Fiber reinforced resin composite pipe manufacturing method and device based on image evaluation

Abstract

The invention relates to a fiber reinforced resin composite pipe manufacturing method and device based on image evaluation. The method comprises a composite pipe inner layer weaving stage, a composite pipe middle layer weaving stage, a composite pipe outer layer weaving stage and a composite pipe surface image evaluation stage. The high-temperature porcelainized modified silicone rubber fiber is used as a weaving material of the composite pipe, so that the surface performance of the composite pipe is improved. In addition, a segmented image acquisition and splicing technology is adopted in the image evaluation process, and the 360-degree circumferential surface image of the composite pipe can be completely obtained; in combination with multi-direction feature extraction, sliding window sampling along the axial direction and the circumferential direction respectively, and calculation based on a relative entropy algorithm, probability distribution differences of image segments are converted into numerical indexes, uniformity of different regions can be quantitatively represented, and by constructing a similarity matrix and in combination with mean deviation analysis, the uniformity of different regions can be quantitatively represented. An abnormal area can be positioned to a single sliding window, accurate positioning of a defect position is realized, and the comprehensiveness of defect identification is improved.

Description

Technical Field

[0001] The present application relates to the technical field of composite pipe manufacturing, and in particular to a method and device for manufacturing a fiber reinforced resin composite pipe based on image evaluation. Background Art

[0002] As a high-performance structural component woven from multiple layers of materials, composite pipes are widely used in aerospace, petrochemical, and high-end equipment manufacturing. Their surface quality (such as fiber distribution uniformity, interlayer bonding strength, shear resistance, specific strength performance, etc.) directly affects the mechanical properties and service life of the product.

[0003] At present, metal fibers are generally used in large quantities in the manufacturing process of traditional composite pipes to enhance the strength of composite pipes. Although they can provide higher mechanical strength, they have insulation and electromagnetic compatibility defects, poor interface compatibility and other problems. With the demand for multifunctional composite pipe materials in modern industry, in some fields, materials are required to have both high insulation and extremely high fire resistance, which traditional metal fiber systems cannot meet at the same time.

[0004] In addition, the existing composite tube manufacturing process cannot monitor the uniformity of fiber distribution in real time during the multi-layer weaving process. Uneven fiber distribution is the main cause of interface defects. Once it occurs, it will have a great impact on the surface performance of the composite tube. Therefore, it is necessary to improve the surface material of the composite tube and perform real-time image evaluation of the surface. Summary of the invention

[0005] In order to overcome the above-mentioned deficiencies of the prior art, the present application provides a method and device for manufacturing a fiber-reinforced resin composite pipe based on image evaluation, which specifically adopts the following technical solutions: A method for manufacturing a fiber reinforced resin composite pipe based on image evaluation, characterized in that the method comprises the following steps: The inner layer of the composite tube is obtained by weaving through a fiber weaving technology; The middle layer of the composite pipe is obtained by weaving the inner surface of the composite pipe using fiber weaving technology; The outer layer of the composite pipe is obtained by weaving the surface of the middle layer of the composite pipe using fiber weaving technology; When weaving the inner layer, middle layer and outer layer of the composite tube, image evaluation is performed on the surface of the composite tube respectively: Using an image acquisition device to acquire images of the current circumferential surface of the composite pipe in sections to obtain a surface image of the composite pipe; Performing image preprocessing on the collected composite tube surface image; the image preprocessing includes image correction, image stitching and grayscale conversion; The preprocessed composite tube surface image is slid along preset directions by a sliding window method to obtain a plurality of image segments; the preset directions include a first direction corresponding to the axial direction of the composite tube and a second direction corresponding to the circumferential direction of the composite tube; Calculate the similarity of adjacent image segments based on the relative entropy algorithm: ; in is the similarity between adjacent image segments P and Q; is the distribution probability of the image segment P; is the distribution probability of image segment Q; Obtaining a similarity matrix of the composite tube surface image according to the similarity calculation result; The uniformity of the current circumferential surface of the composite pipe and the abnormal position are determined based on the similarity matrix.

[0006] Optionally, the step of performing image preprocessing on the collected composite tube surface image includes: Acquire internal parameter data of the image acquisition device and the radius of the current composite tube respectively; the internal parameter data at least includes focal length, principal point coordinates and distortion parameters; A cylindrical coordinate system is established based on the radius and axis position of the composite tube; The pixel coordinates of the composite tube surface images collected by different image acquisition devices are Mapping into cylindrical coordinates: ; in is the pixel coordinate Angle in cylindrical coordinates; is the pixel coordinate Height in cylindrical coordinates; is the principal point coordinate of the composite tube surface image in the horizontal direction; is the principal point coordinate of the composite tube surface image in the vertical direction; is the radius of the composite tube; Convert pixel coordinates in cylindrical coordinate system into planar unfolded coordinates : .

[0007] Optionally, the step of performing image preprocessing on the collected composite tube surface image includes: The feature extraction algorithm is used to extract feature points from the surface image of the composite tube after plane expansion; Finding matching point pairs among the feature points of the composite tube surface images acquired by adjacent image acquisition devices based on feature point matching; According to the matching point pairs, the homography matrix between the composite tube surface images acquired by adjacent image acquisition devices is calculated using the RANSAC algorithm; According to the homography matrix between adjacent composite tube surface images, geometric transformation is performed on the composite tube surface images to align the adjacent composite tube surface images; Find the overlapping area based on the aligned adjacent composite tube surface images to obtain the image fusion area; Based on the image fusion algorithm, the image fusion areas of adjacent composite pipe surface images are spliced ​​and combined to obtain a complete and expanded composite pipe circumferential surface image.

[0008] Optionally, the step of sliding the preprocessed composite tube surface image along a preset direction in a sliding window manner includes: Determine the sliding window size and sliding step length according to the composite tube diameter and image resolution; Based on the first direction and the second direction, the sliding window slides according to the set sliding step length to obtain a plurality of image segments; According to the acquisition positions of the image segments, a plurality of image segments are arranged and distributed in a matrix to obtain an image segment matrix.

