Method and system for processing pipe length detection data

Abstract

The present application provides a method and system for processing rolled pipe length detection data. The detection equipment obtains the feature data set and environmental data of the rolled pipe to be detected, and calibrates the feature data set according to the environmental data to obtain the calibrated feature data set, and sends the calibrated feature data set to the edge device. The edge device performs intelligent analysis and processing on the calibrated feature data set to generate a target detection result. Through the above method, laser radar, depth camera and ultrasonic sensor are used to collect rolled pipe data, and intelligent processing is performed by edge device to obtain accurate length value. It has the advantages of high-precision measurement, diversified data collection, real-time and rapidity, intelligent processing, improved production efficiency and product quality, environmental adaptability and non-contact detection. These advantages together ensure the accuracy and reliability of rolled pipe length detection, and provide strong support for the stable operation of the steel rolling production line and product quality control.

Description

Technical Field

[0001] The present application relates to the field of automation control technology, and in particular to a method and system for processing rolled tube length detection data. Background Art

[0002] Rolled pipes are manufactured by rolling metal bars or billets on a rolling mill. With the continuous advancement of technology, rolled pipes have become essential structural components in industry and construction. To adapt to different applications, rolled pipes can be manufactured using hot and cold rolling methods, and can be rolled into various shapes based on demand. Length testing is crucial to ensure the stability of rolled pipes in various applications.

[0003] In the prior art, the length of rolled tubes is usually detected manually, and on the production line, it is usually detected by a camera or an encoder.

[0004] However, the detection data of the above method is easily interfered by external environmental factors, thereby affecting the accuracy. Summary of the Invention

[0005] The present application provides a method and system for processing rolled tube length detection data, which are used to solve the problems of inaccurate and lack of flexibility in rolled tube length detection in the prior art.

[0006] In a first aspect, the present application provides a method for processing rolled tube length detection data, which is applied to a detection system, wherein the detection system includes a detection device and an edge device, wherein the detection device and the edge device are communicatively connected; the method includes:

[0007] The detection device acquires a feature data set and environmental data of the rolled tube to be detected, wherein the feature data set includes point cloud data, image data and ultrasonic data;

[0008] The detection device calibrates the feature data set according to the environmental data to obtain a calibrated feature data set, and sends the calibrated feature data set and the environmental data to the edge device;

[0009] The edge device performs intelligent analysis and processing on the calibrated feature data set to generate a target detection result, which includes a length value of the rolled pipe to be detected.

[0010] In combination with the first aspect, in some embodiments, the detection device calibrates the feature dataset according to the environmental data to obtain a calibrated feature dataset, including:

[0011] The detection device calibrates the point cloud data according to the environmental data to obtain calibrated point cloud data;

[0012] The detection device calibrates the image data according to the environmental data to obtain calibrated image data;

[0013] The detection device calibrates the ultrasonic data according to the environmental data to obtain calibrated ultrasonic data;

[0014] The detection device aggregates the calibrated point cloud data, the calibrated image data, and the calibrated ultrasonic data to obtain the calibrated feature data set.

[0015] In combination with the first aspect, in some embodiments, the detection device calibrates the point cloud data according to the environmental data to obtain the calibrated point cloud data, including:

[0016] The detection device extracts data from the environmental data based on the point cloud data to generate first environmental data, where the first environmental data is environmental data that is correlated with the point cloud data;

[0017] The detection device extracts the number of point clouds related to the length of the rolled pipe to be detected from the point cloud data as target point cloud data;

[0018] The detection device modifies a preset first pipe rolling compensation formula according to the first environmental data and characteristics of the sensor that collects the point cloud data to obtain a point cloud compensation formula;

[0019] The detection device calculates the target point cloud data and the first environmental data based on the point cloud compensation formula to obtain the calibrated point cloud data.

[0020] In combination with the first aspect, in some embodiments, the detection device calibrates the image data according to the environmental data to obtain the calibrated image data, including:

[0021] The detection device extracts data from the environmental data based on the image data to generate second environmental data, where the second environmental data is environmental data that is correlated with the image data;

[0022] The detection device modifies a preset second pipe rolling compensation formula according to the second environmental data and the image data to obtain an image compensation formula;

[0023] The detection device calculates the image data and the second environmental data based on the image compensation formula to obtain the calibrated image data.

[0024] In combination with the first aspect, in some embodiments, the detection device calibrates the ultrasonic data according to the environmental data to obtain the calibrated ultrasonic data, including:

[0025] The detection device extracts data from the environmental data based on the ultrasonic data to generate third environmental data, where the third environmental data is environmental data correlated with the ultrasonic data;

[0026] The detection device modifies a preset third pipe rolling compensation formula according to the third environmental data and the ultrasonic data to obtain an ultrasonic compensation formula;

[0027] The detection device calculates the ultrasonic data and the third environment data based on the ultrasonic compensation formula to obtain the calibrated ultrasonic data.

[0028] In conjunction with the first aspect, in some embodiments, the edge device performs intelligent analysis and processing on the calibrated feature data set to generate a target detection result, including:

[0029] The edge device performs preprocessing and feature extraction on the calibrated feature data set to obtain a feature set;

[0030] The edge device performs feature fusion processing on the feature set to obtain a fused feature vector;

[0031] The edge device performs intelligent calculation processing on the fused feature vector to obtain the target detection result.

[0032] In combination with the first aspect, in some embodiments, the edge device performs feature fusion processing on the feature set to obtain a fused feature vector, including:

[0033] The edge device performs normalization processing on the feature set to obtain a feature vector set;

[0034] The edge device performs weight calculation on each feature vector in the feature vector set to obtain a feature weight corresponding to each feature vector;

[0035] The edge device performs a fusion calculation on the feature vector set based on the feature weight corresponding to each feature vector and a preset fusion formula to obtain the fused feature vector.

[0036] In combination with the first aspect, in some embodiments, the edge device performs intelligent calculation processing on the fused feature vector to obtain the target detection result, including:

[0037] The edge device calculates the fused feature vector based on a preset calculation model to obtain the target detection result.

