CN116740044B - Copper pipe milling surface processing method and system based on visual detection and control - Google Patents
Copper pipe milling surface processing method and system based on visual detection and control Download PDFInfo
- Publication number
- CN116740044B CN116740044B CN202310802832.4A CN202310802832A CN116740044B CN 116740044 B CN116740044 B CN 116740044B CN 202310802832 A CN202310802832 A CN 202310802832A CN 116740044 B CN116740044 B CN 116740044B
- Authority
- CN
- China
- Prior art keywords
- copper pipe
- milling
- result
- parameter
- visual detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 title claims abstract description 113
- 229910052802 copper Inorganic materials 0.000 title claims abstract description 113
- 239000010949 copper Substances 0.000 title claims abstract description 113
- 238000003801 milling Methods 0.000 title claims abstract description 104
- 238000001514 detection method Methods 0.000 title claims abstract description 56
- 230000000007 visual effect Effects 0.000 title claims abstract description 53
- 238000003672 processing method Methods 0.000 title claims description 11
- 238000005457 optimization Methods 0.000 claims abstract description 67
- 238000000605 extraction Methods 0.000 claims abstract description 47
- 238000012544 monitoring process Methods 0.000 claims abstract description 45
- 238000012545 processing Methods 0.000 claims abstract description 19
- 238000004088 simulation Methods 0.000 claims description 24
- 238000011179 visual inspection Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 9
- 238000007635 classification algorithm Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 8
- 230000004913 activation Effects 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 3
- 230000001131 transforming effect Effects 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 16
- 238000013527 convolutional neural network Methods 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 4
- 238000012549 training Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000003908 quality control method Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000001050 lubricating effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000003746 surface roughness Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the technical field of copper pipe milling, in particular to a copper pipe milling method and a copper pipe milling system based on visual detection and control, wherein the method comprises the following steps: image acquisition is carried out on the outer surface of the copper pipe processed by milling, and a copper pipe milling surface image is obtained; the method comprises the steps of obtaining a feature extraction result by identifying a copper pipe milling surface image; determining necessary adjustment parameter information according to the feature extraction result, establishing a visual detection control model, and carrying out optimization adjustment on the necessary adjustment parameter information through the visual detection control model to obtain a parameter optimization result; and determining a monitoring feedback standard, and evaluating the parameter optimization result according to the monitoring feedback standard. The invention effectively solves the problem that the traditional copper pipe milling surface depends on subjective judgment of operators, and can detect and analyze the precision quality of milling surface processing on line, so that the finish degree and precision of milling surfaces among different copper pipes tend to be stable, the manual intervention is reduced, and the processing efficiency and consistency are improved.
Description
Technical Field
The invention relates to the technical field of copper pipe milling, in particular to a copper pipe milling method and system based on visual detection and control.
Background
The milling surface of the copper pipe is a processing process of the copper pipe, the outer surface of the copper pipe is removed and trimmed by a milling process through a rotary cutter, so that the requirements of flatness, smoothness and precision are met, the quality problem of the milling surface of the copper pipe can affect a plurality of aspects, and the size precision of the copper pipe can exceed a specified range, and the assembly and connection performance of the copper pipe are affected. Meanwhile, poor milling surface quality can cause the increase of the surface roughness of the copper pipe, and the defects of rugged or burrs and the like appear, so that the appearance, touch feeling and lubricating performance are affected.
Conventional quality control methods typically rely on subjective judgment by the operator, and quality control methods typically require offline sampling and subsequent detection and analysis processes, resulting in delays in feedback and problems in the process that cannot be timely found and corrected.
The information disclosed in this background section is only for enhancement of understanding of the general background of the disclosure and is not to be taken as an admission or any form of suggestion that this information forms the prior art that is well known to a person skilled in the art.
Disclosure of Invention
The invention provides a copper pipe milling surface processing method and system based on visual detection and control, which can effectively solve the problems in the background technology.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
a copper pipe milling surface processing method based on visual detection and control comprises the following steps:
image acquisition is carried out on the outer surface of the copper pipe processed by milling, and a copper pipe milling surface image is obtained;
the characteristic extraction result is obtained by identifying the surface image of the milling surface of the copper pipe;
determining necessary adjustment parameter information according to the feature extraction result, establishing a visual detection control model, and carrying out optimization adjustment on the necessary adjustment parameter information through the visual detection control model to obtain a parameter optimization result;
and determining a monitoring feedback standard, and evaluating the parameter optimization result according to the monitoring feedback standard.
