CN116630428B - Pouring position identification method and system based on machine vision - Google Patents
Pouring position identification method and system based on machine vision Download PDFInfo
- Publication number
- CN116630428B CN116630428B CN202310909762.2A CN202310909762A CN116630428B CN 116630428 B CN116630428 B CN 116630428B CN 202310909762 A CN202310909762 A CN 202310909762A CN 116630428 B CN116630428 B CN 116630428B
- Authority
- CN
- China
- Prior art keywords
- casting
- pouring
- solidification
- model
- dimensional grid
- 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
- 238000000034 method Methods 0.000 title claims abstract description 73
- 238000005266 casting Methods 0.000 claims abstract description 309
- 230000008569 process Effects 0.000 claims abstract description 37
- 238000004458 analytical method Methods 0.000 claims abstract description 26
- 238000005457 optimization Methods 0.000 claims abstract description 13
- 238000004088 simulation Methods 0.000 claims abstract description 12
- 238000012545 processing Methods 0.000 claims abstract description 6
- 238000007711 solidification Methods 0.000 claims description 165
- 230000008023 solidification Effects 0.000 claims description 165
- 238000013441 quality evaluation Methods 0.000 claims description 95
- 238000012549 training Methods 0.000 claims description 19
- 239000007788 liquid Substances 0.000 claims description 16
- 238000013528 artificial neural network Methods 0.000 claims description 13
- 238000005192 partition Methods 0.000 claims description 11
- 238000001816 cooling Methods 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 claims description 8
- 238000012216 screening Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 125000004122 cyclic group Chemical group 0.000 claims description 5
- 238000001303 quality assessment method Methods 0.000 claims description 4
- 238000012937 correction Methods 0.000 abstract description 4
- 238000010276 construction Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 230000009467 reduction Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001915 proofreading effect Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000013316 zoning Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B22—CASTING; POWDER METALLURGY
- B22D—CASTING OF METALS; CASTING OF OTHER SUBSTANCES BY THE SAME PROCESSES OR DEVICES
- B22D46/00—Controlling, supervising, not restricted to casting covered by a single main group, e.g. for safety reasons
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B22—CASTING; POWDER METALLURGY
- B22D—CASTING OF METALS; CASTING OF OTHER SUBSTANCES BY THE SAME PROCESSES OR DEVICES
- B22D2/00—Arrangement of indicating or measuring devices, e.g. for temperature or viscosity of the fused mass
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T1/00—General purpose image data processing
- G06T1/0014—Image feed-back for automatic industrial control, e.g. robot with camera
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
- G06T7/75—Determining position or orientation of objects or cameras using feature-based methods involving models
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30116—Casting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/06—Recognition of objects for industrial automation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mechanical Engineering (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Geometry (AREA)
- Computer Graphics (AREA)
- Robotics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a pouring position identification method and a pouring position identification system based on machine vision, which relate to the technical field of image processing, and are used for acquiring mold image information and pouring cup image information, carrying out three-dimensional grid modeling, carrying out pouring position optimization by combining a pouring control process, acquiring a recommended pouring position, and constructing a casting three-dimensional grid first model in a first coordinate system; and acquiring casting image information to construct a casting three-dimensional grid second model, carrying out position deviation analysis to obtain a position deviation coefficient, and if the position deviation coefficient is greater than or equal to a deviation coefficient threshold value, carrying out pouring position adjustment. The casting position setting method solves the technical problems that the casting position in the prior art needs to be set according to experience, has certain subjectivity, cannot guarantee the required compliance of castings and is poor in stability, combines three-dimensional modeling to carry out simulation analysis optimization, determines the recommended casting position and carries out casting deviation analysis, realizes automatic casting position correction by identifying the casting position through machine vision, and improves casting stability.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a pouring position identification method and system based on machine vision.
Background
The pouring position refers to the posture and the position of the casting in the mold during pouring, the quality of the casting is directly influenced by the selection of the pouring position, and the casting method, the casting structure and the like are comprehensively considered. The prior art has the defects that the pouring position needs to be set according to experience, certain subjectivity exists, the required fitting degree of the casting cannot be guaranteed, and the stability is poor.
Disclosure of Invention
The application provides a pouring position identification method and system based on machine vision, which are used for solving the technical problems that the pouring position needs to be set according to experience in the prior art, a certain subjectivity exists, the required conformity of castings cannot be ensured, and the stability is poor.
In view of the above problems, the application provides a pouring position identification method and system based on machine vision.
In a first aspect, the present application provides a machine vision-based pouring location identification method, the method comprising:
based on a vision collector, collecting mold image information and pouring cup image information, and constructing a mold three-dimensional grid model and a pouring cup three-dimensional grid model in a first coordinate system;
based on a pouring control process, combining the three-dimensional grid model of the die and the three-dimensional grid model of the pouring cup to optimize the pouring position, and obtaining a recommended pouring position;
Constructing a casting three-dimensional grid first model in the first coordinate system according to the recommended pouring position;
acquiring casting image information based on the vision collector, and constructing a casting three-dimensional grid second model in the first coordinate system;
performing position deviation analysis on the casting three-dimensional grid first model and the casting three-dimensional grid second model to obtain a position deviation coefficient;
and when the position deviation coefficient is larger than or equal to the deviation coefficient threshold value, carrying out pouring position adjustment on the preset casting based on the vision collector.
