A bending angle detection method and system for pipe bending processing and an electronic device

Abstract

The application discloses a bending angle detection method for elbow pipe machining, and comprises the following steps: binarizing and extracting a region of interest from a to-be-recognized image to obtain two straight line segment region binarization images; the to-be-recognized image is an image of a machined elbow pipe; and the bending angle is calculated according to the two straight line segment region binarization images. The application provides an image-based bending angle online measurement method and system, and has the advantages of low cost, high efficiency and strong versatility.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pipe bending processing, in particular to a bending angle detection method and system for pipe bending processing and an electronic device. BACKGROUND

[0002] A pipe bending machine is a common device for bending metal pipes. Due to factors such as material springback, the actual bending angle is not accurate, the product quality is unstable, and the pass rate of the pipe bending product is reduced.

[0003] By using online measurement technology to detect the pipe bending manufacturing process and obtain the actual bending angle after each bending, the pipe bending product detection efficiency can be effectively improved. By connecting the detection system with the control system, effective data can be provided for timely adjustment of process parameters, thereby effectively improving the quality and pass rate of the pipe bending product and reducing costs.

[0004] In the prior art, the quality of the pipe bending product is usually detected offline according to quality control procedures, which is low in efficiency and high in cost. The processing process cannot be detected online, the production quality trend is unknown, and the unqualified rate is high. With the development of image processing and computer vision technology, some methods for measuring bending angles using optical imaging technology have been proposed in the field of pipe bending forming. For example, Katona et al. (Procedia CIRP, 2016) used a three-dimensional laser scanner to measure the bending angle of the pipe without contact. However, this device is high in cost and time-consuming in measurement, and is difficult to meet the needs of online detection of pipe bending processing. Simonetto et al. (INT J ADV MANUF TECHNOL, 2021) proposed a method of embedding an inertial measurement unit (IMU) into a mandrel for springback measurement. However, this method is only suitable for pipe bending processing using a mandrel. SUMMARY

[0005] The purpose of the present application is to provide a bending angle detection method and system for pipe bending processing, which is low in cost, high in efficiency, and strong in universality.

[0006] To achieve the above-mentioned purpose, the present application provides the following solutions:

[0007] A bending angle detection method for pipe bending processing, the detection method comprising the following steps:

[0008] The binaryzation and the region of interest extraction are performed on the to-be-identified image to obtain two straight line segment region binaryzation images; the to-be-identified image is an image of a processed pipe;

[0009] The bending angle is calculated according to the binaryzation images of the two straight line segment regions.

[0010] Optionally, the binarization and the region of interest extraction on the image to be identified are performed to obtain two straight line segment region binarization images, and specifically include:

[0011] The image to be identified is matched with a sample image in a bent pipe processing image database to obtain a matching transformation matrix; the bent pipe processing image database includes sample images obtained after each bending and translation in the bent pipe processing process, and the sample images include a reference image, a labeled bent pipe region of interest, and two straight line segment regions adjacent to the current bending;

[0012] The bent pipe region of interest and the two straight line segment regions adjacent to the current bending in the sample image are subjected to coordinate transformation by using the transformation matrix to obtain the bent pipe region of interest and the two straight line segment regions adjacent to the current bending in the image to be identified;

[0013] The bent pipe region of interest in the image to be identified is cut from the image to be identified to obtain a bent pipe region of interest image, and the bent pipe region of interest image is binarized to obtain a bent pipe region of interest binarization image;

[0014] The two straight line segment regions adjacent to the current bending in the image to be identified are cut from the bent pipe region of interest binarization image to obtain two straight line segment region binarization images.

[0015] Optionally, the bending angle is calculated according to the two straight line segment region binarization images, and specifically includes:

[0016] The two straight line segment region images are thinned;

[0017] The two thinned pixel coordinates are subjected to straight line fitting to obtain straight line equations of the two line segments;

[0018] The included angle between the two line segments is calculated according to the straight line equations of the two line segments to obtain the bending angle.