[0009] Optionally, the step of obtaining a similarity matrix of the composite tube surface image according to the similarity calculation result includes: Based on the image segment matrix, similarities are calculated for adjacent image segments according to a first direction and a second direction respectively to obtain a similarity matrix along the first direction. D 1 and the similarity matrix along the second direction D 2: ; Among them is Represents the image fragment at row n and column m; is an image segment along the first direction and image fragments The similarity between is the image segment along the second direction and image fragments The similarity between .

[0010] Optionally, the step of judging the uniformity of the current circumferential surface of the composite pipe and the abnormal position based on the similarity matrix includes: Calculate the means of the elements in the first similarity matrix and the second similarity matrix respectively; Calculating the difference between the element at each position in the first similarity matrix and the corresponding mean and calculating the difference between the element at each position in the first similarity matrix and the corresponding mean; Determine whether the absolute values ​​of the calculated differences in the first similarity matrix and the second similarity matrix exceed a preset threshold: When the absolute value of the calculated difference is greater than or equal to a preset threshold, the corresponding element in the corresponding similarity matrix is ​​judged to be abnormal, and the corresponding abnormal image segment position is obtained according to the position of the abnormal element in the corresponding similarity matrix; When the absolute value of the calculated difference is less than the preset threshold, it is determined that there is no abnormality.

[0011] Optionally: the inner layer is woven from a twisted fiber bundle of a mixture of high-temperature porcelain-modified silicone rubber fibers impregnated with resin and glass fibers; the middle layer includes at least two layers, and the middle layer is woven from a twisted fiber bundle of a mixture of high-temperature porcelain-modified silicone rubber fibers impregnated with resin and glass fibers; the outer layer includes a first outer layer and a second outer layer located on the surface of the first outer layer, the first outer layer is woven from a plurality of strands of glass fiber wire impregnated with resin, and the second outer layer is woven from a twisted fiber bundle of a mixture of high-temperature porcelain-modified silicone rubber fibers impregnated with resin and glass fibers; The method for preparing the high temperature porcelain-modified silicone rubber fiber comprises the following steps: (1) Weighing a certain amount of methyl chlorosilicone rubber, heating it and immersing it in a cyclohexane solvent for pretreatment to obtain pretreated methyl chlorosilicone rubber; (2) dissolving triethoxysilane in ethanol, adding ammonia water to adjust the pH to 4-5 to form a hydrolysis solution, adding polydimethylsiloxane to the hydrolysis solution, ultrasonically treating to disperse the polydimethylsiloxane, then adding nano-sized aluminum nitride, stirring to disperse the aluminum nitride, and obtaining a silicon nitride precursor solution; (3) mixing the silicon nitride precursor solution obtained in step (2) with the pretreated methylchlorosilicone rubber obtained in step (1), heating the mixture to 250-300° C., and keeping the mixture to react for 2-4 hours to obtain a ceramic precursor material; (4) heating the ceramic precursor material to a molten state, extruding it to form fibers, cooling and solidifying the fibers, and then sintering the fibers at a high temperature; (5) Spraying a silicon dioxide dispersion onto the surface of the fiber after sintering in step (4), and drying and curing to obtain the high temperature porcelainizable modified silicone rubber fiber.

[0012] Optional: The pretreatment process of step (1) is specifically as follows: heating the methylchlorosilicone rubber to 80-100° C., immersing the heated methylchlorosilicone rubber in cyclohexane solvent, stirring for 10-15 minutes, and then vacuum degassing to obtain pretreated methylchlorosilicone rubber.

[0013] Optionally: in step (2), the mass ratio of triethoxysilane to ethanol is 1:3-5, the mass of polydimethylsiloxane is 1-5% of the hydrolysis solution, and the mass of nano-aluminum nitride is 2-5% of the hydrolysis solution; in step (3), the mass ratio of silicon nitride precursor solution to pretreated methylchlorosilicone rubber is 1:5-10; in step (4), the temperature of high-temperature sintering is 1200-1300°C and maintained for 2-3 hours; in step (5), the silicon dioxide dispersion is a mixture of silicon dioxide and ethanol, and the mass volume ratio of silicon dioxide to ethanol is 1g:5-10mL.

[0014] Optionally, the method for preparing the twisted fiber bundle of the resin-impregnated high-temperature porcelainizable modified silicone rubber fiber mixed with glass fiber comprises the following steps: (1) mixing the high temperature porcelainizable modified silicone rubber fiber prepared above with glass fiber in a mass ratio of 1:1-3, and after mixing, twisting the two fiber bundles together by twisting technology to form a twisted fiber bundle; (2) Immersing the twisted fiber bundle in epoxy resin impregnation liquid to soak the fiber bundle and ensure that the fiber bundle is completely penetrated by the resin. The fiber bundle impregnated with the resin is cured and dried by heating to obtain a twisted fiber bundle of a mixture of high-temperature porcelainizable modified silicone rubber fiber and glass fiber impregnated with the resin.

[0015] In addition, the present application also discloses a fiber-reinforced resin composite pipe manufacturing device based on image evaluation, the device comprising: An inner layer braiding unit, used for braiding an inner layer of the composite tube by using a fiber braiding technology; An intermediate layer braiding unit is used to braid the inner layer surface of the composite pipe by using a fiber braiding technology to obtain an intermediate layer of the composite pipe; An outer layer braiding unit is used to braid the outer layer of the composite pipe by using a fiber braiding technology on the surface of the middle layer of the composite pipe; Image evaluation unit for image evaluation of the composite pipe surface during braiding of the inner layer, the middle layer and the outer layer: Using an image acquisition device to acquire images of the current circumferential surface of the composite pipe in sections to obtain a surface image of the composite pipe; Performing image preprocessing on the collected composite tube surface image; the image preprocessing includes image correction, image stitching and grayscale conversion; The preprocessed composite tube surface image is slid along preset directions by a sliding window method to obtain a plurality of image segments; the preset directions include a first direction corresponding to the axial direction of the composite tube and a second direction corresponding to the circumferential direction of the composite tube; Calculate the similarity of adjacent image segments based on the relative entropy algorithm: ; in is the similarity between adjacent image segments P and Q; is the distribution probability of the image segment P; is the distribution probability of image segment Q; Obtaining a similarity matrix of the composite tube surface image according to the similarity calculation result; The uniformity of the current circumferential surface of the composite pipe and the abnormal position are determined based on the similarity matrix.