Further, the step of obtaining the feature extraction result by identifying the copper pipe milling surface image comprises the following steps:
converting the surface image of the copper pipe milling surface into a feature matrix capable of carrying out convolution operation;
constructing a convolution layer in a convolution neural network, wherein the convolution layer can identify inferior characteristics of a copper pipe milling surface image;
performing convolution operation on the feature matrix through the convolution layer to obtain a convolution operation result;
and obtaining the feature extraction result according to the convolution operation result.
Further, the visual inspection control model includes: an input layer, a full connection layer and an output layer;
flattening the feature extraction result and converting the feature extraction result into a one-dimensional vector;
mapping and transforming the one-dimensional vector as input through the fully connected layer;
wherein the one-dimensional vector is non-linearly transformed between the fully connected layer and the output layer using an activation function.
Further, the establishing the visual inspection control model includes:
collecting historical data and preprocessing the historical data;
selecting a proper classification algorithm to classify the preprocessed historical data, and establishing a classification subset of the historical data;
formulating a standard meeting threshold according to the historical data classification subset;
and establishing the visual detection control model according to the standard meeting threshold.
Further, the visual detection control model performs optimization adjustment on the necessary adjustment parameter information, and obtaining a parameter optimization result includes:
inputting the necessary adjustment parameter information into the visual inspection control model;
carrying out parameter correction on the necessary adjustment parameter information according to the standard-meeting threshold;
and carrying out classification weight parameter compensation on the parameter correction through the historical data classification subset corresponding to the necessary adjustment parameters, and obtaining the parameter optimization result.
Further, by identifying the copper tube milling surface image, a feature extraction result is obtained, which comprises:
converting the copper pipe milling surface image into a characteristic image, wherein pixel points or areas are represented as nodes in the image, and edges are connected with adjacent nodes
Calculating edge weights in the feature graphs;
the edge weight calculation formula is as follows:
wherein w (i, j) is the edge weight between i and j, f (i) and f (j) are the eigenvalues of pixel i and pixel j respectively, i and j are the indexes of adjacent pixels, and sigma is a parameter for controlling the weight attenuation speed;
and obtaining a feature extraction result through calculating the edge weight in the feature graph.
Further, the determining a monitoring feedback criterion, the monitoring feedback criterion evaluating the parameter optimization result includes:
acquiring surface milling parameter data of the historical high-quality copper pipe, and determining a monitoring feedback standard according to the surface milling parameter data acquisition result;
determining the parameter optimization result, and carrying out face milling simulation demonstration according to the parameter optimization result to obtain a simulation demonstration result;
and detecting the simulation demonstration result through the monitoring feedback standard, and evaluating the parameter optimization result.
Copper pipe mills face processing system based on visual detection and control, the system includes:
and an image acquisition module: image acquisition is carried out on the outer surface of the copper pipe processed by milling, and a copper pipe milling surface image is obtained;
and the feature extraction module is used for: the characteristic extraction result is obtained by identifying the surface image of the milling surface of the copper pipe;
and an optimization and adjustment module: determining necessary adjustment parameter information according to the feature extraction result, and establishing a visual detection control model, wherein the visual detection control model performs optimization adjustment on the necessary adjustment parameter information to obtain a parameter optimization result;
monitoring and evaluating module: and determining a monitoring feedback standard, wherein the monitoring feedback standard evaluates the parameter optimization result.
Further, the optimization adjustment module includes:
a data acquisition unit: collecting historical data and preprocessing the historical data;
a data processing unit: selecting a proper classification algorithm to classify the historical data, and establishing a classification subset of the historical data;
threshold setting unit: formulating a standard meeting threshold according to the historical data classification subset;
model building unit: and establishing the visual detection control model according to the standard meeting threshold.
Further, the monitoring and evaluating module includes:
a standard determination unit: acquiring surface milling parameter data of the historical high-quality copper pipe, and determining a monitoring feedback standard according to the surface milling parameter data acquisition result;
simulation demonstration unit: determining the parameter optimization result, and carrying out face milling simulation demonstration according to the parameter optimization result to obtain a simulation demonstration result;
and a monitoring feedback unit: and detecting the simulation demonstration result through the monitoring feedback standard, and evaluating the parameter optimization result.
By the technical scheme of the invention, the following technical effects can be realized:
the problem that the traditional copper pipe milling surface depends on subjective judgment of operators is effectively solved, the accuracy quality of milling surface processing can be detected and analyzed on line, the finish degree and the accuracy of milling surfaces among different copper pipes tend to be stable, manual intervention is reduced, and the processing efficiency and consistency are improved.