[0038] In a second aspect, the present application provides a detection device comprising: a processor, a memory communicatively connected to the processor, a laser radar, a depth camera, an ultrasonic sensor, an environmental sensor, and a communication interface;

[0039] The memory stores computer-executable instructions;

[0040] The processor executes the computer-executable instructions stored in the memory to implement the method for processing the pipe length detection data according to any one of the first aspects.

[0041] In a third aspect, the present application provides a detection system, comprising a detection device and an edge device, wherein the detection device and the edge device are communicatively connected; and configured to execute the method for processing rolled pipe length detection data as described in any one of the first aspects.

[0042] In a fourth aspect, the present application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method for processing pipe length detection data described in any of the aforementioned aspects.

[0043] In a fifth aspect, the present application provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the method for processing the pipe length detection data described in any of the above aspects.

[0044] The present application provides a method and system for processing rolled pipe length detection data. The detection equipment obtains the feature data set and environmental data of the rolled pipe to be detected, and calibrates the feature data set according to the environmental data to obtain a calibrated feature data set. The calibrated feature data set is sent to the edge device, and the edge device performs intelligent analysis and processing on the calibrated feature data set to generate a target detection result. Through the above method, the rolled pipe data is collected using a lidar, a depth camera, and an ultrasonic sensor, and intelligent processing is performed by the edge device to obtain an accurate length value. It has the advantages of high-precision measurement, diversified data collection, real-time and rapidity, intelligent processing, improved production efficiency and product quality, environmental adaptability, and non-contact detection. These advantages together ensure the accuracy and reliability of rolled pipe length detection, and provide strong support for the stable operation of the steel rolling production line and product quality control. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0046] Figure 1A flowchart of a first embodiment of a method for processing pipe length detection data provided in an embodiment of the present application;

[0047] Figure 2 A flow chart of a second embodiment of a method for processing pipe length detection data provided in an embodiment of the present application;

[0048] Figure 3 A flowchart of a third embodiment of a method for processing pipe length detection data provided in an embodiment of the present application;

[0049] Figure 4 A flowchart of a fourth embodiment of a method for processing pipe length detection data provided in an embodiment of the present application;

[0050] Figure 5 A flowchart of a fifth embodiment of a method for processing pipe length detection data provided in an embodiment of the present application;

[0051] Figure 6 A flowchart of a sixth embodiment of a method for processing pipe length detection data provided in an embodiment of the present application;

[0052] Figure 7 A flowchart of a seventh embodiment of a method for processing pipe length detection data provided in an embodiment of the present application;

[0053] Figure 8 A schematic diagram of the structure of the detection device provided in the embodiment of the present application;

[0054] Figure 9 A schematic diagram of the structure of the detection system provided in an embodiment of the present application.

[0055] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0056] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0057] Rolled pipe refers to a pipe made by rolling metal bars or tube blanks on a rolling mill. With the continuous advancement of science and technology, rolled pipes have gradually become the main structural components in the fields of industry and construction. In order to adapt to different application scenarios, rolled pipes can be divided into two manufacturing methods: hot rolling and cold rolling. They can also be rolled into different shapes according to needs. In order to ensure the stability of rolled pipes in application scenarios, length detection is crucial. In the existing technology, measuring tools such as tape measures and laser rangefinders can be used. This method will have human errors and low efficiency. It has certain limitations and cannot be applied to automated production lines or automated measurement systems. Some modern pipe rolling production lines may be equipped with automated length detection systems, which are detected by sensors or visual systems. This method uses a single sensor for measurement and is easily affected by the environment. Therefore, it lacks certain flexibility and accuracy.

[0058] To address the above-mentioned issues, the present application provides a method and system for processing pipe length detection data, thereby improving the accuracy, intelligence, and flexibility of pipe length detection. Specifically, the prior art uses measuring tools or automated measurement systems to detect pipe length. However, the prior art detection is inaccurate, cannot adapt to different detection scenarios, and is easily affected by the environment. Considering these issues, the inventors investigated whether it is possible to design a detection system that is equipped with a detection device and an edge device. The detection device is used to collect pipe data information. The edge device can perform intelligent processing based on the data collected by the detection device to obtain the accurate length of the pipe. On this basis, other information about the pipe, such as morphology and surface information, can also be obtained. Based on this information, the pipe production line can be inspected to improve production accuracy. At the same time, the detection device is also equipped with multiple sensors, which can compensate for the inaccurate data collected by a single sensor in the prior art that is easily affected by the environment. Based on this, the technical solution of the present application is proposed.

[0059] The processing method for rolled pipe length detection data provided in this application is mainly applicable to rolled pipe detection scenarios, which can be production line detection, post-cutting detection, packaging and pre-delivery detection, etc. The above scenarios include at least detection equipment and edge equipment, and the detection equipment and edge equipment are communicatively connected. The monitoring equipment is equipped with a variety of sensors, such as lidar, depth camera and ultrasonic sensor, as well as environmental sensors. The above sensors work in real time and collect rolled pipe information in real time. The monitoring equipment stores the data collected by the sensor, accurately calibrates the collected data, and sends the calibrated data to the edge device. The edge device performs rolled pipe detection based on the calibrated data, such as length detection, surface defect detection, etc.

[0060] Optionally, the communication method between the detection device and the edge device can be wired communication connection, wireless connection, low-power wide area network connection, serial communication protocol connection, message queue telemetry transmission, Bluetooth, etc., which is not specifically limited in this application.

[0061] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0062] Figure 1 The flowchart of the first embodiment of the method for processing the pipe length detection data provided in the embodiment of the present application is as follows: Figure 1 As shown, the method is applied to a detection system, wherein the detection system includes a detection device and an edge device, and the detection device and the edge device are communicatively connected. The method specifically includes:

[0063] S101: The detection equipment obtains a feature data set and environmental data of the rolled pipe to be detected.