[0019] A bending angle detection system for bent pipe processing, the detection system includes:

[0020] An image processing module is configured to binarize and extract a region of interest from an image to be identified to obtain two straight line segment region binarization images; the image to be identified is an image of a processed bent pipe;

[0021] A bending angle analysis module is configured to calculate a bending angle according to the two straight line segment region binarization images.

[0022] Optionally, the image processing module includes:

[0023] The registration unit is configured to register a to-be-identified image with a sample image in a bending pipe processing image database to obtain a registration transformation matrix; the to-be-identified image is an image of a processed bending pipe, and the bending pipe processing image database includes sample images obtained after each bending and translation in a bending pipe processing process, and the sample images include a reference image, a labeled bending pipe region of interest, and two straight line segment regions adjacent to a current bending;

[0024] The coordinate transformation unit is configured to perform coordinate transformation on the bending pipe region of interest and the two straight line segment regions adjacent to the current bending of the sample image by using the transformation matrix to obtain the bending pipe region of interest and the two straight line segment regions adjacent to the current bending in the to-be-identified image.

[0025] The bending pipe region of interest image extraction unit is configured to cut the bending pipe region of interest in the to-be-identified image from the to-be-identified image to obtain a bending pipe region of interest image.

[0026] The binarization unit is configured to binarize the bending pipe region of interest image to obtain a bending pipe region of interest binarization image.

[0027] The two straight line segment region binarization image extraction unit is configured to cut the two straight line segment regions adjacent to the current bending in the to-be-identified image from the bending pipe region of interest binarization image to obtain a two straight line segment region binarization image.

[0028] Optionally, the bending angle analysis module specifically includes:

[0029] The thinning unit is configured to thin the two straight line segment region images.

[0030] The line fitting unit is configured to perform line fitting on the two thinned pixel coordinates to obtain line equations of the two line segments.

[0031] The bending angle calculation unit is configured to calculate an included angle between the two line segments according to the line equations of the two line segments to obtain a bending angle.

[0032] An electronic device includes a memory disposed at each participant, a processor disposed at each participant, and a computer program stored on the memory and executable on the processor, and the processor implements the detection method when executing the computer program.

[0033] An online detection method for bending pipe processing, the online detection method including the following steps:

[0034] Obtaining a control signal of bending pipe processing from a control end of a bending pipe processing device;

[0035] determining whether the image acquisition condition is met according to the control signal; the image acquisition condition is that two adjacent control signals are bending and translation, and the translation movement ends;

[0036] when the image acquisition condition is not met, returning to the step of obtaining the control signal of the bending pipe machining from the control end of the bending pipe machining device;

[0037] when the image acquisition condition is met, acquiring the image of the machined bending pipe as a to-be-recognized image;

[0038] using the above detection method, the bending angle is determined based on the to-be-recognized image;

[0039] the bending angle is compared with the bending angle required by the bending pipe machining process to obtain a bending angle error;

[0040] the bending error is fed back to the control system of the bending pipe machining device for adjustment of the machining parameters.

[0041] Optionally, the image of the machined bending pipe is acquired as the to-be-recognized image, which specifically includes:

[0042] the position of the optical camera is adjusted by using an electric control base, so that the optical camera can acquire an image containing the machined bending pipe;

[0043] the optical camera is controlled to acquire the image of the machined bending pipe as the to-be-recognized image.

[0044] An online detection system for bending pipe machining, the online detection system comprises an online control module, an image acquisition module, an image processing module, a bending angle analysis module and an error feedback module;

[0045] the online control module is used to perform the following steps:

[0046] obtaining the control signal of the bending pipe machining from the control end of the bending pipe machining device;

[0047] determining whether the image acquisition condition is met according to the control signal; the image acquisition condition is that two adjacent control signals are bending and translation, and the translation movement ends;

[0048] the image acquisition module is used to acquire the image of the machined bending pipe as a to-be-recognized image when the image acquisition condition is met;

[0049] the image processing module is used to perform binarization and region of interest extraction on the to-be-recognized image to obtain two straight line segment region binarization images; the to-be-recognized image is the image of the machined bending pipe;

[0050] the bending angle analysis module is used to calculate the bending angle according to the binarization images of the two straight line segments;

[0051] The error feedback module is configured to compare the bending angle with a bending angle required by a bending pipe processing technology to obtain a bending angle error, and feed back the bending error to a control system of the bending pipe processing equipment to adjust processing parameters.