In a second aspect, the present application provides a machine vision based pour location identification system, the system comprising:
the image acquisition module is used for acquiring mold image information and pouring cup image information based on the vision acquisition device, and constructing a mold three-dimensional grid model and a pouring cup three-dimensional grid model in a first coordinate system;
the position acquisition module is used for carrying out pouring position optimization by combining the three-dimensional grid model of the die and the three-dimensional grid model of the pouring cup based on a pouring control process to acquire a recommended pouring position;
the first model building module is used for building a casting three-dimensional grid first model in the first coordinate system according to the recommended pouring position;
The second model building module is used for obtaining casting image information based on the vision collector and building a casting three-dimensional grid second model in the first coordinate system;
the position deviation analysis module is used for carrying out position deviation analysis on the casting three-dimensional grid first model and the casting three-dimensional grid second model to obtain a position deviation coefficient;
and the pouring position adjusting module is used for adjusting the pouring position of the preset casting based on the vision collector when the position deviation coefficient is greater than or equal to the deviation coefficient threshold value.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
according to the pouring position identification method based on machine vision, based on a vision collector, die image information and pouring cup image information are collected, a die three-dimensional grid model and a pouring cup three-dimensional grid model are built in a first coordinate system, pouring position optimization is conducted in combination with a pouring control process, a recommended pouring position is obtained, and a casting three-dimensional grid first model is built in the first coordinate system; and acquiring casting image information based on the vision collector, constructing a casting three-dimensional grid second model in the first coordinate system, analyzing the position deviation to acquire a position deviation coefficient, and adjusting the pouring position of the preset casting based on the vision collector if the position deviation coefficient is larger than or equal to a deviation coefficient threshold. The casting position setting method solves the technical problems that the casting position in the prior art needs to be set according to experience, has certain subjectivity, cannot guarantee the required compliance of castings and is poor in stability, combines three-dimensional modeling to carry out simulation analysis optimization, determines the recommended casting position and carries out casting deviation analysis, realizes automatic casting position correction by identifying the casting position through machine vision, and improves casting stability.
Drawings
FIG. 1 is a schematic flow chart of a pouring position identification method based on machine vision;
FIG. 2 is a schematic diagram of a recommended pouring position acquisition flow in a machine vision-based pouring position identification method according to the present application;
FIG. 3 is a schematic diagram of a flow chart for obtaining M casting quality evaluation results in a pouring position identification method based on machine vision;
fig. 4 is a schematic structural diagram of a pouring position recognition system based on machine vision.
Reference numerals illustrate: the device comprises an image acquisition module 11, a position acquisition module 12, a first model construction module 13, a second model construction module 14, a position deviation analysis module 15 and a pouring position adjustment module 16.
Detailed Description
The application provides a pouring position identification method and a pouring position identification system based on machine vision, which are used for acquiring mold image information and pouring cup image information, carrying out three-dimensional grid modeling, carrying out pouring position optimization by combining a pouring control process, acquiring a recommended pouring position, and constructing a casting three-dimensional grid first model in a first coordinate system; the casting three-dimensional grid second model is constructed by collecting casting image information, position deviation analysis is carried out to obtain a position deviation coefficient, and pouring position adjustment is carried out if the position deviation coefficient is larger than or equal to a deviation coefficient threshold value, so that the technical problems that the pouring position in the prior art needs to be set according to experience, a certain subjectivity exists, the required fit degree of the casting cannot be guaranteed, and the stability is poor are solved.
Example 1
As shown in fig. 1, the present application provides a machine vision-based pouring position identification method, which includes:
step S100: based on a vision collector, collecting mold image information and pouring cup image information, and constructing a mold three-dimensional grid model and a pouring cup three-dimensional grid model in a first coordinate system;
specifically, the pouring position refers to the posture and the position of a casting in a mold during pouring, the quality of the casting is directly influenced by the selection of the pouring position, and the comprehensive consideration of a pouring method, a casting structure and the like is needed.
Specifically, the vision collector is a tool for collecting images, and based on the vision collector, the mold and the pouring cup are subjected to multi-angle image collection so as to ensure the information coverage completeness of the collected images, and the collected images are sequentially integrated based on the change of the collection angles to obtain the image information of the mold and the image information of the pouring cup. Furthermore, the casting space is taken as a coordinate space, and the space dimension is taken as a coordinate axis, so that the first coordinate system is built. And constructing a three-dimensional grid model based on the mold image and the pouring cup image. The method comprises the steps of determining a plurality of key positioning points based on the mold and the pouring cup, performing space positions of the corresponding key positioning points based on the mold image information, determining space coordinates, performing positioning of the coordinate positions in the first coordinate system, and performing three-dimensional reduction of the mold based on the space coordinates and the coordinate positions by combining the mold image information to obtain a three-dimensional network model of the mold; and similarly, combining the pouring cup image information, and carrying out positioning reduction in the first coordinate system to generate the pouring cup three-dimensional grid model. The three-dimensional grid model of the mold and the three-dimensional grid model of the pouring cup are used as the basis for pouring position analysis.
Step S200: based on a pouring control process, combining the three-dimensional grid model of the die and the three-dimensional grid model of the pouring cup to optimize the pouring position, and obtaining a recommended pouring position;
further, as shown in fig. 2, based on the pouring control process, the pouring position is optimized by combining the three-dimensional grid model of the mold and the three-dimensional grid model of the pouring cup, so as to obtain a recommended pouring position, and step S200 of the present application further includes:
step S210: the pouring control process comprises a liquid column flow time sequence parameter, a pouring cup grid position time sequence parameter, a liquid column temperature time sequence parameter and a cooling process time sequence parameter;
step S220: based on the mold three-dimensional grid model and the pouring cup three-dimensional grid model, combining the liquid column flow time sequence parameter, the pouring cup grid position time sequence parameter, the liquid column temperature time sequence parameter and the cooling process time sequence parameter to construct a digital twin pouring model;
step S230: downloading pouring log data according to the mould three-dimensional grid model and the pouring cup three-dimensional grid model, wherein the pouring log data comprises M pouring position record data;
step S240: based on the M pouring position record data, performing simulation pouring on the digital twin pouring model to obtain M groups of casting position solidification sequence parameters;
Step S250: and optimizing the pouring position based on the M groups of casting position solidification sequence parameters, and obtaining the recommended pouring position.