[0064] In this step, in order to accurately detect the length of the rolled pipe in different scenarios and environments, the detection equipment obtains the feature data set and environmental data of the rolled pipe to be detected. The feature data set includes point cloud data, image data and ultrasonic data.

[0065] Specifically, the detection equipment is equipped with laser radar, depth camera, ultrasonic sensor and environmental sensor. The laser radar, depth camera, ultrasonic sensor and environmental sensor collect data of the rolled pipe to be inspected in real time. Among them, the laser radar collects point cloud data, the depth camera collects image data, the ultrasonic sensor collects ultrasonic data, and the environmental sensor collects environmental data of the actual detection environment, such as temperature and humidity, in real time.

[0066] For example, taking a pipe rolling production line as an example, the detection equipment is installed at the detection position. When the pipe enters the detection area, the detection equipment starts to control different sensors to collect data.

[0067] It should be noted that when the above-mentioned different sensors are working, the pipe rolling data and environmental data collected in real time need to be synchronized to ensure the accuracy of the data.

[0068] S102: The detection device calibrates the feature data set according to the environmental data to obtain a calibrated feature data set, and sends the calibrated feature data set to the edge device.

[0069] In this step, after the detection device obtains the feature data set and environmental data, since different sensors have different sensitivities to environmental factors, in order to improve the accuracy of the data, the data in the feature data set is calibrated according to the environmental data, and the calibrated feature data set is sent to the edge device.

[0070] Specifically, the detection equipment calibrates the point cloud data according to the environmental data to obtain calibrated point cloud data; calibrates the image data according to the environmental data to obtain calibrated image data; calibrates the ultrasonic data according to the environmental data to obtain calibrated ultrasonic data; finally, the calibrated point cloud data, calibrated image data and calibrated ultrasonic data are aggregated to obtain a calibrated feature data set.

[0071] S103: The edge device performs intelligent analysis and processing on the calibrated feature data set to generate target detection results.

[0072] In this step, after receiving the calibrated feature data set sent by the detection device, the edge device performs intelligent analysis on the calibrated feature data set to obtain the accurate length of the rolled pipe to be detected, and generates a target detection result, where the target detection result includes the length value of the rolled pipe to be detected.

[0073] Specifically, the edge device preprocesses and extracts features from the calibrated feature data set to obtain a feature set, then performs feature fusion processing on the feature set to obtain a fused feature vector, and then performs intelligent calculation processing on the fused feature vector to obtain the target detection result.

[0074] Optionally, the target detection results may also include information such as the shape and surface information of the rolled pipe, as well as the rolling speed and rolling pressure. Based on the above information, it is also possible to detect rolled pipe defects and the status of the production line, thereby ensuring the product quality of the rolled pipe. Real-time monitoring of the production line can also be used to avoid production failures and improve the stability and production efficiency of the production line.

[0075] The method for processing rolled pipe length detection data provided in this embodiment is that the detection equipment obtains the feature data set and environmental data of the rolled pipe to be detected, and calibrates the feature data set based on the environmental data to obtain a calibrated feature data set. The calibrated feature data set is sent to the edge device, and the edge device performs intelligent analysis and processing on the calibrated feature data set to generate a target detection result. Through the above method, rolled pipe data is collected using lidar, depth camera and ultrasonic sensor, and intelligent processing is performed by edge device to obtain accurate length value. It has the advantages of high-precision measurement, diversified data collection, real-time and rapid performance, intelligent processing, improved production efficiency and product quality, environmental adaptability, and non-contact detection. These advantages jointly ensure the accuracy and reliability of rolled pipe length detection, providing strong support for the stable operation of the steel rolling production line and product quality control.

[0076] Figure 2 The flowchart of the second embodiment of the method for processing the pipe length detection data provided in the embodiment of the present application is as follows: Figure 2 As shown, based on the above embodiment, step S102 specifically includes:

[0077] S201: The detection device calibrates the point cloud data according to the environmental data to obtain calibrated point cloud data.

[0078] In this step, environmental factors vary significantly depending on the pipe inspection scenario. For example, in a pipe production line, pipe rolling can be divided into hot and cold rolling, resulting in significant variations in ambient temperature. LiDAR is easily affected by environmental data, such as ambient temperature. Temperature changes can affect the dimensions of the LiDAR's internal mechanical structure, potentially causing slight shifts in the laser emission and reception positions, affecting ranging accuracy. For example, rising temperatures can cause the mechanical structure to expand, shifting the laser beam's emission angle or path. High humidity increases the amount of water molecules in the air, and the laser interacts with these molecules as it propagates through the air, potentially attenuating or scattering the laser energy, affecting laser ranging accuracy. Light, such as natural light or other artificial light sources, can interfere with the laser beam, causing the laser ranging device to receive erroneous reflected signals, resulting in errors. Therefore, in order to accurately determine the pipe length from the point cloud data collected by the LiDAR, the point cloud data must be calibrated based on the collected environmental data.

[0079] Specifically, the detection equipment extracts environmental data based on point cloud data to generate first environmental data, then extracts the number of point clouds related to the length of the rolled pipe to be detected from the point cloud data as target point cloud data, and then corrects the preset first pipe rolling compensation formula based on the first environmental data and the characteristics of the sensor that collects the point cloud data to obtain the point cloud compensation formula. Finally, based on the point cloud compensation formula, the target point cloud data and the first environmental data are calculated to obtain calibrated point cloud data.

[0080] S202: The detection device calibrates the image data according to the environmental data to obtain calibrated image data.

[0081] In this step, since the depth camera has high measurement accuracy and is a non-contact detection method, it is an important tool for detecting the length of the rolled pipe. However, the data collected by the depth camera is easily affected by environmental factors, such as lighting conditions and strong light. Depth cameras usually use infrared light or structured light technology to obtain depth information. If there is strong light in the environment (such as direct sunlight), it may interfere with the infrared light or structured light of the camera, resulting in inaccurate depth information. Dim environment: In insufficient light, the camera may not be able to capture enough feature points for depth calculation, thereby affecting the accuracy of length detection; Environmental humidity: A high humidity environment may cause the electronic components inside the camera to become damp, thereby affecting its normal operation. In addition, humidity may also affect the propagation of light, thereby indirectly affecting the accuracy of depth calculation. Therefore, in order to obtain the accurate rolled pipe length based on the image data collected by the depth camera, it is necessary to calibrate the image data based on the collected environmental data to obtain calibrated image data.