The foregoing description is only an overview of the technical solutions of the present application, and may be implemented according to the content of the specification in order to make the technical means of the present application more clearly understood, and in order to make the above-mentioned and other objects, features and advantages of the present application more clearly understood, the following detailed description of the present application will be given.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present invention, and other drawings may be obtained according to the drawings without inventive effort to those skilled in the art.
FIG. 1 is a schematic flow chart of a copper tube milling surface processing method based on visual detection and control;
FIG. 2 is a flow chart of obtaining feature extraction results;
FIG. 3 is a schematic flow chart of establishing a visual inspection control model;
FIG. 4 is a flow chart of obtaining a parameter optimization result;
FIG. 5 is a flow chart for evaluating the results of parameter optimization;
fig. 6 is a schematic structural diagram of a copper tube milling surface processing system based on visual detection and control.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.
Example 1
As shown in fig. 1, the present application provides a copper pipe milling surface processing method based on visual detection and control, the method comprising:
s100: image acquisition is carried out on the outer surface of the copper pipe processed by milling, and a copper pipe milling surface image is obtained;
s200: the method comprises the steps of obtaining a feature extraction result by identifying a copper pipe milling surface image;
the method comprises the steps of carrying out image acquisition on the outer surface of a copper pipe subjected to surface milling through image acquisition equipment, wherein the surface image of the surface of the copper pipe is formed by acquiring multiple angles of the copper pipe, carrying out feature extraction and identification on the surface image of the surface of the copper pipe through image processing and identification technology, and extracting details and features of a milling surface, wherein the features can comprise textures, concave-convex, contours, cracks and the like of the copper pipe, and can reflect a series of features affecting the surface finish and the dimensional accuracy of the copper pipe in the surface milling process of the copper pipe.
S300: determining necessary adjustment parameter information according to the feature extraction result, establishing a visual detection control model, and carrying out optimization adjustment on the necessary adjustment parameter information through the visual detection control model to obtain a parameter optimization result;
specifically, a feature extraction result is obtained through a copper pipe milling surface image, the feature extraction result reflects problems existing in the copper pipe milling process, parameters which are required to be adjusted are judged through the problems, for example, the accuracy deviation of the outer contour of the copper pipe in the feature extraction result in size exists, the corresponding required adjustment parameter information is the feeding speed and milling depth of the copper pipe, the copper pipe is adjusted through a visual detection control model, and a parameter optimization result is obtained after adjustment.
S400: and determining a monitoring feedback standard, and evaluating the parameter optimization result according to the monitoring feedback standard.
Specifically, for copper pipes with different specifications and different user demands, the monitoring feedback standards are inconsistent, the specific standard is that the produced high-quality copper pipes with the same type are acquired off-line, then the evaluation standard of the optimization result is definitely determined, at the moment, if the monitoring feedback standards are not met, the monitoring feedback standards are required to be fed back to the visual detection control model, and the necessary adjustment parameter information is selected again or readjusted in the original necessary adjustment parameter information through the visual detection control model.
According to the technical scheme, the problem that the traditional copper pipe milling surface depends on subjective judgment of operators is effectively solved, the accuracy and quality of milling surface processing can be detected and analyzed on line, the smoothness and accuracy of milling surfaces among different copper pipes tend to be stable, manual intervention is reduced, and processing efficiency and consistency are improved.
Further, as shown in fig. 2, by identifying the surface image of the milled surface of the copper pipe, the feature extraction result includes:
s210: converting the surface image of the copper pipe milling surface into a feature matrix capable of carrying out convolution operation;
s220: constructing a convolution layer in a convolution neural network, wherein the convolution layer can identify inferior characteristics of the surface image of the copper pipe milling surface;
s230: performing convolution operation on the feature matrix through the convolution layer to obtain a convolution operation result;
s240: and obtaining a feature extraction result according to the convolution operation result.
Specifically, the convolutional neural network can learn and capture local features in the image, a more accurate and representative feature extraction result can be obtained by carrying out convolutional operation on the copper pipe milling surface image, and by constructing a convolutional layer, the convolutional neural network can identify inferior features of the copper pipe milling surface image, so that the method is beneficial to quickly detecting and identifying defects, flaws or poor features in the milling surface, and further improves the sensitivity and precision of detection. Of course, in the process of constructing the convolution layer, the low-resolution feature map can be restored to the high-resolution feature map through deconvolution operation, so that details and local information in the image can be captured more accurately, the feature extraction precision is improved, the deconvolution operation can reversely operate the result of the convolution operation, the feature map is remapped to the size of the input image, the activation mode in the convolution layer can be visualized, and understanding and analyzing of the features and the characterization capability of the network learned in different layers can be facilitated.