Specifically, the pouring control process is called, parameter characteristic identification extraction is carried out on the pouring control process, the liquid column flow time sequence parameter, the pouring cup grid position time sequence parameter, the liquid column temperature time sequence parameter and the cooling process time sequence parameter can be directly obtained, namely the multi-element control parameters for completing pouring are obtained, and the parameter time sequences correspond to each other. Further, based on the three-dimensional grid model of the mold and the three-dimensional grid model of the pouring cup, combining the liquid column flow time sequence parameter, the pouring cup grid position time sequence parameter, the liquid column temperature time sequence parameter and the cooling process time sequence parameter as pouring simulation data, performing simulated reduction of a pouring process in a simulated vacuum chamber, generating the digital twin pouring model with the hardware structure consistent with soft control parameters, and positioning through the three-dimensional grid model, wherein any grid comprises a preset number of pixels, and the default value is 1. And the digital twin casting model is used for simulation casting, so that the flexibility adjustment can be performed based on the operation requirement, and meanwhile, the cost loss existing in the actual operation test can be avoided.
Further, the downloading of the pouring log data is performed based on the three-dimensional grid model of the mold and the three-dimensional grid model of the pouring cup, namely, the pouring log data comprises the M pouring position record data based on the once pouring records of the mold and the pouring cup. And based on the M pouring position record data, performing simulation pouring in the digital twin pouring model, and performing simulation pouring monitoring to obtain the M casting position solidification sequence parameters. And carrying out pouring position optimization by taking the M casting position solidification sequence parameters as references. Determining the sequence of pouring positions through simulation pouring, and then analyzing the sequence of pouring positions by combining with a casting quality evaluation model to predict the pouring quality.
Further, the step S250 of the present application further includes optimizing the pouring position based on the solidification sequence parameters of the M sets of casting positions, and obtaining the recommended pouring position:
step S251: traversing the M groups of casting position solidification sequence parameters to evaluate casting quality, and obtaining M casting quality evaluation results;
step S252: when the M casting quality evaluation results meet the number of casting quality evaluation thresholds to be zero, screening N groups of casting position solidification sequence parameters from large to small based on the M casting quality evaluation results to perform position neighborhood variation, and acquiring O groups of casting position solidification sequence expansion data;
Step S253: traversing the O-group casting position solidification sequence expansion data to evaluate casting quality, and obtaining O casting quality evaluation results;
step S254: when any one of the O casting quality evaluation results meets the casting quality evaluation threshold, acquiring the recommended pouring position;
step S255: and repeating the position neighborhood variation when the number of the O casting quality evaluation results meeting the casting quality evaluation threshold is zero.
Specifically, traversing the M groups of casting position solidification sequence parameters, carrying out solidification partition of each group of casting position solidification sequence parameters based on casting position solidification time sequence, inputting the solidification partition into a constructed casting quality evaluation model for quality evaluation, and obtaining M casting quality evaluation results corresponding to the parameters. Setting the casting quality evaluation threshold value, and checking the M casting quality evaluation results, wherein if the M casting quality evaluation results do not meet the casting quality evaluation threshold value, the casting products of the casting position solidification sequence parameters are unqualified. And screening N groups of casting position solidification sequence parameters from large to small based on the M casting quality evaluation results, and performing position neighborhood variation on the N groups of casting position solidification sequence parameters, namely performing random adjustment on the casting position sequence based on the number of the sequences adjusted front and back, so as to obtain the O groups of casting position solidification sequence expansion data.
And further, respectively carrying out casting quality evaluation on the O-group casting position solidification sequence expansion data by combining the casting quality evaluation model, and acquiring the O-group casting quality evaluation results by the same specific analysis steps. Further checking the quality evaluation results of the O castings with the quality evaluation threshold value of the castings, and taking any one of the quality evaluation results of the O castings as the recommended pouring position to ensure the qualification degree of the recommended pouring position if any one of the quality evaluation results of the O castings meets the quality evaluation threshold value of the castings; and if all the O casting quality evaluation results do not meet the casting quality evaluation threshold, carrying out positive sequencing sorting screening and position neighborhood variation on the O casting quality evaluation results, carrying out expansion and quality evaluation again, and repeating the steps until at least one casting quality evaluation result meets the casting quality threshold, and taking the casting quality evaluation result as the recommended pouring position.
Further, as shown in fig. 3, the casting quality evaluation is performed by traversing the M sets of casting position solidification sequence parameters, and M casting quality evaluation results are obtained, where step S251 of the present application further includes:
step S2511: acquiring the position solidification sequence parameters of the M-th group of castings according to the position solidification sequence parameters of the M-th group of castings;
Step S2512: according to the casting position solidification time sequence, carrying out solidification partition on the m-th group of casting position solidification sequence parameters to obtain a first sequence solidification region, a second sequence solidification region and a Y-th sequence solidification region;
step S2513: and inputting the casting quality evaluation model from the first sequential solidification region to the second sequential solidification region to the Y sequential solidification region by combining the pouring control process, and obtaining the M casting quality evaluation results.