[0082] Specifically, the detection equipment extracts environmental data based on the image data to generate second environmental data, and then corrects the preset second pipe rolling compensation formula based on the second environmental data and the image data to obtain an image compensation formula. Finally, based on the image compensation formula, the image data and the second environmental data are calculated to obtain calibrated image data.

[0083] S203: The detection device calibrates the ultrasonic data according to the environmental data to obtain calibrated ultrasonic data.

[0084] In this step, the ultrasonic sensor not only enables non-contact detection, but also has the advantages of being able to penetrate materials of a certain thickness, detect deep objects, and have high measurement accuracy. Therefore, it is suitable for accurate measurement of rolled pipe length. However, the ultrasonic sensor is also affected by environmental factors in the rolled pipe length detection scenario, such as ambient temperature. Changes in ambient temperature will affect the propagation speed of ultrasonic waves, thereby affecting the measurement accuracy of the sensor. Generally, the higher the temperature, the faster the speed of sound, and the faster the propagation speed, resulting in a corresponding decrease in the detection distance; the ambient humidity. A high humidity environment may cause an increase in water molecules in the air. Ultrasonic waves interact with water molecules during propagation, resulting in energy attenuation or scattering, affecting measurement accuracy; sound wave interference. If there are other sound sources in the detection environment (such as machine noise, personnel communication, etc.), they may interfere with the measurement of the ultrasonic sensor, resulting in inaccurate measurement data. Therefore, in order to obtain accurate rolled pipe length based on the ultrasonic data collected by ultrasonic sensing, it is necessary to calibrate the ultrasonic data based on the collected environmental data to obtain calibrated ultrasonic data.

[0085] S204: The detection device aggregates the calibrated point cloud data, the calibrated image data, and the calibrated ultrasonic data to obtain a calibrated feature data set.

[0086] In the method for processing rolled tube length detection data provided in this embodiment, the detection equipment calibrates point cloud data based on environmental data to obtain calibrated point cloud data. The image data is then calibrated based on the environmental data to obtain calibrated image data. Finally, the ultrasonic data is calibrated based on the environmental data to obtain calibrated ultrasonic data. Finally, the calibrated point cloud data, calibrated image data, and calibrated ultrasonic data are aggregated to obtain a calibrated feature data set. By calibrating the data collected by different sensors separately, data accuracy is ensured.

[0087] Figure 3 The flowchart of the third embodiment of the method for processing the pipe length detection data provided in the embodiment of the present application is as follows: Figure 3 As shown, based on the above embodiments, step S201 specifically includes:

[0088] S301: The detection device extracts environmental data based on the point cloud data to generate first environmental data.

[0089] In this step, since the environmental data collected by the environmental sensor may contain multiple types of environmental data, such as temperature, humidity, equipment vibration, dust and dirt, light, electromagnetic interference, and pipe material properties, and the pipe is collected as a feature data set of different sensor types, it is affected differently by different environmental factors. In order to make the data collected by the sensor more accurate, the data collected by different sensors can be personalized calibrated. For the point cloud data collected by the lidar, the environmental data can be extracted based on the point cloud data to obtain the first environmental data, wherein the first environmental data is environmental data that is correlated with the point cloud data, which can be one type or multiple types.

[0090] S302: The inspection device extracts point cloud data related to the length of the rolled pipe to be inspected from the point cloud data as target point cloud data.

[0091] In this step, in order to obtain the length of the rolled pipe more accurately, it is necessary to extract the point cloud data, and extract the point cloud data related to the length of the rolled pipe as the target point cloud data.

[0092] Optionally, according to the shape and characteristics of the rolled tube, a certain threshold or algorithm may be set to extract point cloud data related to the rolled tube length.

[0093] For example, a certain spatial range or height range can be set to filter out point cloud data located in the pipe rolling area.

[0094] The density, intensity and other information of the point cloud data can also be used to further extract the boundary or contour line of the rolled pipe through clustering, segmentation and other algorithms, so as to obtain the length information of the rolled pipe more accurately.

[0095] S303: The detection device modifies the preset first pipe rolling compensation formula according to the first environmental data and the characteristics of the sensor that collects the point cloud data to obtain a point cloud compensation formula.

[0096] In this step, in order to accurately calibrate the target point cloud data, the first pipe rolling compensation formula is pre-set according to the characteristics of the point cloud data, and the first pipe rolling compensation formula is characteristically corrected based on the characteristics of the first environmental data and the lidar, thereby obtaining the point cloud compensation formula.

[0097] Specifically, suppose there are n environmental variables, that is, n types of environmental data, represented as x1, x2, x3...x n , then the first tube rolling compensation formula can be expressed as:

[0098] I corrected = original +(x1, x2, x3, …, x n )

[0099] Among them, I corrected Represents the calibrated point cloud data, I original Represents point cloud data, f(x1, x2, x3, ..., x n ) represents the correction function.

[0100] The above correction function f(x1, x2, x3, ..., x n ) formula can be expressed as:

[0101]

[0102] Among them, a i 、n i 、c i d i 、e ij 、f ij 、g ij ,h,k i Represents the correction coefficient, which reflects the degree of influence of different environmental variables on point cloud data.

[0103] In order to obtain the precise length of the rolled pipe, after extracting the target point cloud data and the first environmental data, the correction function is characteristically corrected based on the first environmental data, thereby obtaining a point cloud compensation formula.