[0016] Beneficial Effects The technical solution of this application has the following beneficial effects: (1) The fiber-reinforced resin composite pipe manufacturing method of the present application adopts segmented image acquisition and splicing technology, which breaks through the limitations of traditional single-point detection and can completely obtain the 360° circumferential surface image of the composite pipe to ensure that there are no blind spots in the detection; and based on the multi-directional feature extraction method, it can sample along the axial and circumferential sliding windows respectively, and capture the feature differences of the composite pipe surface along the length direction and the circumferential direction at the same time, thereby improving the comprehensiveness of defect identification.

[0017] (2) The fiber-reinforced resin composite pipe manufacturing method of the present application is based on the similarity calculation of the relative entropy algorithm, which converts the probability distribution differences of image segments into numerical indicators, and can quantitatively characterize the uniformity of different areas. Compared with the traditional threshold segmentation method, it has higher sensitivity to subtle defects (such as 0.1mm fiber misalignment); and by constructing a similarity matrix and combining it with mean deviation analysis, the abnormal area can be located to a single sliding window, achieving centimeter-level precise positioning of the defect position.

[0018] (3) The fiber-reinforced resin composite pipe manufacturing method of the present application can automatically generate uniformity evaluation results and output abnormal positions by thresholding the similarity matrix, providing an intuitive basis for real-time quality monitoring of the production process, and improving efficiency by more than 80% compared to manual visual inspection.

[0019] (4) The fiber-reinforced resin composite pipe manufacturing method of the present application uses high-temperature ceramicizable modified silicone rubber fiber as the braiding material of the composite pipe, which has excellent fire resistance, ablation resistance and high temperature resistance. Through high-temperature sintering, the modified silicone rubber fiber forms a high-temperature resistant ceramic layer in a high-temperature environment, which makes it have extremely strong thermal stability at high temperatures. The chemical stability and oxidation resistance of the silicone rubber matrix enable the fiber to effectively resist combustion and oxidation under extremely high temperature conditions, while the ablation rate is low and good structural integrity can be maintained. In addition, during the combustion process, the combustion characteristics of the modified silicone rubber are low smoke, non-toxic, and the flame propagation speed is slow, which further improves the fire prevention and disaster prevention performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1This is a flow chart of the method for manufacturing a fiber reinforced resin composite pipe in an embodiment of the present application.

[0021] Figure 2 Schematic diagram of the process of capturing image segments using a sliding window in an embodiment of the present application.

[0022] Figure 3 Schematic diagram of the process of obtaining a first similarity matrix from an image segment matrix in an embodiment of the present application.

[0023] Figure 4 Schematic diagram of the process of obtaining a second similarity matrix from an image segment matrix in an embodiment of the present application.

[0024] Figure 5 It is a schematic diagram of the structure of the fiber reinforced resin composite pipe manufacturing device in the embodiment of the present application.

[0025] Figure 6 This is a structural diagram of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0026] The present application is further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present application, and cannot be used to limit the scope of protection of the present application. It should be noted that the following detailed descriptions are all exemplary and are intended to provide further explanation of the present application.

[0027] Fiber composite pipes have been widely used in many fields, such as petrochemicals, aerospace, and construction engineering, due to their excellent properties, such as high strength, light weight, and corrosion resistance. At present, the quality of traditional composite pipes is affected by many factors, including fiber materials, weaving processes, and impregnation effects. In order to ensure the quality stability and reliability of composite pipes, this application proposes a method and device for manufacturing fiber-reinforced resin composite pipes based on image evaluation. By introducing high-temperature porcelain-modified silicone rubber fiber materials in the composite pipe manufacturing process and combining image evaluation technology, the surface quality of the composite pipe is monitored in real time, which effectively improves the strength and quality of the composite pipe products.

[0028] Combination Figure 1 As shown, the fiber reinforced resin composite pipe manufacturing method based on image evaluation in this application specifically includes the following steps: Inner layer weaving: The inner layer of the composite tube is obtained by weaving through fiber weaving technology; the inner layer material is preferably a twisted fiber bundle of a mixture of high-temperature porcelain-modified silicone rubber fiber impregnated with resin and glass fiber. The high-temperature porcelain-modified silicone rubber fiber gives the composite tube good flexibility and high temperature and chemical corrosion resistance, while the glass fiber provides high-strength support. Before weaving, the fiber bundle needs to be pretreated and placed under specific temperature and humidity conditions for a period of time to make the performance of the fiber bundle more stable. During the weaving process, two-dimensional weaving or three-dimensional weaving technology is generally used. According to the design requirements of the composite tube, the weaving angle, weaving density and other parameters are precisely controlled. During the weaving process, it is necessary to ensure that the fiber bundles are evenly distributed to avoid problems such as overlapping fiber bundles and excessive gaps.

[0029] Intermediate layer weaving: The intermediate layer of the composite pipe is obtained by weaving the inner surface of the composite pipe with fiber weaving technology; the intermediate layer in this embodiment includes at least two layers, and the intermediate layer adopts a twisted fiber bundle of a mixture of high-temperature porcelain-modified silicone rubber fiber impregnated with resin and glass fiber; the high-temperature porcelain-modified silicone rubber fiber can be converted into a ceramic material under high temperature environment to enhance the high temperature resistance and fire resistance of the composite pipe. The weaving process of the intermediate layer is similar to that of the inner layer. The weaving angle and density can be adjusted according to the mechanical properties requirements of the composite pipe. When weaving two adjacent intermediate layers, a cross-weaving method is adopted to improve the overall strength and impact resistance of the composite pipe. Specifically, the intermediate layer of this embodiment is from the inside to the outside: the first intermediate layer adopts a twisted fiber bundle of a mixture of high-temperature porcelain-modified silicone rubber fiber impregnated with resin and glass fiber, which is continuously wound around the inner layer, and then the axial glass fiber impregnated with resin is laid along the axial direction of the pipe on the first intermediate layer, and then a layer of glass fiber impregnated with resin is continuously wound around the circumference to form a second intermediate layer. After weaving is completed, the middle layer needs to be checked for quality and the distribution of the fiber bundles needs to be observed. If there is any abnormality, adjustments should be made in time.