Further, the visual inspection control model includes: an input layer, a full connection layer and an output layer;
flattening the feature extraction result and converting the feature extraction result into a one-dimensional vector;
taking the one-dimensional vector as input and mapping and transforming through the full connection layer;
wherein between the fully connected layer and the output layer, a one-dimensional vector is non-linearly transformed using an activation function.
Specifically, the step of converting the feature extraction result into a one-dimensional vector retains the spatial relationship and structural information of the feature, and converts the spatial relationship and structural information into a form that can be processed by a fully connected layer, wherein each neuron of the fully connected layer has a connection relationship with each neuron in a previous layer, so that the high-level abstract feature can be learned, the number of neurons of an output layer is determined according to a specific task, and two classification (such as roughness/non-roughness of the surface of a copper tube) or multiple classification is performed by using one neuron, wherein an activation function between the fully connected layer and the output layer can be selected as ReLU, sigmoid, softmax, and an appropriate activation function can be selected according to requirements.
Further, referring to fig. 3, establishing the visual inspection control model includes:
s311: collecting historical data and preprocessing the historical data;
s312: selecting a proper classification algorithm to classify the preprocessed historical data, and establishing a classification subset of the historical data;
s313: formulating a standard meeting threshold according to the historical data classification subset;
s314: and establishing a visual detection control model according to the standard threshold.
Specifically, the historical data can be data records of the copper pipe before and after surface milling, the data records comprise image data, feature extraction results, milling parameters and corresponding processing quality of the copper pipe before and after surface milling, and the preprocessing of the historical data can comprise operations such as data cleaning, denoising and the like, so that noise, abnormal values and error data in some historical data can be removed, the quality and accuracy of the data can be improved, and the reliability of subsequent analysis and modeling can be ensured; the method comprises the steps of selecting a proper classification algorithm, such as a decision tree, a support vector machine, a neural network and the like, classifying historical data into different categories and characteristics, classifying the preprocessed historical data based on the classification algorithm, wherein the classification rule is to process copper pipes with different production specifications or classify different requirements on the quality of the copper pipes facing different customers, classifying the preprocessed historical data into different data subsets according to the classification rule, analyzing and preparing thresholds meeting standards according to the data of the different subsets, processing quality in the process of data acquisition corresponding to the standards, taking processed offline copper pipes as samples, matching the processed offline copper pipes with the classified subsets, converting the characteristics acquired by surface images of the milled surfaces of the copper pipes into the form of data sets based on a visual detection control model established according to the standard thresholds, and realizing monitoring and control on the processing quality of the milled surfaces of the copper pipes by the model, and improving consistency and stability of the processing quality of the milled surfaces of the copper pipes, updating the historical data and optimizing the model continuously, and improving and conforming to the standard threshold gradually.
Further, referring to fig. 4, the visual inspection control model performs optimization adjustment on necessary adjustment parameter information, and obtaining a parameter optimization result includes:
s321: inputting necessary adjustment parameter information into a visual detection control model;
s322: carrying out parameter correction on necessary adjustment parameter information according to the standard threshold;
s323: and carrying out classification weight parameter compensation on parameter correction through the historical data classification subset corresponding to the necessary adjustment parameters, and obtaining a parameter optimization result.
Specifically, in the previous step, the obtained necessary adjustment parameter information is input into a visual detection control model, the visual detection control model corrects the necessary adjustment parameter information by conforming to a standard threshold, the correction principle is to correct the necessary adjustment parameter information to be the median value of two endpoints conforming to the standard threshold, and it should be noted that the standard threshold conforming to the standard threshold in the visual detection control model is adjusted at any time according to the updating of the historical data, so that the range conforming to the standard threshold and the endpoint value are updated in a period of time; because the establishment of the classification subset is faced with copper pipes with different specifications or different quality requirements of clients, and correction meeting the standard threshold has generality for the necessary adjustment parameters, the correction for the necessary adjustment parameters needs to consider the classification subset corresponding to the copper pipe, and certain compensation is carried out on the necessary adjustment parameters based on the classification subset, for example, the necessary adjustment parameters are the feeding speed of the copper pipe, if the copper pipe has higher requirements on the roughness of the outer surface of the copper pipe, after the feeding speed is adjusted to meet the median value of the two end points of the standard threshold, fine adjustment compensation is carried out on the slower end threshold, and the compensated adjustment can better meet the requirements on the roughness of the outer surface of the copper pipe during face milling. After the visual detection control model is established, the visual detection control model can be trained and verified, the visual detection control model can be ensured to have high accuracy through training and verification, and the training model uses a large amount of marking data to learn characteristics and modes, so that the milling surface of the copper pipe can be accurately identified and analyzed; the verification model evaluates the performance of the model through an independent test data set, verifies whether the model can accurately detect and control the milling surface quality, and can improve the reliability and stability of the visual detection control model through training and verification, and discover and correct the deviation and error of the model under different conditions, thereby improving the robustness and stability of the model.