[0104] For example, taking the first environmental data including temperature and humidity as an example, the correction function is corrected, that is, only the linear term and the product term in the correction function are retained, so that the corrected correction function is expressed as follows:

[0105] f(T,H)=1T+a2H+a 12 TH

[0106] Among them, T represents temperature, H represents humidity, a1, a2, a 12 Indicates the correction factor.

[0107] Alternatively, if the rolled tube is made of a specific material and the length-temperature behavior of the material is known (for example, the coefficient of thermal expansion of a certain metal), this property can be incorporated into the correction function to ensure data accuracy:

[0108] f(T,H)=(L0+T)·(a1T+a2H+a 12 TH)

[0109] Wherein, L0 represents the length of the rolled tube at the reference temperature, and α represents the thermal expansion coefficient of the rolled tube material.

[0110] Optionally, the LiDAR itself may also be affected by environmental factors. For example, the propagation speed of laser light in the air may vary with changes in temperature and humidity. Therefore, based on the characteristics of the sensor that collects the point cloud data, the above correction formula is further modified:

[0111]

[0112] in, It shows the relationship between the speed of laser propagation in air and temperature and humidity.

[0113] Then, v(T,H) can be expressed as:

[0114] v(T,H)= std 1+ T ·(TT std )+ H ·(HH std ))

[0115] Among them, v std is the speed of light under standard conditions, T std and H std are the standard values ​​of temperature and humidity, k T and k H Represents the correlation coefficient.

[0116] S304: The detection device calculates the target point cloud data and the first environment data based on the point cloud compensation formula to obtain calibrated point cloud data.

[0117] In this step, after the point cloud compensation formula is obtained through the above steps, the target point cloud data and the first environment data are calculated based on the point cloud compensation formula to obtain calibrated point cloud data.

[0118] Specifically, after modifying the first pipe rolling compensation formula, the point cloud compensation formula is obtained:

[0119]

[0120] Among them, I corrected Represents the calibrated point cloud data, I original represents the target point cloud data, T and H represent the first environment data, and the other symbols have the same meanings as in the previous steps and are not repeated here.

[0121] In the method for processing rolled pipe length detection data provided in this embodiment, a detection device extracts environmental data based on point cloud data to generate first environmental data. Point cloud data related to the length of the rolled pipe to be detected is extracted from the point cloud data and used as target point cloud data. A preset first rolled pipe compensation formula is corrected based on the characteristics of the first environmental data and the sensor that collects the point cloud data to obtain a point cloud compensation formula. Based on the point cloud compensation formula, calculations are performed on the target point cloud data and the first environmental data to obtain calibrated point cloud data. By designing a compensation formula and correcting the compensation formula based on the characteristics of the point cloud data, the accuracy of the data correction is ensured, and a data foundation is provided for subsequent rolled pipe length detection.

[0122] Figure 4 The flowchart of the fourth embodiment of the method for processing the pipe length detection data provided in the embodiment of the present application is as follows: Figure 4 As shown, based on the above embodiments, step S202 specifically includes:

[0123] S401: The detection device extracts environmental data based on the image data to generate second environmental data.

[0124] This step is the same as step S301 in the aforementioned embodiment, and extracts environmental data that is correlated with the image data from the environmental data as the second environmental data.

[0125] S402: The detection device modifies the preset second pipe rolling compensation formula according to the second environment data and the image data to obtain an image compensation formula.

[0126] In this step, based on the characteristics of the depth camera, a second tube rolling compensation formula is pre-designed:

[0127]

[0128] Among them, I`(x,y) represents the corrected pixel value, I(x,y) represents the pixel value of the image data at the coordinate (x,y), α i (x i ) is the same as the i-th environmental data x i Related compensation coefficient functions, max(I) and min(I) represent the maximum and minimum pixel values, respectively, which are used to normalize image pixels to the range of [0,1], and β represents the global offset, which is used to adjust the overall brightness or contrast.

[0129] In order to obtain the precise length of the rolled pipe, after extracting the second environmental data, the second pipe rolling compensation formula is corrected based on the second environmental data, thereby obtaining an image compensation formula.

[0130] Optionally, taking the second environmental data including light, temperature, humidity and dust as an example, the second pipe rolling compensation formula is modified to obtain:

[0131]

[0132] Among them, L represents light intensity, T represents temperature, H represents humidity, D represents dust quantization value, α L (L), α T (T), α H (H), α D (D) Compensation coefficient functions corresponding to the quantized values ​​of light, temperature, humidity, and dust, respectively.

[0133] Optional, α L (L), α T (T), α H (H), α D The formulas for (D) are respectively expressed as:

[0134]

[0135] α T (T) = 1 + λ2 (T - μ2)

[0136] α H (H) = 1 + λ3 (H - μ3)

[0137]

[0138] Among them, λ1 controls the steepness of the function, μ1 represents the baseline value of light intensity, λ2 represents the sensitivity coefficient of temperature change, μ2 represents the baseline temperature, λ3 represents the sensitivity coefficient of humidity change, μ3 represents the baseline humidity, and λ4 represents the coefficient of the influence of dust quantization value on image quality.

[0139] S403: The detection device calculates the image data and the second environment data based on the image compensation formula to obtain calibrated image data.

[0140] In this step, after the image compensation formula is obtained through the above steps, the image data and the second environment data are calculated based on the image compensation formula to obtain calibrated image data.

[0141] Specifically, after correcting the second tube rolling compensation formula, the image compensation formula is obtained:

[0142]

[0143] The method for processing rolled pipe length detection data provided in this embodiment extracts environmental data based on image data to generate second environmental data. A preset second rolled pipe compensation formula is corrected based on the second environmental data and the image data to obtain an image compensation formula. Based on the image compensation formula, the image data and the second environmental data are calculated to obtain calibrated image data. By designing a compensation formula and correcting it based on the characteristics of the image data, the accuracy of the data correction is ensured, and a data basis is provided for subsequent rolled pipe length detection.

[0144] Figure 5 The flowchart of the fifth embodiment of the method for processing the pipe length detection data provided in the embodiment of the present application is as follows: Figure 5 As shown, based on the above embodiments, step S203 specifically includes

[0145] S501: The detection device extracts environmental data based on the ultrasonic data to generate third environmental data.