Further, inputting the casting quality evaluation model from the first sequential solidification region to the second sequential solidification region to the Y sequential solidification region in combination with the pouring control process to obtain the M casting quality evaluation results, and the step S2513 further includes:
step S25131: collecting pouring record data, wherein the pouring record data comprises quality record data before pouring, solidification area identification data, pouring process record data and quality record data after pouring, and training a reference predictor based on a BP neural network;
step S25132: acquiring joint training record data, wherein the joint training record data comprises casting position solidification sequence record data and casting process record data;
Step S25133: according to the casting position solidification time sequence, carrying out solidification partition on the casting position solidification sequence record data to obtain a first sequence solidification record area, a second sequence solidification record area and an H sequence solidification record area;
step S25134: in the cyclic neural network, building H hidden layers connected in series based on the reference predictor, wherein the reference predictor is a processor of any hidden layer;
step S25135: and inputting the first sequential solidification recording area, the second sequential solidification recording area to the H sequential solidification recording area and casting process recording data into the circulating neural network for joint training, and obtaining the casting quality evaluation model.
Specifically, based on the M sets of casting position solidification sequence parameters, one set of casting position solidification sequence parameters is extracted to serve as the M-th set of casting position solidification sequence parameters. And based on the casting position solidification time sequence, carrying out solidification partition on the m-th group of casting position solidification sequence parameters, determining a solidification coverage area under the same time sequence, and obtaining the first sequential solidification area, the second sequential solidification area and the Y-th sequential solidification area. And further constructing the casting quality evaluation model, and carrying out pouring quality prediction by combining the first sequential solidification region and the second sequential solidification region until the Y-th sequential solidification region.
Specifically, the pre-pouring quality record data, the solidification area identification data, the pouring process record data and the post-pouring quality record data are collected and used as the pouring record data, mapping association among all groups of data is carried out to be used as training data, and BP neural network training is carried out by taking the mapping association as a reference to generate the reference predictor. And further calling the casting position solidification sequence record data, combining the casting process record data to serve as the combined training record data, wherein the casting record data and the combined training record data can be called and determined based on data retrieval. Further, based on the casting position solidification time sequence, based on time sequence transition and the time sequence, solidification partitioning is carried out, and the first sequential solidification recording area, the second sequential solidification recording area and the H sequential solidification recording area are obtained. Since each subsequent casting quality evaluation needs to be applied to the previous casting quality, a recurrent neural network with memory is used as a model macroscopic frame. And taking the reference predictor as a general processor, and building the H hidden layers connected in series in the cyclic neural network, wherein the H hidden layers correspond to the H sequential solidification recording areas.
Further, the first sequential solidification recording region, the second sequential solidification recording region, up to the H sequential solidification recording region, and casting process recording data are input into the cyclic neural network, stepwise combined training is performed based on the serial sequence of hidden layers, for example, compactness, uniform metallographic structure, dimensional accuracy, and the like are used as indexes for performing quality assessment, and the casting quality assessment model after training is obtained. The input of each hidden layer is the output of the last hidden layer, the solidification area corresponding to the sequence and the casting technological parameters.
And further, inputting a casting quality evaluation model from the first sequential solidification region to the second sequential solidification region to the Y sequential solidification region by combining the pouring control process, and obtaining a casting quality evaluation result of the m-th group of casting position solidification sequence parameters. And similarly, respectively carrying out solidification partition on the M groups of casting position solidification sequence parameters, and carrying out quality evaluation by combining with the casting quality evaluation model. Before quality detection is performed on the zoning results of the solidification sequence parameters of different groups of castings, a plurality of hidden layers which are connected in series and correspond to the number of hidden layers are built according to the sequence of the solidification areas, and the method is applicable to pouring quality analysis of different parameters and can acquire the quality evaluation results of the M castings.
Further, when the number of the M casting quality evaluation results satisfying the casting quality evaluation threshold is zero, screening N sets of casting position solidification sequence parameters from large to small based on the M casting quality evaluation results to perform position neighborhood variation, and obtaining O sets of casting position solidification sequence expansion data, step S252 of the present application further includes:
step S2521: setting a position variation neighborhood region, wherein the position variation neighborhood region represents a sequence region with a position capable of being adjusted back and forth;
step S2522: and traversing the N groups of casting position solidification sequence parameters, and randomly adjusting the position solidification sequence based on the position variation neighborhood region to obtain the O groups of casting position solidification sequence expansion data.
Specifically, the casting quality evaluation threshold value is set, namely critical quality data for measuring the qualified casting quality. And (3) correcting the M casting quality evaluation results and the casting quality evaluation threshold, counting the number of the casting quality evaluation threshold, and if the number is zero, performing expansion optimization of the casting position solidification sequence. Specifically, the quality evaluation results of the M castings are subjected to positive sequencing from large to small, N groups of casting position solidification sequence parameters are screened, and N is a positive integer smaller than M. And carrying out position neighborhood variation on the N groups of casting position solidification sequence parameters so as to carry out parameter expansion.
Specifically, the sequence interval in which the positions can be adjusted back and forth is defined, that is, the number of position sequences that can be adjusted back and forth is used as the position variation neighborhood interval. And randomly adjusting the sequence under the constraint of the N groups of casting position solidification sequence parameters by taking the position variation neighborhood region as the constraint, integrating and regulating the adjusted casting position solidification sequence parameters, and generating the O groups of casting position solidification sequence expansion data. And further carrying out casting quality evaluation on the casting, and obtaining casting position solidification sequence expansion data meeting casting quality evaluation threshold.
Step S300: constructing a casting three-dimensional grid first model in the first coordinate system according to the recommended pouring position;
step S400: acquiring casting image information based on the vision collector, and constructing a casting three-dimensional grid second model in the first coordinate system;
specifically, the recommended pouring position is a determined pouring position meeting pouring requirements, gridding modeling of the casting is conducted in the first coordinate system based on the recommended pouring position, the casting three-dimensional grid first model is generated, and the casting grid three-dimensional model is a casting in an ideal state determined based on the pouring position.