[0030] Outer layer weaving: The outer layer of the composite pipe is obtained by weaving the fiber weaving technology on the surface of the middle layer of the composite pipe; the outer layer includes a first outer layer and a second outer layer located on the surface of the first outer layer. The first outer layer uses a multi-strand glass fiber line impregnated with resin, which mainly provides the surface hardness and wear resistance of the composite pipe; the second outer layer uses a twisted fiber bundle mixed with a high-temperature porcelain-modified silicone rubber fiber impregnated with resin and glass fiber to further enhance the fire resistance and weather resistance of the composite pipe. In this embodiment, the first outer layer is woven into a cylindrical structure by a three-dimensional weaving technology using a multi-strand glass fiber line impregnated with resin, and the entire first outer layer is seamless, the whole layer structure has no overlap, there is no edge stress defect, and the fiber space distribution is uniform and dense, especially, the multi-strand glass fiber line should ensure that there are more than 3 strands of glass fiber lines at the intersection point in the space, and extend in 3 directions, so that its directional force capacity is greatly enhanced, and the pipeline shear resistance and specific strength performance are improved. The second outer layer uses a twisted fiber bundle mixed with a high-temperature porcelain-modified silicone rubber fiber and glass fiber, and is formed in combination with two-dimensional or three-dimensional weaving technology.

[0031] Image evaluation: During the weaving process, it is necessary to ensure that the fiber bundles of each layer of the composite tube are evenly distributed. Therefore, when weaving the inner layer, middle layer and outer layer of the composite tube, the surface of the composite tube is image evaluated respectively: (1) First, use an image acquisition device to capture the current circumferential surface of the composite tube in sections to obtain the surface image of the composite tube; during the weaving process of each layer of the composite tube, the image acquisition device preferably uses a high-precision industrial camera, which has high resolution, high frame rate and good optical performance, and can clearly capture the detailed features of the surface of the composite tube. In this embodiment, the industrial cameras are generally set at least three, and the industrial cameras are distributed around the axis of the composite tube, and multiple industrial cameras are fixed relative to the surface of the composite tube for shooting. Alternatively, the industrial camera can be installed on a robotic arm that can move around the axis of the composite tube, and the motion trajectory of the robotic arm can be controlled by programming to achieve a comprehensive scan of the surface of the composite tube.

[0032] It should be noted that when installing an industrial camera, the camera needs to be calibrated to ensure the accuracy of image acquisition. The calibration process generally includes determining the camera's internal parameter data, such as focal length, principal point coordinates, and distortion parameters, while measuring the radius of the current composite tube. It is also necessary to establish a cylindrical coordinate system based on the radius and axis position of the composite tube, and accurately map the image pixel coordinates collected by the camera to the cylindrical coordinate system, providing an accurate data basis for subsequent image analysis.

[0033] (2) Secondly, the collected composite tube surface image is subjected to image preprocessing; the image preprocessing includes image correction, image stitching and grayscale conversion.

[0034] Specifically, the step of performing image correction on the collected composite tube surface image in this embodiment includes: The internal parameter data of the image acquisition device and the radius of the current composite tube are obtained respectively; the internal parameter data at least includes focal length, principal point coordinates and distortion parameters.

[0035] A cylindrical coordinate system is established based on the radius and axis position of the composite pipe; it should be noted that, since the composite pipe is generally long during the composite pipe manufacturing process, the composite pipe surface image is acquired in segments in this embodiment. For example, for a 2-meter-long composite pipe, the circumferential surface image of the composite pipe can be acquired every 20cm-30cm. When establishing a cylindrical coordinate system, data acquired at different intervals can establish a cylindrical coordinate system based on the same axis position.

[0036] The pixel coordinates of the composite tube surface image acquired by the image acquisition device are Mapping into cylindrical coordinates: ; in is the pixel coordinate Angle in cylindrical coordinates; is the pixel coordinate Height in cylindrical coordinates; is the principal point coordinate of the composite tube surface image in the horizontal direction; is the principal point coordinate of the composite tube surface image in the vertical direction; is the radius of the composite tube; Convert pixel coordinates in cylindrical coordinate system into planar unfolded coordinates : .

[0037] Through the above process, the composite tube surface image acquired by the image acquisition device can be converted into a plane unfolding state to facilitate subsequent sliding window processing.

[0038] Furthermore, due to the size of the composite tube, a single camera cannot capture the entire circumferential surface at one time. In each composite tube surface image acquisition process, multiple image acquisition devices are required to capture the image. In order to obtain complete composite tube surface images of different segments, the collected composite tube surface images are also spliced ​​in this embodiment. The specific steps include: A feature extraction algorithm (such as SIFT (Scale Invariant Feature Transform) or SURF (Speeded Up Robust Feature) algorithm) is used to extract feature points from the surface image of the composite tube after the plane is unfolded.

[0039] Based on the feature point matching algorithm, matching point pairs are found among the feature points of the composite tube surface images collected by adjacent image collection devices.

[0040] According to the matching point pairs, the RANSAC (random sampling consensus) algorithm is used to calculate the homography matrix between the composite tube surface images collected by adjacent image acquisition devices.

[0041] According to the homography matrix between adjacent composite tube surface images, the composite tube surface images are geometrically transformed to align the adjacent composite tube surface images.

[0042] Based on the aligned adjacent composite tube surface images, the overlapping area is found to obtain the image fusion area.

[0043] Based on an image fusion algorithm (such as a weighted average fusion or a multi-resolution fusion algorithm), image fusion regions of adjacent composite pipe surface images are spliced ​​and combined to obtain a complete and expanded composite pipe circumferential surface image of the current segment.