[0052] According to the specific embodiments of the present application, the following technical effects are disclosed:

[0053] The present application discloses a bending angle detection method for bending pipe processing, which comprises the following steps: binarizing and extracting a region of interest from a to-be-identified image to obtain two straight line segment region binarization images; the to-be-identified image is an image of a processed bending pipe; and a bending angle is calculated according to the two straight line segment region binarization images. The present application provides an image-based bending angle online measurement method and system, which has the advantages of low cost, high efficiency and strong versatility. BRIEF DESCRIPTION OF DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0055] Figure 1 The flowchart of image processing provided for the embodiments of the present application is provided.

[0056] Figure 2 The flowchart of bending degree analysis provided for the embodiments of the present application is provided.

[0057] Figure 3 The flowchart of the online detection method for bending pipe processing provided for the embodiments of the present application is provided.

[0058] Figure 4 The structure diagram of the online detection system for bending pipe processing provided for the embodiments of the present application is provided.

[0059] Figure 5 The flowchart of image acquisition provided for the embodiments of the present application is provided. DETAILED DESCRIPTION

[0060] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments only constitute some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0061] The application aims to provide a bending angle detection method and system for pipe bending processing, which is low in cost, high in efficiency and strong in versatility.

[0062] In order to make the above-mentioned purposes, features and advantages of the application more obvious and easy to understand, the application will be further described in detail below in combination with the drawings and specific embodiments.

[0063] Embodiment 1

[0064] The application provides a bending angle detection method for pipe bending processing, which comprises the following steps:

[0065] The to-be-recognized image is binarized and the region of interest is extracted to obtain two straight line segment region binarization images; the to-be-recognized image is an image of a processed pipe.

[0066] The bending angle is calculated according to the two straight line segment region binarization images.

[0067] There are many ways to binarize and extract the region of interest of the to-be-recognized image, such as edge detection, neural network model, etc., but none of them is suitable for the extraction of the straight line segment region in pipe bending processing. Figure 1 As shown in the figure, in the embodiment of the application, the to-be-recognized image is binarized and the region of interest is extracted to obtain two straight line segment region binarization images, which specifically comprises:

[0068] The to-be-recognized image is registered with a sample image in a pipe bending image database to obtain a registration transformation matrix; the pipe bending image database comprises sample images obtained after each bending and translation in the pipe bending process, and the sample images comprise a reference image, a labeled region of interest of the pipe and two straight line segment regions adjacent to the current bending.

[0069] The region of interest of the pipe and the two straight line segment regions adjacent to the current bending of the sample image are subjected to coordinate transformation by using the transformation matrix to obtain the region of interest of the pipe and the two straight line segment regions adjacent to the current bending in the to-be-recognized image.

[0070] The region of interest of the pipe in the to-be-recognized image is cut from the to-be-recognized image to obtain a region of interest of the pipe image, and the region of interest of the pipe image is binarized to obtain a region of interest of the pipe binarization image.

[0071] The two straight line segment regions adjacent to the current bending in the to-be-recognized image are cut from the region of interest of the pipe binarization image to obtain two straight line segment region binarization images.

[0072] As shown in the figure, Figure 2As shown, the bending angle is calculated according to the two straight line segment region binary images, and specifically includes:

[0073] The two straight line segment region images are thinned.

[0074] The two thinned pixel coordinates are fitted with straight lines to obtain the straight line equations of the two line segments.

[0075] The included angle between the two line segments is calculated according to the straight line equations of the two line segments to obtain the bending angle.