And further, carrying out pouring operation based on the recommended pouring position, and carrying out multi-angle image acquisition on the poured casting based on the vision acquisition device to acquire the image information of the casting. Based on the casting image information. And extracting casting characteristics based on the casting image information, and performing modeling reduction in the first coordinate system to generate a casting three-dimensional grid second model, wherein the casting three-dimensional grid second model is a casting in an actual state determined based on the pouring position actual operation. Because the casting is affected by uncontrollable factors in the casting process, the expected casting and the actually produced casting have certain differences, and the casting position is corrected based on the differences.
Step S500: performing position deviation analysis on the casting three-dimensional grid first model and the casting three-dimensional grid second model to obtain a position deviation coefficient;
step S600: and when the position deviation coefficient is larger than or equal to the deviation coefficient threshold value, carrying out pouring position adjustment on the preset casting based on the vision collector.
Specifically, based on the casting three-dimensional grid first model and the casting three-dimensional grid second model, grid coordinate extraction and coordinate mapping correspondence are respectively carried out, a plurality of coordinate sets are determined grid by grid, grid offset distance calculation is carried out, and a grid offset distance set is obtained. And further, carrying out classification and quantity statistics on the grid deviation distance set by combining with a set distance deviation threshold value, and calculating and obtaining the position deviation coefficient by combining with a deviation coefficient evaluation formula.
And further setting the deviation coefficient threshold, namely, customizing the set critical deviation coefficient based on the casting precision requirement. The position deviation coefficient is calibrated with the deviation coefficient threshold value, if the position deviation coefficient is smaller than the deviation coefficient threshold value, the position deviation coefficient is indicated to be in an error allowable range of castings obtained by actual pouring operation, and adjustment is not needed; and if the position deviation coefficient is larger than or equal to the deviation coefficient threshold value, indicating that the pouring deviation of the casting is overlarge, taking the grid deviation distance set as a correction standard, combining the vision collector to perform pouring monitoring, and realizing automatic pouring position correction by identifying the pouring position through machine vision, thereby improving the pouring stability.
Further, the step S500 of the present application further includes performing a position deviation analysis on the first model of the casting three-dimensional grid and the second model of the casting three-dimensional grid to obtain a position deviation coefficient:
step S510: acquiring a first grid coordinate set and a second grid coordinate set of the casting three-dimensional grid first model and the casting three-dimensional grid second model, wherein the first grid coordinate set and the second grid coordinate set are in one-to-one correspondence;
Step S520: performing coordinate distance calculation according to the first grid coordinate set and the second grid coordinate set to obtain a grid deviation distance set;
step S530: constructing a deviation coefficient evaluation formula:the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Characterizing the ith grid offset distance, +.>Characterizing a distance deviation threshold, ">Characterizing the number of grid deviation distances less than or equal to the distance deviation threshold, +.>Characterizing the number of grid deviation distances greater than a distance deviation threshold, and characterizing a position deviation coefficient;
step S540: and processing the grid deviation distance set according to the deviation coefficient evaluation formula to acquire the position deviation coefficient.
Specifically, grid coordinates of the casting three-dimensional grid first model and the casting three-dimensional grid second model in the first coordinate system are recognized regularly, and the first grid coordinate set and the second grid coordinate set are generated. Further mapping the first grid coordinate set and the second grid coordinate set, calculating the difference between the horizontal coordinate and the vertical coordinate of the mapped coordinates, and obtaining the distance between the corresponding grid coordinates based on the calculation to serve as a grid deviation distance; and calculating grid by grid to obtain the grid deviation distance set.
Further, a distance deviation threshold is set, i.e. a preset critical deviation distance for measuring whether the distance is within the allowable limit. Traversing the grid deviation distance set, performing proofreading and dividing with the distance deviation threshold, and counting the number of the distance deviation threshold which is larger than or equal to the distance deviation threshold. And then combining the deviation coefficient evaluation formula:performing position deviation coefficient calculation, wherein +_>Characterization of the ith grid biasDistance from (I)>Characterizing a distance deviation threshold, ">Characterizing the number of grid deviation distances less than or equal to the distance deviation threshold, +.>The number of the grid deviation distances larger than the distance deviation threshold value is represented, the position deviation coefficient is represented by D, and the parameters can be obtained based on the earlier processing of the embodiment of the application.
The pouring position identification method based on machine vision provided by the embodiment of the application has the following technical effects:
1. and (3) carrying out three-dimensional grid modeling on the mould and the pouring cup, carrying out twin simulation pouring by combining a pouring process, carrying out quality evaluation and expansion analysis on the simulation pouring casting by taking the casting position solidification sequence as a reference, and carrying out recommended pouring positions meeting the casting quality requirement by optimizing the pouring positions.
2. Dividing the area based on the solidification sequence, constructing a series predictor according to the sequence of the solidification area as a casting quality evaluation model so as to adapt to the casting analysis requirements under different states, and carrying out adaptability adjustment so as to ensure the model adaptation degree.
3. Modeling and positioning analysis are carried out by taking the three-dimensional grid as a reference, so that the positioning accuracy can be effectively improved.