Further, the feature extraction result is obtained by identifying the surface image of the milled surface of the copper pipe, and the feature extraction result comprises the following steps:
converting the copper pipe milling surface image into a characteristic image, wherein pixel points or areas are represented as nodes in the image, and edges are connected with adjacent nodes
Calculating edge weights in the feature graphs;
the edge weight calculation formula is as follows:
wherein w (i, j) is the edge weight between i and j, f (i) and f (j) are the eigenvalues of pixel i and pixel j respectively, i and j are the indexes of adjacent pixels, and sigma is a parameter for controlling the weight attenuation speed; exp is the calculation of an exponential function, which represents the power x of e, where x is the fractional content;
and obtaining a feature extraction result through calculating the edge weight in the feature graph.
Specifically, edge weight calculation plays an important role in image recognition, and can emphasize edge information, distinguish targets from backgrounds, promote local consistency, and capture texture and structure information at the same time, so that accuracy and stability of image recognition and segmentation are improved, and the edge weight calculation can be used in combination with a convolutional neural network to improve performance and effect of image recognition. For example, in the target detection task, a convolutional neural network may be used as a feature extractor, and then the extracted features are subjected to image segmentation using calculation of edge weights, so as to obtain accurate boundary information of the target; can be used as a preprocessing or post-processing step of the convolutional neural network. For example, prior to convolving the neural network, the image may also be initially segmented using edge weight calculations to extract regions of interest or reduce background noise.
Further, as shown in fig. 5, determining a monitoring feedback criterion, the monitoring feedback criterion evaluating the parameter optimization result includes:
s410: acquiring surface milling parameter data of the historical high-quality copper pipe, and determining a monitoring feedback standard according to the surface milling parameter data acquisition result;
s420: determining a parameter optimization result, and carrying out face milling simulation demonstration according to the parameter optimization result to obtain a simulation demonstration result;
s430: and detecting the simulation demonstration result through a monitoring feedback standard, and evaluating the parameter optimization result.
Specifically, a certain number of samples are selected from the historical high-quality copper pipe to collect milling parameter data, the parameters can comprise milling speed, feeding speed, cutting depth and other parameters related to milling, a monitoring feedback standard is determined according to collected data analysis and statistics, and a reliable reference standard can be established through collecting and analyzing the parameter data of the historical high-quality copper pipe to evaluate the quality of the current copper pipe milling. The establishment of the monitoring feedback standard can help to identify possible deviation or abnormality in the milling process, provide targets and guidance for subsequent parameter optimization and control, carry out face milling simulation demonstration according to the parameter optimization result, predict the optimization quality, obtain the quality state of the copper pipe face milling after optimization treatment through the simulation demonstration result, thereby judging whether the parameter optimization result has deviation, if the deviation exists, feeding back the simulation demonstration result to the visual detection control model, and optimize the visual detection control model while adjusting the parameter optimization result.
Embodiment two:
as shown in fig. 6, the copper pipe milling surface processing system based on visual detection and control comprises:
and an image acquisition module: image acquisition is carried out on the outer surface of the copper pipe processed by milling, and a copper pipe milling surface image is obtained;
and the feature extraction module is used for: the method comprises the steps of obtaining a feature extraction result by identifying a copper pipe milling surface image;
and an optimization and adjustment module: determining necessary adjustment parameter information according to the feature extraction result, establishing a visual detection control model, and optimizing and adjusting the necessary adjustment parameter information by the visual detection control model to obtain a parameter optimization result;
monitoring and evaluating module: and determining a monitoring feedback standard, and evaluating the parameter optimization result by the monitoring feedback standard.
The adjusting system can effectively realize the purifying and purifying adjusting method of the metal melt, and has the technical effects as described in the embodiment, and the description is omitted here.
Further, the optimization adjustment module includes:
a data acquisition unit: collecting historical data and preprocessing the historical data;
a data processing unit: selecting a proper classification algorithm to classify the historical data, and establishing a classification subset of the historical data;
threshold setting unit: formulating a standard meeting threshold according to the historical data classification subset;
model building unit: and establishing a visual detection control model according to the standard threshold.