[0044] In addition, before sliding window processing, the image needs to be grayed out to convert the color image into a grayscale image. Generally, the weighted average method is used to calculate the grayscale value of each pixel based on the contribution of different color channels in the RGB color model to human vision to obtain a grayscale image.

[0045] (3) Then, the pre-processed composite tube surface image is slid along preset directions by using a sliding window method to obtain a plurality of image segments; the preset directions include a first direction corresponding to the axial direction of the composite tube and a second direction corresponding to the circumferential direction of the composite tube. Figure 2 As shown, it is a process of using a sliding window to intercept image segments of the composite tube surface image, wherein Figure 2 (a) is the surface image of the composite tube. Figure 2 (b) is the obtained 3×3 matrix image fragment matrix, and the specific operation process includes: The sliding window size and sliding step length are determined according to the composite tube diameter and image resolution. The selection of the sliding window size should comprehensively consider the composite tube surface feature size and image resolution to ensure that the window can contain sufficient detail information. The sliding step length needs to be adjusted according to the computational efficiency and image analysis accuracy.

[0046] Based on the first direction and the second direction, the sliding window slides according to the set sliding step length respectively, and during the sliding process, a number of image segments are intercepted from the preprocessed composite tube surface image.

[0047] Finally, several image segments are arranged and distributed in a matrix according to the acquisition positions of the image segments to obtain an image segment matrix for subsequent similarity calculation and analysis.

[0048] (4) Next, the similarity of adjacent image segments is calculated based on the relative entropy algorithm: ; in is the similarity between adjacent image segments P and Q; is the distribution probability of the image segment P; is the distribution probability of image segment Q.

[0049] Generally, a picture is a matrix composed of several pixels, each of which has a specific color value. The relative entropy algorithm can regard the pixel distribution of a picture as a probability distribution, and use relative entropy to measure the difference between two pictures. For example, for two pictures of the same size, their pixel value distributions in each color channel (RGB channel) are statistically calculated to form a probability distribution and, then the relative entropy is calculated to evaluate their degree of difference. Generally, the larger the relative entropy value, the greater the difference in pixel distribution between the two pictures, and the greater the difference in picture content. In this embodiment, the relative entropy algorithm can effectively measure the degree of difference between the probability distributions of the two pictures, and calculate the similarity value of adjacent pictures. In this embodiment, if the fibers on the surface of the composite tube are generally in a uniform state, the relative entropy values ​​of adjacent image segments are low. If the fibers on the surface of the composite tube are in a non-uniform state, the relative entropy values ​​of adjacent image segments will increase.

[0050] Based on the image segment matrix, similarities are calculated for adjacent image segments in a first direction and a second direction respectively to obtain a first similarity matrix along the first direction. D 1 and the second similarity matrix along the second direction D 2: ; Among them is Represents the image fragment at row n and column m; is the image segment along the first direction and image fragments The similarity between is the image segment along the second direction and image fragments The similarity between .

[0051] Based on the above matrix transformation, the image fragment matrix can be converted into a similarity value matrix in two directions. In the subsequent abnormal value judgment stage, the abnormal position of the composite pipe surface can be judged and prompted through the first similarity matrix and the second similarity matrix, which can facilitate operators to locate surface abnormalities in a timely and accurate manner.

[0052] Combination Figure 3 As shown, according to Figure 2 The image fragment matrix of the 3×3 matrix in can obtain the similarity matrix along the first direction (ie, the axial direction of the composite tube); Figure 4 As shown, according to Figure 2The 3×3 matrix image segment matrix in can obtain the similarity matrix along the second direction (ie, the axial direction of the composite tube).

[0053] (6) Finally, the uniformity of the current circumferential surface of the composite tube and the abnormal position are determined based on the similarity matrix. The specific steps include: First, the means of the elements in the first similarity matrix and the second similarity matrix are calculated respectively; Then, the difference between the element at each position in the first similarity matrix and the corresponding mean is calculated, and the difference between the element at each position in the first similarity matrix and the corresponding mean is calculated; Determine whether the absolute values ​​of the calculated differences in the first similarity matrix and the second similarity matrix exceed a preset threshold: When the absolute value of the calculated difference is greater than or equal to a preset threshold, the corresponding element in the corresponding similarity matrix is ​​judged to be abnormal, indicating that the uniformity of the current image segment is poor. At this time, the corresponding abnormal image segment position is obtained according to the position of the abnormal element in the corresponding similarity matrix and the position of the abnormal element in the second similarity matrix; When the absolute value of the calculated difference is less than the preset threshold, it is judged that there is no abnormality, indicating that the uniformity of the current image segment is good.

[0054] by Figure 3 and Figure 4 For example, when the image fragment in the first row and first column of the image fragment matrix is ​​abnormal, Figure 3 The value of T1 in the matrix obviously exceeds the mean value of the elements in the first similarity matrix. Figure 4 The value of H1 in the second similarity matrix obviously exceeds the mean value of the elements in the second similarity matrix. Conversely, the position of the abnormal image segment in the image segment matrix can be located by determining the abnormal element T1 and the abnormal element H1.

[0055] This embodiment converts the probability distribution differences of image segments into numerical indicators, which can quantitatively characterize the uniformity of different areas, thereby having a higher sensitivity to subtle defects and achieving high-quality evaluation; and by constructing a similarity matrix and combining it with mean deviation analysis, the abnormal area can be located to a single sliding window, thereby achieving centimeter-level precise positioning of the defect position.