Example two
Based on the same inventive concept as the machine vision-based pouring position recognition method in the foregoing embodiment, as shown in fig. 4, the present application provides a machine vision-based pouring position recognition system, which includes:
the image acquisition module 11 is used for acquiring mold image information and pouring cup image information based on a vision acquisition device, and constructing a mold three-dimensional grid model and a pouring cup three-dimensional grid model in a first coordinate system;
the position acquisition module 12 is used for carrying out pouring position optimization by combining the mold three-dimensional grid model and the pouring cup three-dimensional grid model based on a pouring control process, so as to acquire a recommended pouring position;
a first model building module 13, wherein the first model building module 13 is used for building a casting three-dimensional grid first model in the first coordinate system according to the recommended pouring position;
A second model building module 14, wherein the second model building module 14 is used for obtaining casting image information based on the vision collector and building a casting three-dimensional grid second model in the first coordinate system;
the position deviation analysis module 15 is used for carrying out position deviation analysis on the first model of the casting three-dimensional grid and the second model of the casting three-dimensional grid to obtain a position deviation coefficient;
and the pouring position adjusting module 16 is used for adjusting the pouring position of the preset casting based on the vision collector when the position deviation coefficient is greater than or equal to the deviation coefficient threshold value.
Further, the location acquisition module 12 further includes:
the process analysis module is used for controlling the pouring control process to comprise a liquid column flow time sequence parameter, a pouring cup grid position time sequence parameter, a liquid column temperature time sequence parameter and a cooling process time sequence parameter;
the model construction module is used for constructing a digital twin pouring model based on the mould three-dimensional grid model and the pouring cup three-dimensional grid model by combining the liquid column flow time sequence parameter, the pouring cup grid position time sequence parameter, the liquid column temperature time sequence parameter and the cooling process time sequence parameter;
The data downloading module is used for downloading pouring log data according to the mould three-dimensional grid model and the pouring cup three-dimensional grid model, wherein the pouring log data comprises M pouring position record data;
the parameter acquisition module is used for carrying out simulation pouring on the digital twin pouring model based on the M pouring position record data to acquire M groups of casting position solidification sequence parameters;
and the position optimization module is used for optimizing the pouring position based on the M groups of casting position solidification sequence parameters and obtaining the recommended pouring position.
Further, the location optimization module further includes:
the quality evaluation module is used for traversing the M groups of casting position solidification sequence parameters to evaluate the casting quality and obtaining M casting quality evaluation results;
the expansion data acquisition module is used for carrying out position neighborhood variation on the basis of the M casting quality evaluation results and screening N groups of casting position solidification sequence parameters from large to small when the M casting quality evaluation results meet the number of casting quality evaluation thresholds to be zero, so as to acquire O groups of casting position solidification sequence expansion data;
The casting quality evaluation module is used for traversing the O-group casting position solidification sequence expansion data to evaluate the casting quality and obtain O casting quality evaluation results;
the threshold judging module is used for acquiring the recommended pouring position when any one of the casting quality evaluation results meets the casting quality evaluation threshold;
and the position neighborhood mutation module is used for repeatedly carrying out position neighborhood mutation when the number of the O casting quality evaluation results meeting the casting quality evaluation threshold is zero.
Further, the quality evaluation module further includes:
the casting position solidification sequence parameter acquisition module is used for acquiring the M-th group of casting position solidification sequence parameters according to the M-group casting position solidification sequence parameters;
the solidification partition module is used for carrying out solidification partition on the position solidification sequence parameters of the m-th group of castings according to the casting position solidification time sequence to obtain a first sequence solidification region, a second sequence solidification region and a Y-th sequence solidification region;
And the casting quality evaluation result acquisition module is used for inputting casting quality evaluation models into the casting quality evaluation models by combining the pouring control process from the first sequential solidification region to the second sequential solidification region to the Y sequential solidification region to acquire the M casting quality evaluation results.
Further, the casting quality evaluation result obtaining module further comprises:
the pouring record acquisition module is used for acquiring pouring record data, wherein the pouring record data comprise pre-pouring quality record data, solidification area identification data, pouring process record data and post-pouring quality record data, and a reference predictor is trained based on a BP neural network;
the training record acquisition module is used for acquiring combined training record data, wherein the combined training record data comprises casting position solidification sequence record data and casting process record data;
the region acquisition module is used for carrying out solidification partition on the casting position solidification sequence record data according to the casting position solidification time sequence to acquire a first sequence solidification record region, a second sequence solidification record region and an H sequence solidification record region;
The hidden layer building module is used for building H hidden layers connected in series in the cyclic neural network based on the reference predictor, wherein the reference predictor is a processor of any hidden layer;
the casting quality evaluation model training module is used for inputting the first sequential solidification recording area, the second sequential solidification recording area to the H sequential solidification recording area and casting process recording data into the circulating neural network for joint training, and obtaining the casting quality evaluation model.
Further, the extended data acquisition module further includes:
the interval setting module is used for setting a position variation neighborhood interval, wherein the position variation neighborhood interval represents a sequence interval of which the position can be adjusted back and forth;
the casting position solidification sequence expansion data acquisition module is used for traversing the N groups of casting position solidification sequence parameters, and carrying out position solidification sequence random adjustment based on the position variation neighborhood region to acquire the O groups of casting position solidification sequence expansion data.
Further, the positional deviation analysis module 15 further includes:
the coordinate set acquisition module is used for acquiring a first grid coordinate set and a second grid coordinate set of the casting three-dimensional grid first model and the casting three-dimensional grid second model, wherein the first grid coordinate set and the second grid coordinate set are in one-to-one correspondence;
the distance calculation module is used for calculating coordinate distances according to the first grid coordinate set and the second grid coordinate set, and acquiring a grid deviation distance set;
the formula construction module is used for constructing a deviation coefficient evaluation formula:the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Characterizing the ith grid offset distance, +.>A distance deviation threshold value is characterized,characterizing the number of grid deviation distances less than or equal to the distance deviation threshold, +.>Characterizing the number of grid deviation distances greater than a distance deviation threshold, and characterizing a position deviation coefficient;
and the position deviation coefficient acquisition module is used for processing the grid deviation distance set according to the deviation coefficient evaluation formula to acquire the position deviation coefficient.