[0076] The embodiment 1 of the present application also provides a bending angle detection system for bending pipe machining, and the detection system comprises:

[0077] An image processing module is configured to binarize a to-be-recognized image and extract a region of interest to obtain two straight line segment region binary images; the to-be-recognized image is an image of a machined bending pipe.

[0078] The image processing module comprises:

[0079] A registration unit is configured to register the to-be-recognized image with a sample image in a bending pipe machining image database to obtain a registration transformation matrix; the to-be-recognized image is an image of a machined bending pipe, and the bending pipe machining image database comprises sample images obtained after each bending and translation in a bending pipe machining process, and the sample images comprise a reference image, a labeled bending pipe region of interest, and two straight line segment regions adjacent to a current bending;

[0080] A coordinate transformation unit is configured to perform coordinate transformation on the bending pipe region of interest and the two straight line segment regions adjacent to the current bending of the sample image by using the transformation matrix to obtain the bending pipe region of interest and the two straight line segment regions adjacent to the current bending in the to-be-recognized image.

[0081] A bending pipe region of interest image extraction unit is configured to cut the bending pipe region of interest in the to-be-recognized image from the to-be-recognized image to obtain a bending pipe region of interest image.

[0082] A binarization unit is configured to binarize the bending pipe region of interest image to obtain a bending pipe region of interest binary image.

[0083] A two straight line segment region binary image extraction unit is configured to cut the two straight line segment regions adjacent to the current bending in the to-be-recognized image from the bending pipe region of interest binary image to obtain two straight line segment region binary images.

[0084] A bending angle analysis module is configured to calculate a bending angle according to the two straight line segment region binary images.

[0085] The bending angle analysis module specifically comprises:

[0086] a thinning unit configured to thin the two straight line segment region images.

[0087] a line fitting unit configured to fit lines to the two thinned pixel coordinates to obtain line equations of the two line segments.

[0088] a bend angle calculation unit configured to calculate an included angle between the two line segments according to the line equations of the two line segments to obtain a bend angle.

[0089] Embodiment 1 also provides an electronic device, comprising a memory disposed at each participant, a processor disposed at each participant, and a computer program stored on the memory and executable on the processor, wherein the processor implements the detection method when executing the computer program.

[0090] Embodiment 2

[0091] Embodiment 2 provides an online detection method for bend processing, as shown in the following figure, the online detection method comprises the following steps: Figure 3

[0092] obtaining a control signal of bend processing from a control end of a bend processing device.

[0093] judging whether the image acquisition condition is met according to the control signal; the image acquisition condition is that adjacent two control signals are bending and translation, and the translation movement ends.

[0094] when the image acquisition condition is not met, returning to the step of obtaining the control signal of bend processing from the control end of the bend processing device.

[0095] when the image acquisition condition is met, collecting an image of the processed bend pipe as a to-be-recognized image,

[0096] using the above detection method, determining a bend angle based on the to-be-recognized image,

[0097] comparing the bend angle with a bend angle required by a bend processing process to obtain a bend error,

[0098] feeding back the bend error to a control system of the bend processing device to adjust processing parameters.

[0099] wherein the image of the processed bend pipe is collected as the to-be-recognized image, and specifically comprises:

[0100] adjusting a position of an optical camera by using an electrically controlled base so that the optical camera can obtain an image containing the processed bend pipe;

[0101] obtaining a control signal of bend processing from a control end of a bend processing device.​

[0102] Determine if the conditions for image acquisition are met. If two adjacent control signals are bending and translation, and the translation movement has ended, then the conditions for image acquisition are met; otherwise, the conditions for image acquisition are not met.

[0103] A control signal for image acquisition is generated. If the conditions for image acquisition are met, a control signal for image acquisition is generated.

[0104] Perform image acquisition and save the image to the storage medium.