[0056] Furthermore, in order to improve the performance of the composite pipe, the present application improves the braiding material in the composite pipe manufacturing stage. In the braiding stage of the inner layer, middle layer and outer layer of the composite pipe, a twisted fiber bundle of high-temperature porcelain-modified silicone rubber fiber impregnated with resin and mixed with glass fiber is used, wherein the high-temperature porcelain-modified silicone rubber fiber is the core material, which can improve the surface performance of the composite pipe. The specific preparation method includes the following steps: (1) Weighing a certain amount of methyl chlorosilicone rubber, heating it and immersing it in a cyclohexane solvent for pretreatment to obtain pretreated methyl chlorosilicone rubber; (2) dissolving triethoxysilane in ethanol, adding ammonia water to adjust the pH to 4-5 to form a hydrolysis solution, adding polydimethylsiloxane to the hydrolysis solution, ultrasonically treating to disperse the polydimethylsiloxane, then adding nano-sized aluminum nitride, stirring to disperse the aluminum nitride, and obtaining a silicon nitride precursor solution; (3) mixing the silicon nitride precursor solution obtained in step (2) with the pretreated methylchlorosilicone rubber obtained in step (1), heating the mixture to 250-300° C., and keeping the mixture to react for 2-4 hours to obtain a ceramic precursor material; (4) heating the ceramic precursor material to a molten state, extruding it to form fibers, cooling and solidifying the fibers, and then sintering the fibers at a high temperature; (5) Spraying a silicon dioxide dispersion onto the surface of the fiber after sintering in step (4), and drying and curing to obtain the high temperature porcelainizable modified silicone rubber fiber.

[0057] Preferably, the pretreatment process of step (1) is specifically as follows: heating the methyl chlorosilicone rubber to 80-100° C., immersing the heated methyl chlorosilicone rubber in a cyclohexane solvent, stirring for 10-15 minutes, and then vacuum degassing to obtain the pretreated methyl chlorosilicone rubber.

[0058] Preferably, in step (2), the mass ratio of triethoxysilane to ethanol is 1:3-5, the mass of polydimethylsiloxane is 1-5% of the hydrolysis solution, and the mass of nano-aluminum nitride is 2-5% of the hydrolysis solution; in step (3), the mass ratio of silicon nitride precursor solution to pretreated methylchlorosilicone rubber is 1:5-10; in step (4), the temperature of high-temperature sintering is 1200-1300°C and maintained for 2-3 hours; in step (5), the silicon dioxide dispersion is a mixture of silicon dioxide and ethanol, and the mass volume ratio of silicon dioxide to ethanol is 1g:5-10mL.

[0059] Example: (1) Weigh a certain amount of methyl chlorosilicone rubber, cut it into small pieces, place the methyl chlorosilicone rubber in a heating tank, heat it to 90°C to soften it and enhance its fluidity, immerse the heated methyl chlorosilicone rubber in cyclohexane solvent, stir it for 15 minutes to remove impurities, use vacuum degassing equipment to remove air bubbles, ensure that there are no bubbles in the material, and obtain pretreated methyl chlorosilicone rubber; (2) Add 20 g of triethoxysilane and 80 g of ethanol as a solvent to a beaker, mix them thoroughly, and use an ultrasonic cleaner to treat the solution for 20 minutes to ensure that triethoxysilane is completely dissolved in the solution to form a transparent solution. Add ammonia water to the solution to adjust the pH to 4.5 to promote the hydrolysis and gelation reaction of the silicon nitride precursor to form a hydrolysis solution; add 4 g of polydimethylsiloxane to the hydrolysis solution, treat it with ultrasound for 30 minutes to completely disperse it in the solution, then add 4 g of nano-aluminum nitride, and use a high-speed mixer to mix for 10 minutes to ensure that the powder is completely dispersed and uniform, thereby obtaining a silicon nitride precursor solution; (3) 10 g of the silicon nitride precursor solution obtained in step (2) was mixed with 80 g of the pretreated methylchlorosilicone rubber obtained in step (1), and mixed using a high-speed mixer for 10 minutes to ensure that the solution and the silicone rubber were completely mixed. The mixture was heated to 290° C. and kept warm for 3 hours. The silicone rubber matrix began to crosslink, and the silicon nitride precursor reacted with the silicone rubber to obtain a ceramic precursor material. (4) The ceramic precursor material is heated to about 230°C to fully melt it and achieve good fluidity. The molten material is extruded using a porous nozzle with a pore size of about 0.5 mm to form continuous fibers. The extruded fibers are cooled and solidified by an air cooling system to ensure rapid fiber molding. The obtained fibers are placed in a high-temperature furnace in an oxygen atmosphere and heated to 1200°C at a rate of 5°C / min, and then heated to 1300°C at a rate of 2°C / min and maintained for 2 hours to convert the silicon nitride precursor into silicon nitride ceramics and react chemically with the silicone rubber matrix. (5) Silica and ethanol are mixed in a mass volume ratio of 1 g:8 mL to obtain a silica dispersion, and the silica dispersion is evenly sprayed on the surface of the fiber sintered in step (4) using a spray gun to ensure that the coating is uniform and free of bubbles. The sprayed fiber is placed in an oven at 160° C. and dried for 2 hours to solidify the coating and firmly bond it to the fiber surface, thereby obtaining the high-temperature porcelainizable modified silicone rubber fiber.

[0060] The high-temperature porcelainizable modified silicone rubber fiber and glass fiber prepared by the above embodiment are mixed in a mass ratio of 1:2. After mixing, the two fiber bundles are twisted together by twisting technology to form a uniform twisted fiber bundle, and the twisted fiber bundle is immersed in the epoxy resin impregnation liquid to soak the fiber bundle to ensure that the fiber bundle is completely penetrated by the resin. The fiber bundle impregnated with the resin is cured and dried by heating to obtain a resin-impregnated fiber bundle. Subsequently, the obtained resin-impregnated fiber bundle is used for the manufacture of composite pipes to obtain a finished composite pipe. After performance testing, it is found that the composite pipe has excellent chemical stability, high and low temperature resistance, weather resistance and insulation. At the same time, the skin is less smoke and non-toxic when burning, and has the characteristics of low combustion calorific value and slow flame propagation speed. Whether the composite pipe is internal or external, the composite pipe has excellent fire prevention and disaster prevention performance.