[0146] This step is the same as step S301 and step S401 in the aforementioned embodiment, and extracts environmental data that is correlated with the ultrasonic wave data from the environmental data as the third environmental data.

[0147] S502: The detection device modifies the preset third pipe rolling compensation formula according to the third environment data and the ultrasonic data to obtain an ultrasonic compensation formula.

[0148] In this step, based on the characteristics of the ultrasonic sensor, a third pipe rolling compensation formula is pre-designed:

[0149]

[0150] Among them, D compenstated represents the compensated ultrasonic data, D raw represents ultrasonic data, ω i represents the weight factor of the i-th environmental data, f i (x i ) represents the i-th environmental data x i The nonlinear compensation function.

[0151] In order to obtain the precise length of the rolled pipe, after extracting the third environmental data, the third pipe rolling compensation formula is corrected based on the third environmental data, thereby obtaining the ultrasonic compensation formula.

[0152] Optionally, taking the third environmental data including temperature, humidity and equipment vibration as an example, the third pipe rolling compensation formula is modified to obtain:

[0153] D compenstated =raw + T · T (T)+ H · H (H)+ V · V (V)

[0154] Among them, T, H, and V represent temperature, humidity, and equipment vibration respectively, ω T 、ω H 、ω V Represents the weighting factors of temperature, humidity, and equipment vibration, f T (T), f H (H), f V (V) represents the compensation functions of temperature, humidity, and equipment vibration respectively.

[0155] Optional, f T (T), f H (H), f V (V) are respectively expressed as:

[0156] f T (T) = α·(T-T0)

[0157] f H (H) = β·(H-H0)

[0158] f V (V)=MovingAverageFilter(V0,window_size)

[0159] Among them, α represents the temperature compensation coefficient, T0 represents the reference temperature, β represents the humidity compensation coefficient, H0 represents the reference humidity, MovingAverageFilter is the moving average filter function, and window_size is the size of the filter window.

[0160] S503: The detection device calculates the ultrasonic data and the third environment data based on the ultrasonic compensation formula to obtain calibrated ultrasonic data.

[0161] In this step, after the ultrasonic image compensation formula is obtained through the above steps, calculations are performed based on the ultrasonic compensation formula, the ultrasonic data, and the third environment data to obtain calibrated ultrasonic data.

[0162] Specifically, after correcting the third tube rolling compensation formula, the ultrasonic compensation formula is obtained:

[0163] D compenstated = raw + T ·(α·(T-T0))+ω H·(β·(H-H0))+ V

[0164] ·(MovingAverageFilter(V0,window_size))

[0165] In the method for processing rolled pipe length detection data provided in this embodiment, the detection equipment extracts environmental data based on ultrasonic data to generate third environmental data. Based on the third environmental data and the ultrasonic data, a preset third rolled pipe compensation formula is modified to obtain an ultrasonic compensation formula. Based on the ultrasonic compensation formula, the ultrasonic data and the third environmental data are calculated to obtain calibrated ultrasonic data. By designing a compensation formula and modifying it based on the characteristics of the ultrasonic data, the accuracy of the data correction is ensured and a data foundation is provided for subsequent rolled pipe length detection.

[0166] Figure 6 The flowchart of the sixth embodiment of the method for processing the pipe length detection data provided in the embodiment of the present application is as follows: Figure 6 As shown, based on the above embodiments, step S103 specifically includes:

[0167] S601: The edge device preprocesses and extracts features from the calibrated feature data set to obtain a feature set.

[0168] In this step, after receiving the calibrated feature data set sent by the detection equipment, the edge device needs to preprocess and extract features from the calibrated feature data set in order to better perform intelligent processing on the calibrated feature data set and thus obtain the accurate length of the rolled pipe, thereby obtaining a feature set.

[0169] Specifically, the calibrated point cloud data can be cleaned to remove invalid points due to noise, outliers, or occlusion. Coordinate transformation unifies the coordinate systems of different sensors to ensure spatial consistency of the data. Downsampling reduces the number of points while maintaining the shape characteristics of the point cloud, reducing the computational complexity of subsequent processing. Feature extraction can be performed to extract the contour of the rolled tube through methods such as point cloud fitting and surface reconstruction. Dimension measurement calculates the length, diameter, and other dimensions of the rolled tube based on the contour.

[0170] Image enhancement can be performed on the calibrated image data to improve contrast and sharpness, making the tube outline clearer. Filtering uses methods such as median filtering and Gaussian filtering to remove noise from the image. Binarization converts the image into a black and white binary image to facilitate tube outline extraction. Feature extraction includes edge detection using algorithms such as Canny and Sobel to extract the tube's edges. Morphological processing uses operations such as dilation and erosion to further extract the tube's shape features.

[0171] Calibrated ultrasonic data can be used for time-to-distance conversion, calculating distance based on ultrasonic propagation speed and reception time. Outlier rejection removes outliers caused by interference or errors. Feature extraction allows for distance analysis, analyzing changes in the distance between the pipe and the sensor based on ultrasonic data.

[0172] S602: The edge device performs feature fusion processing on the feature set to obtain a fused feature vector.

[0173] In this step, in order to improve the accuracy of pipe rolling detection, the features extracted by different sensors can be fused to obtain a fused feature vector.

[0174] Specifically, the edge device standardizes the feature set to obtain a feature vector set, performs weight calculation on each feature vector in the feature vector set to obtain the feature weight corresponding to each feature vector, and performs fusion calculation on the feature vector set based on the feature weight corresponding to each feature vector and a preset fusion formula to obtain a fused feature vector.

[0175] Optionally, data fusion can be performed using methods such as weighted averaging, Kalman filtering, and Bayesian networks.

[0176] S603: The edge device performs intelligent calculations on the fused feature vector to obtain target detection results.

[0177] In this step, after the fused feature vector is obtained, the length of the rolled pipe can be calculated based on the fused feature vector, thereby obtaining a target detection result.