Further, the monitoring and evaluating module includes:
a standard determination unit: acquiring surface milling parameter data of the historical high-quality copper pipe, and determining a monitoring feedback standard according to the surface milling parameter data acquisition result;
simulation demonstration unit: determining a parameter optimization result, and carrying out face milling simulation demonstration according to the parameter optimization result to obtain a simulation demonstration result;
and a monitoring feedback unit: and detecting the simulation demonstration result through a monitoring feedback standard, and evaluating the parameter optimization result.
Similarly, the above-mentioned optimization schemes of the system may also respectively correspond to the optimization effects corresponding to the methods in the first embodiment, which are not described herein again.
Although the present application has been described in connection with specific features and embodiments thereof, it will be apparent that various modifications and combinations can be made without departing from the spirit and scope of the application. Accordingly, the specification and figures are merely exemplary illustrations of the application as defined in the appended claims and are to be construed as covering any and all modifications, variations, combinations, or equivalents that are within the scope of the application. It will be apparent to those skilled in the art that various modifications and variations can be made in the present application without departing from the scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the present application and the equivalents thereof, the present application is intended to cover such modifications and variations.
Claims (7)
1. The copper pipe milling surface processing method based on visual detection and control is characterized by comprising the following steps of:
image acquisition is carried out on the outer surface of the copper pipe processed by milling, and a copper pipe milling surface image is obtained;
the feature extraction result is obtained by identifying the surface image of the milled surface of the copper pipe, and the feature extraction result comprises the following steps:
converting the copper pipe milling surface image into a characteristic image, wherein pixel points or areas are represented as nodes in the image, and edges are connected with adjacent nodes
Calculating edge weights in the feature graphs;
the edge weight calculation formula is as follows:
；
wherein w (i, j) is the edge weight between i and j, f (i) and f (j) are the eigenvalues of pixel i and pixel j respectively, i and j are the indexes of adjacent pixels, and sigma is a parameter for controlling the weight attenuation speed;
obtaining a feature extraction result through calculating the edge weight in the feature graph;
determining necessary adjustment parameter information according to the feature extraction result, establishing a visual detection control model, and carrying out optimization adjustment on the necessary adjustment parameter information through the visual detection control model to obtain a parameter optimization result;
determining a monitoring feedback standard, evaluating the parameter optimization result according to the monitoring feedback standard, and comprising the following steps:
acquiring surface milling parameter data of the historical high-quality copper pipe, and determining a monitoring feedback standard according to the surface milling parameter data acquisition result;
determining the parameter optimization result, and carrying out face milling simulation demonstration according to the parameter optimization result to obtain a simulation demonstration result;
and detecting the simulation demonstration result through the monitoring feedback standard, and evaluating the parameter optimization result.
2. The copper tube face milling processing method based on visual detection and control according to claim 1, wherein the step of obtaining a feature extraction result by identifying the copper tube face milling surface image comprises the steps of:
converting the surface image of the copper pipe milling surface into a feature matrix capable of carrying out convolution operation;
constructing a convolution layer in a convolution neural network, wherein the convolution layer can identify inferior characteristics of a copper pipe milling surface image;
performing convolution operation on the feature matrix through the convolution layer to obtain a convolution operation result;
and obtaining the feature extraction result according to the convolution operation result.
3. The copper pipe face milling processing method based on visual detection and control according to claim 2, wherein the visual detection control model comprises: an input layer, a full connection layer and an output layer;
flattening the feature extraction result and converting the feature extraction result into a one-dimensional vector;
mapping and transforming the one-dimensional vector as input through the fully connected layer;
wherein the one-dimensional vector is non-linearly transformed between the fully connected layer and the output layer using an activation function.
4. The copper pipe milling surface processing method based on visual detection and control according to claim 1, wherein the establishing a visual detection control model comprises the following steps:
collecting historical data and preprocessing the historical data;
selecting a proper classification algorithm to classify the preprocessed historical data, and establishing a classification subset of the historical data;
formulating a standard meeting threshold according to the historical data classification subset;
and establishing the visual detection control model according to the standard meeting threshold.
5. The copper pipe face milling processing method based on visual inspection and control according to claim 4, wherein the visual inspection control model performs optimization adjustment on the necessary adjustment parameter information, and obtaining a parameter optimization result comprises:
inputting the necessary adjustment parameter information into the visual inspection control model;
carrying out parameter correction on the necessary adjustment parameter information according to the standard-meeting threshold;
and carrying out classification weight parameter compensation on the parameter correction through the historical data classification subset corresponding to the necessary adjustment parameters, and obtaining the parameter optimization result.