The foregoing detailed description of the pouring position identification method based on machine vision will be clear to those skilled in the art, and the pouring position identification method and system based on machine vision in this embodiment, for the apparatus disclosed in the embodiments, corresponds to the method disclosed in the embodiments, so that the description is relatively simple, and relevant places refer to the method part.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. The pouring position identification method based on machine vision is characterized by comprising the following steps of:
based on a vision collector, collecting mold image information and pouring cup image information, and constructing a mold three-dimensional grid model and a pouring cup three-dimensional grid model in a first coordinate system;
based on a pouring control process, combining the three-dimensional grid model of the die and the three-dimensional grid model of the pouring cup to optimize the pouring position, and obtaining a recommended pouring position;
constructing a casting three-dimensional grid first model in the first coordinate system according to the recommended pouring position;
acquiring casting image information based on the vision collector, and constructing a casting three-dimensional grid second model in the first coordinate system;
Performing position deviation analysis on the casting three-dimensional grid first model and the casting three-dimensional grid second model to obtain a position deviation coefficient;
and when the position deviation coefficient is larger than or equal to the deviation coefficient threshold value, carrying out pouring position adjustment on the preset casting based on the vision collector.
2. The method of claim 1, wherein performing a pouring location optimization based on a pouring control process in conjunction with the mold three-dimensional mesh model and the cup three-dimensional mesh model to obtain a recommended pouring location comprises:
the pouring control process comprises a liquid column flow time sequence parameter, a pouring cup grid position time sequence parameter, a liquid column temperature time sequence parameter and a cooling process time sequence parameter;
based on the mold three-dimensional grid model and the pouring cup three-dimensional grid model, combining the liquid column flow time sequence parameter, the pouring cup grid position time sequence parameter, the liquid column temperature time sequence parameter and the cooling process time sequence parameter to construct a digital twin pouring model;
downloading pouring log data according to the mould three-dimensional grid model and the pouring cup three-dimensional grid model, wherein the pouring log data comprises M pouring position record data;
Based on the M pouring position record data, performing simulation pouring on the digital twin pouring model to obtain M groups of casting position solidification sequence parameters;
and optimizing the pouring position based on the M groups of casting position solidification sequence parameters, and obtaining the recommended pouring position.
3. The method of claim 2, wherein optimizing the pouring location based on the M sets of casting location solidification sequence parameters, the obtaining the recommended pouring location comprises:
traversing the M groups of casting position solidification sequence parameters to evaluate casting quality, and obtaining M casting quality evaluation results;
when the M casting quality evaluation results meet the number of casting quality evaluation thresholds to be zero, screening N groups of casting position solidification sequence parameters from large to small based on the M casting quality evaluation results to perform position neighborhood variation, and acquiring O groups of casting position solidification sequence expansion data;
traversing the O-group casting position solidification sequence expansion data to evaluate casting quality, and obtaining O casting quality evaluation results;
when any one of the O casting quality evaluation results meets the casting quality evaluation threshold, acquiring the recommended pouring position;
And repeating the position neighborhood variation when the number of the O casting quality evaluation results meeting the casting quality evaluation threshold is zero.
4. A method according to claim 3, wherein traversing the M sets of casting position solidification sequence parameters for casting quality assessment, obtaining M casting quality assessment results, comprises:
acquiring the position solidification sequence parameters of the M-th group of castings according to the position solidification sequence parameters of the M-th group of castings;
according to the casting position solidification time sequence, carrying out solidification partition on the m-th group of casting position solidification sequence parameters to obtain a first sequence solidification region, a second sequence solidification region and a Y-th sequence solidification region;
and inputting the casting quality evaluation model from the first sequential solidification region to the second sequential solidification region to the Y sequential solidification region by combining the pouring control process, and obtaining the M casting quality evaluation results.
5. The method of claim 4, wherein inputting the first sequential solidification region, the second sequential solidification region, and up to the Y-th sequential solidification region in conjunction with the pouring control process into a casting quality evaluation model to obtain the M casting quality evaluation results, further comprising:
Collecting pouring record data, wherein the pouring record data comprises quality record data before pouring, solidification area identification data, pouring process record data and quality record data after pouring, and training a reference predictor based on a BP neural network;
acquiring joint training record data, wherein the joint training record data comprises casting position solidification sequence record data and casting process record data;
according to the casting position solidification time sequence, carrying out solidification partition on the casting position solidification sequence record data to obtain a first sequence solidification record area, a second sequence solidification record area and an H sequence solidification record area;
in the cyclic neural network, building H hidden layers connected in series based on the reference predictor, wherein the reference predictor is a processor of any hidden layer;
and inputting the first sequential solidification recording area, the second sequential solidification recording area to the H sequential solidification recording area and casting process recording data into the circulating neural network for joint training, and obtaining the casting quality evaluation model.
6. The method of claim 3, wherein when the number of M casting quality evaluation results satisfying the casting quality evaluation threshold is zero, performing a position neighborhood variation based on the M casting quality evaluation results from large to small screening of N sets of casting position solidification order parameters, obtaining O sets of casting position solidification order expansion data, comprising:
Setting a position variation neighborhood region, wherein the position variation neighborhood region represents a sequence region with a position capable of being adjusted back and forth;
and traversing the N groups of casting position solidification sequence parameters, and randomly adjusting the position solidification sequence based on the position variation neighborhood region to obtain the O groups of casting position solidification sequence expansion data.