[0105] This invention also provides an online inspection system for pipe bending processes, such as... Figure 4 As shown, the online detection system includes: an online control module, an image acquisition module, an image processing module, a bending angle analysis module, and an error feedback module. The online control module receives control signals from the pipe bending equipment, performs logical discrimination on the control signals, generates control signals for image acquisition, and sends them to the image acquisition module. The image acquisition module acquires images of the pipe bending process. The image processing module processes the acquired images to obtain images of the current bending area. The bending angle analysis module processes the images of the region of interest to obtain the bending angle and bending angle error, and sends them to the control terminal of the pipe bending equipment.

[0106] like Figure 5 As shown, the image acquisition process specifically includes:

[0107] The position of the optical camera is adjusted using an electronically controlled base, enabling the optical camera to acquire images including those of the processed bent pipe.

[0108] The control signal for pipe bending is obtained from the control terminal of the pipe bending equipment.

[0109] Determine if the conditions for image acquisition are met. If two adjacent control signals are bending and translation, and the translation movement has ended, then the conditions for image acquisition are met; otherwise, the conditions for image acquisition are not met.

[0110] A control signal for image acquisition is generated. If the conditions for image acquisition are met, a control signal for image acquisition is generated.

[0111] Perform image acquisition and save the image to the storage medium.

[0112] like Figure 1 As shown, the image processing procedure includes the following steps:

[0113] The current image is registered with images in the labeled pipe bending image library to obtain the optimal labeled reference image and image registration transformation matrix.

[0114] According to the transformation matrix and the labeled reference image, a region of interest of the bent pipe and two straight line regions in the current acquisition image are obtained.

[0115] The image of the region of interest of the bent pipe in the current acquisition image is binarized, and two binarized images of the two straight line regions are obtained according to the two straight line regions in the current acquisition image.

[0116] As shown in the figure, Figure 2 The bending angle analysis includes the following steps:

[0117] The binarized images of the two straight line regions are thinned;

[0118] The two straight line equations are obtained by fitting the two straight line regions after thinning.

[0119] The included angle between the two straight lines, i.e., the bending angle, is calculated according to the two straight line equations.

[0120] The bending angle error is calculated by comparing the calculated bending angle with the bending angle required by the processing technology.

[0121] The embodiment of the present application can effectively improve the quality of the product and timely grasp the bending pipe processing condition by acquiring the image after each bending, registering the image with the image in the image database, determining the two line segment regions adjacent to the bending region, obtaining the straight line equations of the two line segments by using image processing, and then obtaining the actual bending angle, which has the characteristics of low cost, high efficiency and strong universality.

[0122] Each embodiment in the specification is described in a progressive manner, and each embodiment focuses on the difference from other embodiments, and the same or similar parts between each embodiment can be referred to each other.

[0123] The principles and implementation modes of the present application are described by applying specific examples in this paper, and the above embodiment description is only used to help understand the method and core idea of the present application; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range will be changed. In conclusion, the content of the specification should not be understood as a limitation of the present application.

Claims

1. An online inspection method for pipe bending processes, characterized in that, The online detection method includes the following steps: The control signal for pipe bending is obtained from the control terminal of the pipe bending equipment; The control signal is used to determine whether the image acquisition conditions are met; the image acquisition conditions are that two adjacent control signals are bending and translation, and the translation movement ends. If the image acquisition conditions are not met, return to the step "Obtain the control signal for pipe bending from the control terminal of the pipe bending equipment"; When the image acquisition conditions are met, an image of the processed bent pipe is acquired as the image to be identified. A bending angle detection method for pipe bending is adopted to determine the bending angle based on the image to be identified; The bending angle is compared with the bending angle required by the pipe bending process to obtain the bending angle error; The bending angle error is fed back to the control system of the pipe bending equipment for adjustment of processing parameters; The detection method includes the following steps: The image to be identified is binarized and the region of interest is extracted to obtain two binarized images of straight line segment regions; the image to be identified is an image of a processed bent pipe. Calculate the bending angle based on the binarized images of the two straight line segment regions; The process of binarizing the image to be recognized and extracting the region of interest to obtain two binary images of line segment regions specifically includes: The image to be identified is registered with the sample images in the pipe bending processing image database to obtain a registration transformation matrix; the pipe bending processing image database includes sample images obtained after each bend and translation during the pipe bending process, and the sample images include: a reference image, a labeled region of interest, and two straight line segment regions adjacent to the current bend; The region of interest in the bend and the two line segments adjacent to the current bend in the sample image are transformed using a transformation matrix to obtain the region of interest in the bend and the two line segments adjacent to the current bend in the image to be identified. The region of interest (ROI) is extracted from the image to be identified to obtain an image of the ROI. The image of the ROI is then binarized to obtain a binarized image of the ROI. From the binarized image of the region of interest bend, extract the two adjacent straight line segments from the image to be identified, and obtain the two binarized images of the straight line segment regions.