[0178] Specifically, the calculation formula for the rolled tube length is expressed as:

[0179]

[0180] Among them, f i Represents the fused feature vector, k represents k features, which are k features extracted from the fused feature vector, ω i is the weight coefficient of each feature, nonlinear(x) is a nonlinear function, a i and b iis the parameter of the nonlinear function, c is a positive real number used to adjust the scaling of the final length value, λ is the regularization weight coefficient, and p is the type of regularization term.

[0181] In the method for processing rolled pipe length detection data provided in this embodiment, edge devices preprocess and extract features from a calibrated feature dataset to obtain a feature set. This feature set is then fused to produce a fused feature vector, which is then intelligently computed to produce the target detection result. This fusion of data from different sensors ensures more accurate pipe length detection. Intelligent processing by edge devices improves processing efficiency and adapts to any detection scenario, enhancing flexibility.

[0182] Figure 7 The flowchart of the seventh embodiment of the method for processing the pipe length detection data provided in the embodiment of the present application is as follows: Figure 7 As shown, based on the above embodiments, step S602 specifically includes S701: the edge device normalizes the feature set to obtain a feature vector set.

[0183] In this step, in order to process different feature data collected by different sensors in the feature set together, it is necessary to unify different types of data into one data form, and the feature set needs to be standardized to obtain a feature vector set.

[0184] Specifically, the feature set is standardized using the following formula:

[0185]

[0186] Among them, X norm, represents the set of feature vectors obtained after normalization, X i represents the feature set, μ i represents the mean vector, δ i represents the standard deviation vector.

[0187] The mean vector and standard deviation formula can be expressed as follows:

[0188]

[0189] Where N represents the number of samples, represents the feature vector of the jth sample of sensor i.

[0190] The final normalization formula is:

[0191]

[0192] S702: The edge device performs weight calculation on each feature vector in the feature vector set to obtain a feature weight corresponding to each feature vector.

[0193] In this step, since different sensor data have different influences on the rolled tube length, a feature weight is set for each feature vector in order to make the detected length more accurate.

[0194] Specifically, the weight coefficient formula is expressed as:

[0195] ω i =(X norm,i )

[0196] S703: The edge device performs a fusion calculation on the feature vector set based on the feature weight corresponding to each feature vector and a preset fusion formula to obtain a fused feature vector.

[0197] In this step, after the weight calculation in the above steps, the feature vector set is subjected to feature weighted fusion.

[0198] Specifically, the fusion formula is expressed as:

[0199]

[0200] Optionally, in order to make the fused feature vector set more accurate, the fusion result is optimized using the following formula:

[0201]

[0202] Where L represents the loss function, such as mean squared error, λ represents the regularization coefficient, and y is the true label.

[0203] In the method for processing rolled pipe length detection data provided in this embodiment, the edge device normalizes a feature set to obtain a feature vector set. Each feature vector in the feature vector set is weighted to obtain a corresponding feature weight. Based on the feature weights corresponding to each feature vector and a preset fusion formula, the feature vector set is fused to obtain a fused feature vector. By assigning feature weights to different types of feature data, the method more flexibly adapts to the varying effects of different data on rolled pipe length, improving detection flexibility and ensuring more accurate data detection.

[0204] Figure 8 A schematic diagram of the structure of the detection device provided in the embodiment of the present application is shown as follows: Figure 8 As shown, the detection device 800 includes: a processor 802, a memory 801 in communication with the processor 802, a laser radar 803, a depth camera 804, an ultrasonic sensor 805, an environmental sensor 806, and a communication interface 807;

[0205] The memory 801 stores computer-executable instructions;

[0206] The processor 802 executes the computer-executable instructions stored in the memory 801 to implement the method for processing the pipe length detection data in any one of the method embodiments.

[0207] Multi-laser radar 803 is used to collect point cloud data of rolled pipes.

[0208] The depth camera 804 is used to collect image data of the rolled pipe.

[0209] The ultrasonic sensor 805 is used to collect ultrasonic data of pipe rolling.

[0210] The environmental sensor 806 is used to collect environmental data.

[0211] The communication interface 807 is used for data communication.

[0212] Figure 9 A schematic diagram of the structure of the detection system provided in the embodiment of the present application is shown in FIG. Figure 9 As shown, the detection system 900 includes a detection device 901 and an edge device 902 , which are communicatively connected to execute the method for processing the pipe length detection data in any of the aforementioned embodiments.

[0213] The embodiments of the present application further provide a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are executed by a processor, the computer-executable instructions are used to execute the method for processing pipe length detection data provided in the various embodiments described above.

[0214] The computer-readable storage medium mentioned above may be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory, electrically erasable programmable read-only memory, erasable programmable read-only memory, programmable read-only memory, read-only memory, magnetic storage, flash memory, magnetic disk, or optical disk. The computer-readable storage medium may be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0215] Optionally, a readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0216] An embodiment of the present application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium, and when at least one processor executes the computer program, it can implement the technical solution provided by any of the above method embodiments.

[0217] In this application, "at least one" means one or more, and "more" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: the existence of A alone, the existence of A and B at the same time, and the existence of B alone, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship; in the formula, the character " / " indicates that the previous and next associated objects are in a "division" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0218] It is understood that the various numerical numbers involved in the embodiments of the present application are only for the convenience of description and are not intended to limit the scope of the embodiments of the present application. In the embodiments of the present application, the order of the sequence numbers of the above-mentioned processes does not mean the order of execution. 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.