[0061] In addition, Figure 5 As shown, the present application also discloses a fiber reinforced resin composite pipe manufacturing device based on image evaluation, the device comprising: An inner layer braiding unit, used for braiding an inner layer of the composite tube by using a fiber braiding technology; An intermediate layer braiding unit is used to braid the inner layer surface of the composite pipe by using a fiber braiding technology to obtain an intermediate layer of the composite pipe; An outer layer braiding unit is used to braid the outer layer of the composite pipe by using a fiber braiding technology on the surface of the middle layer of the composite pipe; Image evaluation unit for image evaluation of the composite pipe surface during braiding of the inner layer, the middle layer and the outer layer: Using an image acquisition device to acquire images of the current circumferential surface of the composite pipe in sections to obtain a surface image of the composite pipe; Performing image preprocessing on the collected composite tube surface image; the image preprocessing includes image correction, image stitching and grayscale conversion; The preprocessed composite tube surface image is slid along preset directions by a sliding window method to obtain a plurality of image segments; the preset directions include a first direction corresponding to the axial direction of the composite tube and a second direction corresponding to the circumferential direction of the composite tube; Calculate the similarity of adjacent image segments based on the relative entropy algorithm: ; in is the similarity between adjacent image segments P and Q; is the distribution probability of the image segment P; is the distribution probability of image segment Q; Obtaining a similarity matrix of the composite tube surface image according to the similarity calculation result; The uniformity of the current circumferential surface of the composite pipe and the abnormal position are determined based on the similarity matrix.

[0062] The device provided in the embodiment of the present application can achieve Figure 1 To avoid repetition, the various processes implemented by the method embodiment will not be described again here.

[0063] like Figure 6 As shown, the embodiment of the present application also provides an electronic device, including a processor and a memory, a program or instruction stored in the memory and executable on the processor, and when the program or instruction is executed by the processor, the following is achieved: Figure 1 The various processes of the method embodiment shown in the figure can achieve the same technical effect. To avoid repetition, they will not be described again here.

[0064] The embodiment of the present application also provides a readable storage medium on which a program or instruction is stored. When the program or instruction is executed by a processor, the above Figure 1 The various processes of the method embodiments described above can achieve the same technical effect, and will not be described again here to avoid repetition.

[0065] The present application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the above Figure 1 The various processes of the method embodiments described above can achieve the same technical effect, and will not be described again here to avoid repetition.

[0066] It should be understood that "one embodiment" or "an embodiment" mentioned throughout the specification means that specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present application, the size of the sequence number of the above-mentioned processes does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. The above-mentioned sequence numbers of the embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.

[0067] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article 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, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0068] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another device, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0069] 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; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0070] In addition, all functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may be a separate unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0071] A person skilled in the art can understand that all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM), magnetic disks or optical disks, etc., various media that can store program codes.

[0072] Alternatively, if the above-mentioned integrated unit of the present application is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application can essentially or in other words, the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a device (which can be a terminal or a platform, etc.) to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0073] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for manufacturing a fiber-reinforced resin composite pipe based on image evaluation, characterized in that: The method comprises the following steps: The inner layer of the composite tube is obtained by weaving through a fiber weaving technology; The middle layer of the composite pipe is obtained by weaving the inner surface of the composite pipe using fiber weaving technology; The outer layer of the composite pipe is obtained by weaving the surface of the middle layer of the composite pipe using fiber weaving technology; When weaving the inner layer, middle layer and outer layer of the composite tube, image evaluation is performed on the surface of the composite tube respectively: Using an image acquisition device to acquire images of the current circumferential surface of the composite pipe in sections to obtain a surface image of the composite pipe; Performing image preprocessing on the collected composite tube surface image; the image preprocessing includes image correction, image stitching and grayscale conversion; The preprocessed composite tube surface image is slid along preset directions by a sliding window method to obtain a plurality of image segments; the preset directions include a first direction corresponding to the axial direction of the composite tube and a second direction corresponding to the circumferential direction of the composite tube; Calculate the similarity of adjacent image segments based on the relative entropy algorithm: ； in is the similarity between adjacent image segments P and Q; is the distribution probability of the image segment P; is the distribution probability of image segment Q; Obtaining a similarity matrix of the composite tube surface image according to the similarity calculation result; The uniformity of the current circumferential surface of the composite pipe and the abnormal position are determined based on the similarity matrix.

2. The method for manufacturing a fiber-reinforced resin composite pipe according to claim 1, characterized in that: The step of performing image preprocessing on the collected composite tube surface image comprises: Acquire internal parameter data of the image acquisition device and the radius of the current composite tube respectively; the internal parameter data at least includes focal length, principal point coordinates and distortion parameters; A cylindrical coordinate system is established based on the radius and axis position of the composite tube; The pixel coordinates of the composite tube surface images collected by different image acquisition devices are Mapping into cylindrical coordinates: ； in is the pixel coordinate Angle in cylindrical coordinates; is the pixel coordinate Height in cylindrical coordinates; is the principal point coordinate of the composite tube surface image in the horizontal direction; is the principal point coordinate of the composite tube surface image in the vertical direction; is the radius of the composite tube; Convert pixel coordinates in cylindrical coordinate system into planar unfolded coordinates : 。 3. The method for manufacturing a fiber-reinforced resin composite pipe according to claim 2, characterized in that: The step of performing image preprocessing on the collected composite tube surface image comprises: The feature extraction algorithm is used to extract feature points from the surface image of the composite tube after plane expansion; Finding matching point pairs among the feature points of the composite tube surface images acquired by adjacent image acquisition devices based on feature point matching; According to the matching point pairs, the homography matrix between the composite tube surface images acquired by adjacent image acquisition devices is calculated using the RANSAC algorithm; According to the homography matrix between adjacent composite tube surface images, geometric transformation is performed on the composite tube surface images to align the adjacent composite tube surface images; Find the overlapping area based on the aligned adjacent composite tube surface images to obtain the image fusion area; Based on the image fusion algorithm, the image fusion areas of adjacent composite pipe surface images are spliced ​​and combined to obtain a complete and unfolded composite pipe circumferential surface image.

4. The method for manufacturing a fiber-reinforced resin composite pipe according to claim 1, characterized in that: The step of sliding the pre-processed composite pipe surface image along the preset direction in a sliding window manner comprises: Determine the sliding window size and sliding step length according to the composite tube diameter and image resolution; Based on the first direction and the second direction, the sliding window slides according to the set sliding step length to obtain a plurality of image segments; According to the acquisition positions of the image segments, a plurality of image segments are arranged and distributed in a matrix to obtain an image segment matrix.

