independent single-particle images; each single-particle image only includes one particle; Based on the single-particle image and the image to be recognized, determine the light transmittance, pixel size and surface roughness of the particle; Based on the conversion ratio and the pixel size of the particle, calculate the actual size of the particle; Based on the light transmittance, actual size and surface roughness of the particle, determine the target type of the particle.

2. The method for identifying industrial oil pollution particles according to claim 1, wherein The determining the target type of the particle based on the light transmittance, actual size and surface roughness of the particle includes: Based on the light transmittance, actual size and surface roughness of the particle, construct a particle information array; Input the particle information array into a trained particle recognition model to obtain the target type.

3. The method for identifying industrial oil pollution particles according to claim 2, characterized in that, The training steps of the particle recognition model include: Obtain a variety of training samples and the corresponding true classifications of the training samples; among them, the true classifications include cutting abrasive particles, fatigue particles, sliding wear particles, fiber particles; a variety of the training samples include a training information array of cutting abrasive particles, a training information array of fatigue particles, a training information array of sliding wear particles and a training information array of fiber particles; the training information array is constructed by the light transmittance, actual size and surface roughness of the particle; Input the training samples into an initial particle recognition model to obtain the training classification of the training samples; Based on the training classification of the training samples and the true classification, calculate a loss value; Based on the loss value, perform iterative optimization on the particle recognition model to obtain a trained particle recognition model.

4. The method for identifying industrial oil pollution particles according to claim 1, characterized in that, The preprocessing the image to be recognized to obtain a binary image includes: Perform gray-scale transformation on the image to be recognized to obtain a gray-scale image; Perform binary processing on the gray-scale image to obtain the binary image.

5. The method for identifying industrial oil pollution particles according to claim 4, characterized in that The determining steps of the light transmittance of the particle include: Calculate the first average gray value of the non-particle area in the gray-scale image, and calculate the second average gray value of the single-particle image; Based on the first average gray value and the second average gray value, calculate the light transmittance.

6. The method for identifying industrial oil pollution particles according to claim 1, wherein, The determining steps of the pixel size of the particle include: Based on the single-particle image, calculate the minimum bounding rectangle of the particle; Take the length and width of the minimum bounding rectangle as the pixel size.

7. The method for identifying industrial oil contamination particles according to claim 1, characterized in that, The steps for determining the surface roughness of the particle include: Perform noise removal processing on the single-particle image to obtain a denoised image; Perform sharpening processing on the denoised image to obtain a sharpened image; Convert the sharpened image into gray-scale contour data, and calculate the approximate contour arithmetic mean deviation of the gray-scale contour data. The approximate contour arithmetic mean deviation is used as the surface roughness of the particle.

8. An identification device for industrial oil pollution particles, characterized in that, Applied to an identification device for industrial oil pollution particles, the identification device includes a transparent pipeline, a glass lens, a backlight assembly, a macro lens, and an image acquisition board. The glass lens is provided with scale lines. The transparent pipeline is used for the oil to be detected to pass through. The backlight assembly is located below the transparent pipeline. The glass lens is provided on the outer side wall of the bottom of the transparent pipeline; The macro lens is located below the glass lens, and the image acquisition board is connected to the macro lens; The device includes: An image acquisition module, configured to control the backlight assembly to emit light and control the image acquisition board to perform image acquisition when it is detected that there is oil in the transparent pipeline, so as to obtain an image to be identified with scale lines; A preprocessing module, configured to preprocess the image to be identified to obtain a binary image; A first calculation module, configured to calculate the pixel distance between any two adjacent scale lines in the binary image; A second calculation module, configured to obtain a conversion ratio based on the pixel distance and the actual distance between two scale lines; An image segmentation module, configured to perform image segmentation processing on the binary image to obtain a plurality of independent single-particle images; each of the single-particle images includes only one particle; A first determination module, configured to determine the light transmittance, pixel size, and surface roughness of the particle based on the single-particle image and the image to be identified; A third calculation module, configured to calculate the actual size of the particle based on the conversion ratio and the pixel size of the particle; A second determination module, configured to determine the target type of the particle based on the light transmittance, actual size, and surface roughness of the particle.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the identification method for industrial oil pollution particles 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 identification method for industrial oil pollution particles according to any one of claims 1 to 7.

Images