[0219] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0220] It should be understood that the present application is not limited to the exact structure described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for processing pipe length detection data, characterized in that: Applied to a detection system, the detection system includes a detection device and an edge device, the detection device and the edge device are communicatively connected; the method includes: The detection device acquires a feature data set and environmental data of the rolled tube to be detected, wherein the feature data set includes point cloud data, image data and ultrasonic data; The detection device calibrates the feature data set according to the environmental data to obtain a calibrated feature data set, and sends the calibrated feature data set to the edge device; The edge device performs intelligent analysis and processing on the calibrated feature data set to generate a target detection result, wherein the target detection result includes a length value of the rolled pipe to be detected; The calculation formula for the length of the rolled tube to be tested is expressed as follows: Among them, f i represents the fused feature vector, k represents k features, which are the k features extracted from the fused feature vector, ωx is the weight coefficient of each feature, nonlinear(x) is a nonlinear function, and a i and b i is the parameter of the nonlinear function, c is a positive real number used to adjust the scaling of the final length value, λ is the regularization weight coefficient, and p is the type of regularization term; The detection device calibrates the feature data set according to the environmental data to obtain a calibrated feature data set, including: The detection device calibrates the point cloud data according to the environmental data and the material properties of the rolled pipe to be detected to obtain calibrated point cloud data; The formula is: Among them, I corrected Represents the calibrated point cloud data, I original represents the target point cloud data, T and H represent the first environment data, v std is the speed of light under standard conditions, T std and H std are the standard values ​​of temperature and humidity, k T and k H represents the correlation coefficient, L0 represents the length of the rolled tube at the reference temperature, α represents the thermal expansion coefficient of the rolled tube material, a1, a2, a 12 represents the correction factor; The detection device calibrates the image data according to the environmental data to obtain calibrated image data; The formula is: Where I`(x,y) represents the corrected pixel value, I(x,y) represents the pixel value of the image data at the coordinate (x,y), max(I) and min(I) represent the maximum and minimum pixel values, respectively, β represents the global offset, L represents the light intensity, T represents the temperature, H represents the humidity, D represents the dust quantization value, λ1 controls the steepness of the function, μ1 represents the baseline value of the light intensity, λ2 represents the sensitivity coefficient of temperature change, μ2 represents the baseline temperature, λ3 represents the sensitivity coefficient of humidity change, μ3 represents the baseline humidity, and λ4 represents the coefficient of the influence of the dust quantization value on the image quality; The detection device calibrates the ultrasonic data according to the environmental data to obtain calibrated ultrasonic data; The formula is: D compenstated =D raw +ω T ·(α·(T-T0))+ω H ·(β·(H-H0))+ω V ·(MovingAverageFilter(V0,window_size)) Among them, D compenstated represents the compensated ultrasonic data, D raw represents ultrasonic data, T, H, and V represent temperature, humidity, and device vibration respectively, ω T 、ω H 、ω V Represents the weighting factors of temperature, humidity, and equipment vibration, α represents the temperature compensation coefficient, T0 represents the reference temperature, β represents the humidity compensation coefficient, H0 represents the reference humidity, MovingAverageFilter is the moving average filter function, and window_size is the size of the filter window; The detection device aggregates the calibrated point cloud data, the calibrated image data, and the calibrated ultrasonic data to obtain the calibrated feature data set.

2. The method according to claim 1, characterized in that The detection device calibrates the point cloud data according to the environmental data to obtain calibrated point cloud data, including: The detection device extracts data from the environmental data based on the point cloud data to generate first environmental data, where the first environmental data is environmental data that is correlated with the point cloud data; The detection device extracts the number of point clouds related to the length of the rolled pipe to be detected from the point cloud data as target point cloud data; The detection device modifies a preset first pipe rolling compensation formula according to the first environmental data and characteristics of the sensor that collects the point cloud data to obtain a point cloud compensation formula; The detection device calculates the target point cloud data and the first environmental data based on the point cloud compensation formula to obtain the calibrated point cloud data.

3. The method according to claim 1, characterized in that The detection device calibrates the image data according to the environmental data to obtain calibrated image data, including: The detection device extracts data from the environmental data based on the image data to generate second environmental data, where the second environmental data is environmental data that is correlated with the image data; The detection device modifies a preset second pipe rolling compensation formula according to the second environmental data and the image data to obtain an image compensation formula; The detection device calculates the image data and the second environmental data based on the image compensation formula to obtain the calibrated image data.

4. The method according to claim 1, wherein The detection device calibrates the ultrasonic data according to the environmental data to obtain calibrated ultrasonic data, including: The detection device extracts data from the environmental data based on the ultrasonic data to generate third environmental data, where the third environmental data is environmental data correlated with the ultrasonic data; The detection device modifies a preset third pipe rolling compensation formula according to the third environmental data and the ultrasonic data to obtain an ultrasonic compensation formula; The detection device calculates the ultrasonic data and the third environment data based on the ultrasonic compensation formula to obtain the calibrated ultrasonic data.

5. The method according to claim 1, wherein The edge device performs intelligent analysis and processing on the calibrated feature data set to generate a target detection result, including: The edge device performs preprocessing and feature extraction on the calibrated feature data set to obtain a feature set; The edge device performs feature fusion processing on the feature set to obtain a fused feature vector; The edge device performs intelligent calculation processing on the fused feature vector to obtain the target detection result.

6. The method according to claim 5, characterized in that The edge device performs feature fusion processing on the feature set to obtain a fused feature vector, including: The edge device performs normalization processing on the feature set to obtain a feature vector set; The edge device performs weight calculation on each feature vector in the feature vector set to obtain a feature weight corresponding to each feature vector; The edge device performs a fusion calculation on the feature vector set based on the feature weight corresponding to each feature vector and a preset fusion formula to obtain the fused feature vector.

7. The method according to claim 6, characterized in that The edge device performs intelligent calculation processing on the fused feature vector to obtain the target detection result, including: The edge device calculates the fused feature vector based on a preset calculation model to obtain the target detection result.

8. A detection device, characterized in that: include: A processor, a memory, a lidar, a depth camera, an ultrasonic sensor, an environmental sensor, and a communication interface in communication with the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method for processing pipe length detection data according to any one of claims 1 to 7.

9. A detection system, characterized in that: include: A detection device and an edge device, wherein the detection device and the edge device are communicatively connected; A method for processing pipe length detection data according to any one of claims 1 to 7.

Images