5. The method for manufacturing a fiber-reinforced resin composite pipe according to claim 1, characterized in that: The step of obtaining a similarity matrix of the composite tube surface image according to the similarity calculation result comprises: Based on the image segment matrix, similarities are calculated for adjacent image segments according to a first direction and a second direction respectively to obtain a similarity matrix along the first direction. D 1 and the similarity matrix along the second direction D 2: ； Among them is Represents the image fragment at row n and column m; is an image segment along the first direction and image fragments The similarity between is the image segment along the second direction and image fragments The similarity between .

6. The method for manufacturing a fiber-reinforced resin composite pipe according to claim 1, characterized in that: The step of judging the uniformity of the current circumferential surface of the composite pipe and the abnormal position based on the similarity matrix comprises: Calculate the means of the elements in the first similarity matrix and the second similarity matrix respectively; Calculating the difference between the element at each position in the first similarity matrix and the corresponding mean and calculating the difference between the element at each position in the first similarity matrix and the corresponding mean; Determine whether the absolute values ​​of the calculated differences in the first similarity matrix and the second similarity matrix exceed a preset threshold: When the absolute value of the calculated difference is greater than or equal to a preset threshold, the corresponding element in the corresponding similarity matrix is ​​judged to be abnormal, and the corresponding abnormal image segment position is obtained according to the position of the abnormal element in the corresponding similarity matrix; When the absolute value of the calculated difference is less than the preset threshold, it is determined that there is no abnormality.

7. The method for manufacturing a fiber-reinforced resin composite pipe according to claim 1, characterized in that: The inner layer is woven from a twisted fiber bundle of a mixture of high-temperature porcelain-modified silicone rubber fibers impregnated with resin and glass fibers; the middle layer includes at least two layers, and the middle layer is woven from a twisted fiber bundle of a mixture of high-temperature porcelain-modified silicone rubber fibers impregnated with resin and glass fibers; the outer layer includes a first outer layer and a second outer layer located on the surface of the first outer layer, the first outer layer is woven from a plurality of strands of glass fiber wire impregnated with resin, and the second outer layer is woven from a twisted fiber bundle of a mixture of high-temperature porcelain-modified silicone rubber fibers impregnated with resin and glass fibers; The method for preparing the high temperature porcelain-modified silicone rubber fiber comprises the following steps: (1) Weighing a certain amount of methyl chlorosilicone rubber, heating it and immersing it in a cyclohexane solvent for pretreatment to obtain pretreated methyl chlorosilicone rubber; (2) dissolving triethoxysilane in ethanol, adding ammonia water to adjust the pH to 4-5 to form a hydrolysis solution, adding polydimethylsiloxane to the hydrolysis solution, ultrasonically treating to disperse the polydimethylsiloxane, then adding nano-sized aluminum nitride, stirring to disperse the aluminum nitride, and obtaining a silicon nitride precursor solution; (3) mixing the silicon nitride precursor solution obtained in step (2) with the pretreated methylchlorosilicone rubber obtained in step (1), heating the mixture to 250-300° C., and keeping the mixture to react for 2-4 hours to obtain a ceramic precursor material; (4) heating the ceramic precursor material to a molten state, extruding it to form fibers, cooling and solidifying the fibers, and then sintering the fibers at a high temperature; (5) Spraying a silicon dioxide dispersion onto the surface of the fiber after sintering in step (4), and drying and curing to obtain the high temperature porcelainizable modified silicone rubber fiber.

8. The method for manufacturing a fiber-reinforced resin composite pipe according to claim 7, characterized in that: The pretreatment process of step (1) is specifically as follows: heating the methyl chlorosilicone rubber to 80-100° C., immersing the heated methyl chlorosilicone rubber in a cyclohexane solvent, stirring for 10-15 minutes, and then vacuum degassing to obtain the pretreated methyl chlorosilicone rubber.

9. The method for manufacturing a fiber-reinforced resin composite pipe according to claim 7, characterized in that: In step (2), the mass ratio of triethoxysilane to ethanol is 1:3-5, the mass of polydimethylsiloxane is 1-5% of the hydrolysis solution, and the mass of nano-aluminum nitride is 2-5% of the hydrolysis solution; in step (3), the mass ratio of silicon nitride precursor solution to pretreated methylchlorosilicone rubber is 1:5-10; in step (4), the temperature of high-temperature sintering is 1200-1300° C. and maintained for 2-3 hours; in step (5), the silicon dioxide dispersion is a mixture of silicon dioxide and ethanol, and the mass volume ratio of silicon dioxide to ethanol is 1g:5-10mL.

10. A fiber-reinforced resin composite pipe manufacturing device based on image evaluation, characterized in that: The device comprises: An inner layer braiding unit, used for braiding an inner layer of the composite tube by using a fiber braiding technology; An intermediate layer braiding unit is used to braid the inner layer surface of the composite pipe by using a fiber braiding technology to obtain an intermediate layer of the composite pipe; An outer layer braiding unit is used to braid the outer layer of the composite pipe by using a fiber braiding technology on the surface of the middle layer of the composite pipe; Image evaluation unit for image evaluation of the composite pipe surface during braiding of the inner layer, the middle layer and the outer layer: Using an image acquisition device to acquire images of the current circumferential surface of the composite pipe in sections to obtain a surface image of the composite pipe; Performing image preprocessing on the collected composite tube surface image; the image preprocessing includes image correction, image stitching and grayscale conversion; The preprocessed composite tube surface image is slid along preset directions by a sliding window method to obtain a plurality of image segments; the preset directions include a first direction corresponding to the axial direction of the composite tube and a second direction corresponding to the circumferential direction of the composite tube; Calculate the similarity of adjacent image segments based on the relative entropy algorithm: ； in is the similarity between adjacent image segments P and Q; is the distribution probability of the image segment P; is the distribution probability of image segment Q; Obtaining a similarity matrix of the composite tube surface image according to the similarity calculation result; The uniformity of the current circumferential surface of the composite pipe and the abnormal position are determined based on the similarity matrix.

Images