7. The method of claim 1, wherein performing a position deviation analysis on the first model of the casting three-dimensional grid and the second model of the casting three-dimensional grid to obtain a position deviation coefficient comprises:
acquiring a first grid coordinate set and a second grid coordinate set of the casting three-dimensional grid first model and the casting three-dimensional grid second model, wherein the first grid coordinate set and the second grid coordinate set are in one-to-one correspondence;
performing coordinate distance calculation according to the first grid coordinate set and the second grid coordinate set to obtain a grid deviation distance set;
constructing a deviation coefficient evaluation formula:the method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Characterizing the ith grid offset distance, +.>Characterizing a distance deviation threshold, ">Characterizing the number of grid deviation distances less than or equal to the distance deviation threshold, +.>Characterizing the number of grid deviation distances greater than a distance deviation threshold, and characterizing a position deviation coefficient;
And processing the grid deviation distance set according to the deviation coefficient evaluation formula to acquire the position deviation coefficient.
8. Pouring position recognition system based on machine vision, characterized by comprising:
the image acquisition module is used for acquiring mold image information and pouring cup image information based on the vision acquisition device, and constructing a mold three-dimensional grid model and a pouring cup three-dimensional grid model in a first coordinate system;
the position acquisition module is used for carrying out pouring position optimization by combining the three-dimensional grid model of the die and the three-dimensional grid model of the pouring cup based on a pouring control process to acquire a recommended pouring position;
the first model building module is used for building a casting three-dimensional grid first model in the first coordinate system according to the recommended pouring position;
the second model building module is used for obtaining casting image information based on the vision collector and building a casting three-dimensional grid second model in the first coordinate system;
the position deviation analysis module is used for carrying out position deviation analysis on the casting three-dimensional grid first model and the casting three-dimensional grid second model to obtain a position deviation coefficient;
And the pouring position adjusting module is used for adjusting the pouring position of the preset casting based on the vision collector when the position deviation coefficient is greater than or equal to the deviation coefficient threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310909762.2A CN116630428B (en) | 2023-07-24 | 2023-07-24 | Pouring position identification method and system based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310909762.2A CN116630428B (en) | 2023-07-24 | 2023-07-24 | Pouring position identification method and system based on machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116630428A CN116630428A (en) | 2023-08-22 |
| CN116630428B true CN116630428B (en) | 2023-10-03 |
Family
ID=87592491
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310909762.2A Active CN116630428B (en) | 2023-07-24 | 2023-07-24 | Pouring position identification method and system based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116630428B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116975770A (en) * | 2023-09-22 | 2023-10-31 | 南通钢安机械制造有限公司 | Cast steel defect detection method and system |
| CN117008557B (en) * | 2023-09-28 | 2023-12-15 | 苏州顶材新材料有限公司 | Production control method and system for blending type interpenetrating network thermoplastic elastomer |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115906557A (en) * | 2022-10-25 | 2023-04-04 | 吉林大学 | Intelligent detection method for judging forming defects by outline deviation of blank of stamping part |
| CN116167236A (en) * | 2023-03-03 | 2023-05-26 | 上海弘智金属制品有限公司 | Intelligent optimization method and system for aluminum alloy die casting molding process |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8661752B2 (en) * | 2006-09-21 | 2014-03-04 | Ahmed Phuly | Foundation with slab, pedestal and ribs for columns and towers |
-
2023
- 2023-07-24 CN CN202310909762.2A patent/CN116630428B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115906557A (en) * | 2022-10-25 | 2023-04-04 | 吉林大学 | Intelligent detection method for judging forming defects by outline deviation of blank of stamping part |
| CN116167236A (en) * | 2023-03-03 | 2023-05-26 | 上海弘智金属制品有限公司 | Intelligent optimization method and system for aluminum alloy die casting molding process |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116630428A (en) | 2023-08-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116630428B (en) | Pouring position identification method and system based on machine vision | |
| CN109492940A (en) | A kind of casting furnace lodge follow-up of quality method for moulding pouring product line | |
| CN115659284B (en) | Big data fusion platform | |
| CN115860592B (en) | Quality detection and evaluation method and system for gypsum block | |
| CN117274259A (en) | Keyboard production equipment management method and system based on Internet of things | |
| CN114462646B (en) | Pole number plate identification method and system based on contact network safety inspection | |
| CN116050678A (en) | Die-casting product processing test system and method based on cloud computing | |
| CN114564880A (en) | Method for constructing digital twin module in additive manufacturing process | |
| CN115625317A (en) | Surface water wave optimization processing method and system for die-casting regulation | |
| CN112906155A (en) | Virtual measurement method for injection molding product information | |
| CN116295141A (en) | Fuel cell bipolar plate flatness detection method and system | |
| CN113408916B (en) | Fire-fighting facility detection and field acceptance assessment system based on intelligent AI and mobile APP | |
| CN116227149A (en) | System and method for detecting matching degree of die products | |
| CN114170427A (en) | Wireless microwave rain attenuation model SSIM image similarity evaluation method based on rain cells | |
| CN111627018B (en) | Steel plate surface defect classification method based on double-flow neural network model | |
| CN114494240A (en) | Ballastless track slab crack measurement method based on multi-scale cooperation deep learning | |
| CN116452840B (en) | Automobile part assembly position vision checking system based on numerical control machine | |
| CN116728783B (en) | Simulation method and system based on 3D printer | |
| CN113628182B (en) | Automatic fish weight estimation method and device, electronic equipment and storage medium | |
| CN117173364B (en) | Slicing and plotting method and system based on building three-dimensional plotting | |
| CN117171922B (en) | Method and system for parallel correction in steel structure manufacturing | |
| CN115034370A (en) | Method for predicting activity of blast furnace hearth based on BP network model | |
| CN116434066B (en) | Deep learning-based soybean pod seed test method, system and device | |
| CN116824093A (en) | Auxiliary three-dimensional model construction method based on point cloud | |
| CN115524156A (en) | Sampling control method and system |
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 |