6. Copper pipe mills face processing system based on visual detection and control, its characterized in that, the system includes:
and an image acquisition module: image acquisition is carried out on the outer surface of the copper pipe processed by milling, and a copper pipe milling surface image is obtained;
and the feature extraction module is used for: the feature extraction result is obtained by identifying the surface image of the milled surface of the copper pipe, and the feature extraction result comprises the following steps:
converting the copper pipe milling surface image into a characteristic image, wherein pixel points or areas are represented as nodes in the image, and edges are connected with adjacent nodes
Calculating edge weights in the feature graphs;
the edge weight calculation formula is as follows:
；
wherein w (i, j) is the edge weight between i and j, f (i) and f (j) are the eigenvalues of pixel i and pixel j respectively, i and j are the indexes of adjacent pixels, and sigma is a parameter for controlling the weight attenuation speed;
obtaining a feature extraction result through calculating the edge weight in the feature graph;
and an optimization and adjustment module: determining necessary adjustment parameter information according to the feature extraction result, and establishing a visual detection control model, wherein the visual detection control model performs optimization adjustment on the necessary adjustment parameter information to obtain a parameter optimization result;
monitoring and evaluating module: determining a monitoring feedback standard, wherein the monitoring feedback standard evaluates the parameter optimization result;
the monitoring and evaluating module comprises:
a standard determination unit: acquiring surface milling parameter data of the historical high-quality copper pipe, and determining a monitoring feedback standard according to the surface milling parameter data acquisition result;
simulation demonstration unit: determining the parameter optimization result, and carrying out face milling simulation demonstration according to the parameter optimization result to obtain a simulation demonstration result;
and a monitoring feedback unit: and detecting the simulation demonstration result through the monitoring feedback standard, and evaluating the parameter optimization result.
7. The copper tube face milling system based on visual inspection and control of claim 6, wherein the optimization adjustment module comprises:
a data acquisition unit: collecting historical data and preprocessing the historical data;
a data processing unit: selecting a proper classification algorithm to classify the historical data, and establishing a classification subset of the historical data;
threshold setting unit: formulating a standard meeting threshold according to the historical data classification subset;
model building unit: and establishing the visual detection control model according to the standard meeting threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310802832.4A CN116740044B (en) | 2023-06-30 | 2023-06-30 | Copper pipe milling surface processing method and system based on visual detection and control |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310802832.4A CN116740044B (en) | 2023-06-30 | 2023-06-30 | Copper pipe milling surface processing method and system based on visual detection and control |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116740044A CN116740044A (en) | 2023-09-12 |
| CN116740044B true CN116740044B (en) | 2024-02-27 |
Family
ID=87911439
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310802832.4A Active CN116740044B (en) | 2023-06-30 | 2023-06-30 | Copper pipe milling surface processing method and system based on visual detection and control |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116740044B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117314262B (en) * | 2023-11-27 | 2024-03-08 | 欧派家居集团股份有限公司 | Method and system for detecting multi-procedure automatic integrated processing production line of wooden door |
| CN117972487B (en) * | 2024-01-26 | 2024-09-03 | 常州润来科技有限公司 | Copper pipe milling face cutter disc defect detection method and system |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101532926A (en) * | 2008-12-12 | 2009-09-16 | 齐齐哈尔华工机床制造有限公司 | On-line vision detecting system for automatic impact specimen processing device and image processing method thereof |
| CN103659602A (en) * | 2012-09-25 | 2014-03-26 | 宁波工程学院 | Method and device for detecting cylindrical grinding surface roughness through data fusion technology |
| CN108535264A (en) * | 2018-04-09 | 2018-09-14 | 西京学院 | A kind of Machine Vision Detection milling cutter positioning clamping equipment |
| CN112371752A (en) * | 2020-10-20 | 2021-02-19 | 安徽德诠新材料科技有限公司 | Processing and preparation method of ultrathin-wall copper pipe |
| CN113341882A (en) * | 2021-06-28 | 2021-09-03 | 成都飞机工业（集团）有限责任公司 | Numerical control process design and optimization method based on processing knowledge |
| CN114139772A (en) * | 2021-11-12 | 2022-03-04 | 江苏师范大学 | Method for predicting surface quality of workpiece based on optimized generalized regression neural network |