2. The online inspection method for pipe bending processing according to claim 1, characterized in that, The step of calculating the bending angle based on the binarized images of two straight line segment regions specifically includes: Thin the image of the two line segment regions; By fitting a straight line to the two thinned pixel coordinates, the equations of the two line segments are obtained. The angle between the two line segments is calculated by using their straight-line equations, thus obtaining the bending angle.

3. The online inspection method for pipe bending processing according to claim 1, characterized in that, The acquisition of images of the processed bent pipes, used as images to be identified, specifically includes: The position of the optical camera is adjusted using an electronically controlled base, enabling the optical camera to acquire images including those of the bent pipe being processed. The optical camera is controlled to acquire images of the processed bent pipe, which are then used as images to be identified.

4. An online inspection system for pipe bending processes, characterized in that, The online detection system includes: an online control module, an image acquisition module, an image processing module, a bending angle analysis module, and an error feedback module; The online control module is used to perform the following steps: The control signal for pipe bending is obtained from the control terminal of the pipe bending equipment; The control signal is used to determine whether the image acquisition conditions are met; the image acquisition conditions are that two adjacent control signals are bending and translation, and the translation movement ends. The image acquisition module is used to acquire an image of the processed bent pipe when the image acquisition conditions are met, and use it as the image to be identified. The image processing module is used to binarize the image to be recognized and extract the region of interest to obtain two binarized images of straight line segment regions; the image to be recognized is an image of a processed bent pipe; The bending angle analysis module is used to calculate the bending angle based on the binarized images of the two straight line segment regions. The error feedback module is used to compare the bending angle with the bending angle required by the pipe bending process to obtain the bending angle error; and to feed the bending angle error back to the control system of the pipe bending equipment for adjustment of the processing parameters. The image processing module includes: The registration unit is used to register the image to be identified with the sample images in the pipe bending processing image database to obtain a registration transformation matrix; the image to be identified is an image of a pipe bending process, and the pipe bending processing image database includes sample images obtained after each bend and translation during the pipe bending process, and the sample images include: a reference image, a labeled region of interest, and two straight line segment regions adjacent to the current bend; The coordinate transformation unit is used to perform coordinate transformation on the region of interest (ROI) and the two line segments adjacent to the current bend in the sample image using a transformation matrix, so as to obtain the region of interest (ROI) and the two line segments adjacent to the current bend in the image to be identified. The region of interest (ROI) image extraction unit is used to extract the ROI region from the image to be identified and obtain the ROI region image. The binarization unit is used to binarize the image of the region of interest bend to obtain a binarized image of the region of interest bend. The two line segment region binarized image extraction unit is used to extract two adjacent line segment regions from the binarized image of the bend region of interest in the image to be identified, and obtain two line segment region binarized images.

5. The online inspection system for pipe bending processing according to claim 4, characterized in that, The bending angle analysis module specifically includes: The thinning unit is used to thin the image of two line segment regions; The line fitting unit is used to perform line fitting on the coordinates of two thinned pixels to obtain the line equations of the two line segments. The bending angle calculation unit is used to calculate the included angle between two line segments based on their straight line equations, thus obtaining the bending angle.

6. An electronic device, characterized in that, It includes a memory deployed in each participating party, a processor deployed in each participating party, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the online detection method for pipe bending processing as described in any one of claims 1 to 3.

Images