| CN114357851A (en) * | 2021-10-23 | 2022-04-15 | 西北工业大学 | Numerical control milling surface roughness prediction method based on DAE-RNN |
| CN115056074A (en) * | 2022-08-18 | 2022-09-16 | 潍坊潍尔达石油机械有限公司 | Steel pipe tip clout processingequipment based on machine vision |
| CN115146761A (en) * | 2022-05-26 | 2022-10-04 | 腾讯科技（深圳）有限公司 | Defect detection model training method and related device |
| CN115415549A (en) * | 2022-09-05 | 2022-12-02 | 苏州中科煜宸激光智能科技有限公司 | Laser cladding leveling control system and method based on nonlinear autoregressive neural network |
| CN116167968A (en) * | 2022-12-12 | 2023-05-26 | 天津大学 | H-shaped steel surface defect detection method and system |
-
2023
- 2023-06-30 CN CN202310802832.4A patent/CN116740044B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101532926A (en) * | 2008-12-12 | 2009-09-16 | 齐齐哈尔华工机床制造有限公司 | On-line vision detecting system for automatic impact specimen processing device and image processing method thereof |
| CN103659602A (en) * | 2012-09-25 | 2014-03-26 | 宁波工程学院 | Method and device for detecting cylindrical grinding surface roughness through data fusion technology |
| CN108535264A (en) * | 2018-04-09 | 2018-09-14 | 西京学院 | A kind of Machine Vision Detection milling cutter positioning clamping equipment |
| CN112371752A (en) * | 2020-10-20 | 2021-02-19 | 安徽德诠新材料科技有限公司 | Processing and preparation method of ultrathin-wall copper pipe |
| CN113341882A (en) * | 2021-06-28 | 2021-09-03 | 成都飞机工业（集团）有限责任公司 | Numerical control process design and optimization method based on processing knowledge |
| CN114357851A (en) * | 2021-10-23 | 2022-04-15 | 西北工业大学 | Numerical control milling surface roughness prediction method based on DAE-RNN |
| CN114139772A (en) * | 2021-11-12 | 2022-03-04 | 江苏师范大学 | Method for predicting surface quality of workpiece based on optimized generalized regression neural network |
| CN115146761A (en) * | 2022-05-26 | 2022-10-04 | 腾讯科技（深圳）有限公司 | Defect detection model training method and related device |
| CN115056074A (en) * | 2022-08-18 | 2022-09-16 | 潍坊潍尔达石油机械有限公司 | Steel pipe tip clout processingequipment based on machine vision |
| CN115415549A (en) * | 2022-09-05 | 2022-12-02 | 苏州中科煜宸激光智能科技有限公司 | Laser cladding leveling control system and method based on nonlinear autoregressive neural network |
| CN116167968A (en) * | 2022-12-12 | 2023-05-26 | 天津大学 | H-shaped steel surface defect detection method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116740044A (en) | 2023-09-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116740044B (en) | Copper pipe milling surface processing method and system based on visual detection and control | |
| CN109977808B (en) | Wafer surface defect mode detection and analysis method | |
| CN113870235B (en) | Circular stamping part defect detection method based on quantum firework arc edge extraction | |
| CN116205919B (en) | Hardware part production quality detection method and system based on artificial intelligence | |
| CN109003275B (en) | Segmentation method of weld defect image | |
| CN112862744B (en) | Intelligent detection method for internal defects of capacitor based on ultrasonic image | |
| CN104463199A (en) | Rock fragment size classification method based on multiple features and segmentation recorrection | |
| CN112907519A (en) | Metal curved surface defect analysis system and method based on deep learning | |
| CN113393426B (en) | Steel rolling plate surface defect detection method | |
| CN114706358B (en) | Method and system for processing welding process information of straight welded pipe | |
| CN108647722B (en) | Zinc ore grade soft measurement method based on process size characteristics | |
| CN116758491A (en) | Printing monitoring image analysis method and system applied to 3D printing | |
| CN109859199B (en) | Method for detecting quality of freshwater seedless pearls through SD-OCT image | |
| CN116630329B (en) | Online visual detection method for multi-axis multi-channel numerical control system | |
| CN117350967B (en) | Optical detection-based wafer appearance defect detection system and method | |
| CN108931621B (en) | Zinc ore grade soft measurement method based on process texture characteristics | |
| CN114998214A (en) | Sampling speed control method and system for cable defect detection | |
| CN115018790A (en) | Workpiece surface defect detection method based on anomaly detection | |
| CN116777836A (en) | Multi-mode data driving-based injection molding process product quality detection method | |
| CN117817111A (en) | Method and system for intelligently identifying and matching process parameters in laser welding | |
| CN111695582A (en) | Flutter texture detection method and device | |
| CN115457323A (en) | Visual inspection system-based method for classifying non-uniform surface defects of mobile phone glass screen | |
| CN113919396A (en) | Vibration signal and image characteristic machine tool cutter wear state monitoring method based on semi-supervised learning | |
| CN117649387B (en) | Defect detection method suitable for object with complex surface texture | |
| CN117474916B (en) | Image detection method, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |