CN114519797A - Casting process control method and equipment based on artificial intelligence technology - Google Patents
Casting process control method and equipment based on artificial intelligence technology Download PDFInfo
- Publication number
- CN114519797A CN114519797A CN202210111350.XA CN202210111350A CN114519797A CN 114519797 A CN114519797 A CN 114519797A CN 202210111350 A CN202210111350 A CN 202210111350A CN 114519797 A CN114519797 A CN 114519797A
- Authority
- CN
- China
- Prior art keywords
- casting
- defect
- image
- determining
- model
- 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.)
- Pending
Links
- 238000005266 casting Methods 0.000 title claims abstract description 376
- 238000000034 method Methods 0.000 title claims abstract description 91
- 238000005516 engineering process Methods 0.000 title claims abstract description 17
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 16
- 230000007547 defect Effects 0.000 claims abstract description 332
- 238000012545 processing Methods 0.000 claims abstract description 61
- 238000004886 process control Methods 0.000 claims abstract description 40
- 238000003754 machining Methods 0.000 claims abstract description 12
- 238000012549 training Methods 0.000 claims abstract description 7
- 230000008569 process Effects 0.000 claims description 54
- 238000004519 manufacturing process Methods 0.000 claims description 12
- 230000014759 maintenance of location Effects 0.000 claims description 10
- 238000011176 pooling Methods 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 3
- 230000008901 benefit Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 238000010924 continuous production Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000009966 trimming Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30116—Casting
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Mechanical Engineering (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The application provides a casting process control method and equipment based on artificial intelligence technology, and the method is used for acquiring a plurality of first casting images with first defects. And inputting the plurality of first casting images serving as samples into a denoising model, and training the denoising model. And receiving the surface image of the casting, and inputting the surface image of the casting into a denoising model. And determining a demarcating area of the denoising model for the casting surface image. And determining a second defect in the non-demarcated area of the casting surface image through a pre-trained second defect identification model. And determining first matching degrees respectively corresponding to the second defects and the plurality of casting machining devices according to the defect types of the second defects and the position information of the second defects in the castings. And generating a process control instruction under the condition that at least one first matching degree meets a preset condition, and sending the process control instruction and the information of each casting processing device corresponding to each first matching degree to a corresponding process control terminal.
Description
Technical Field
The application relates to the technical field of casting processes, in particular to a casting process control method and equipment based on an artificial intelligence technology.
Background
The casting industry plays an important role in people's life, and various cast products, such as wheel hubs, pots and the like, are produced by casting. At present, the technology is continuously developed, people have higher and higher requirements on required appliances, and people are in transition from pursuit of product use to pursuit of high-quality products. The traditional casting process is mainly carried out in a manual mode, and therefore manpower is wasted excessively. In the existing large-scale casting industry, a robot is used for operation, continuous production of casting processing equipment is realized, and labor is saved.
However, once the casting processing equipment generates product surface defects, such as surface cracks affecting the use experience of consumers, the subsequent casting processing equipment may also have product problems which have occurred historically. If the problems cannot be found in time in the casting process and the technological process causing the surface defects of the product is adjusted in time, the casting production efficiency and the casting production quality are seriously influenced.
Based on this, a technical scheme that the casting processing equipment can be adjusted and controlled in time according to the casting problems generated in history and the casting benefit and the casting quality are ensured is urgently needed.
Disclosure of Invention
The embodiment of the application provides a casting process control method and equipment based on an artificial intelligence technology, which are used for adjusting and controlling casting processing equipment in time, ensuring the casting benefit and the casting quality and improving the use experience of consumers.
In one aspect, an embodiment of the present application provides a casting process control method based on an artificial intelligence technology, where the method includes:
a number of first casting images are acquired of the first defect. And inputting the plurality of first casting images serving as samples into a denoising model so as to train the denoising model. And receiving the surface image of the casting, and inputting the surface image of the casting into a denoising model. And determining a demarcating area of the denoising model for the casting surface image. The demarcated area includes a first defect. And determining a second defect in the non-demarcated area of the casting surface image through a pre-trained second defect identification model. And determining first matching degrees respectively corresponding to the second defects and the plurality of casting machining devices according to the defect types of the second defects and the position information of the second defects in the castings. And correspondingly storing the defect type, the position information of the second defect in the casting and the casting processing equipment information in a defect process comparison table. And generating a process control instruction under the condition that at least one first matching degree meets a preset condition, and sending the process control instruction and each casting machining device corresponding to each first matching degree to a corresponding process control terminal.
In one implementation of the present application, a number of second casting images are input into a second defect identification model to determine whether a second defect exists in each of the second casting images. And removing the second casting images with the second defects from the second casting images to obtain a plurality of first sample images. And screening the first sample images, and sending the screened preset number of first sample images to the user terminal so as to determine the noise retention result of each first sample image based on the operation of the user at the user terminal. Wherein the noise retention result is a result of whether the first defect is included in the first sample image. And under the condition that each noise retention result is matched with the preset result, each first sample image is used as a first casting image.
In one implementation of the present application, a casting type corresponding to each first casting image is determined. It is determined whether the casting types are consistent. And under the condition that the types of the castings are consistent, determining a denoising model corresponding to the type of the casting. And under the condition that the casting types are not consistent, classifying the casting types according to the casting types. And determining each denoising model matched with each classified casting type.
In one implementation of the present application, a casting type corresponding to the casting surface image is determined by an image recognition model. Or sending a casting type acquisition instruction to a sending terminal of the casting surface image so as to determine the casting type of the casting surface image. And determining a denoising model corresponding to the casting type of the casting surface image according to the casting type. Determining a demarcated area of the denoising model for the casting surface image, which specifically comprises: and determining the edge characteristic point of the first defect in the casting surface image through the denoising model. And connecting the edge characteristic points according to a preset mode to obtain an edge line of the first defect. And determining a defect area positioned in the edge line according to the denoising model and the edge line, and taking the defect area as a demarcated area.
In one implementation of the present application, a defect region of the first defect is determined according to the denoising model, and a defect region area is determined. The area of the closure zone of the edge line is determined. And taking the area of the closed region corresponding to the maximum value of the area coverage of the area of the defect region as a demarcating region. And uploading the surface image of the casting to a preset database, and determining casting parameters of the casting corresponding to the surface image of the casting stored in the preset database. Wherein the casting parameters include: casting size, casting density. The ratio of the area of the defect region inside the edge line to the surface area of the casting is determined based on the casting size. And determining whether the ratio of the area of the defect region inside the edge line to the surface area of the casting is larger than a preset threshold value. And under the condition that the area of the defect region in the edge line is determined, and the ratio of the area of the defect region to the surface area of the casting is greater than a preset threshold value, sending the image corresponding to the defect region to a production terminal.
In one implementation of the present application, a plurality of process defect images are obtained from a preset defect image database. The process defect image includes a second defect generated by the casting processing equipment. And inputting each process defect image into the convolution layer of the second defect identification model through the model input layer so as to perform convolution processing on each process defect image to obtain a characteristic image of each process defect image. And inputting each characteristic image into a pooling layer for pooling. And outputting the model identification image, the defect type and the position information of the second defect in the casting through an output layer of the second defect identification model. Wherein the model identifies an image including the second defect in the image. And uploading the identification images of the models, the corresponding defect types and the position information of the second defects in the casting to an image comparison database to determine whether the second defects in the images of the models, the corresponding defect types and the position information of the second defects in the casting are matched or not, and accumulating the matching times. And calculating a second matching degree according to the accumulated matching times and the corresponding matching result. And under the condition that the second matching degree is smaller than a second preset threshold value, acquiring a plurality of process defect images, retraining the second defect identification model until the second matching degree is larger than or equal to the second preset threshold value, and finishing the training of the second defect identification model.
In one implementation of the present application, a defect process comparison table of a previous time period in a preset database is determined. The defect process comparison table is generated according to the history records of the casting processing equipment, the defect types and the position information of the second defects in the castings. And comparing the position information of the defect type and the second defect in the casting with the first sequence group in the defect process comparison table to determine whether the position information of the defect type and the second defect in the casting has the defect process comparison table. The first sequence group comprises a plurality of defect types and defect position information. And if so, determining a matching sequence in a second sequence group corresponding to the defect type and the position information of the second defect in the casting, and determining a plurality of first matching degrees according to the matching sequence. The second sequence group comprises casting machining equipment and corresponding matching degree, wherein the casting machining equipment is matched with the defect types in the first sequence group and the position information of the second defects in the casting. In an implementation manner of the present application, at least one first matching degree corresponding to a maximum value of the plurality of first matching degrees is determined, and whether the at least one first matching degree is greater than a preset value is determined. And under the condition that the at least one first matching degree is larger than a preset value, determining that the ratio of the at least one first matching degree meets a preset condition, and generating a process control instruction. And the process control instruction is used for adjusting equipment operation of casting processing equipment corresponding to the process control terminal.
In an implementation manner of the present application, under the condition that there is no first matching degree satisfying a preset condition in each first matching degree, a plurality of process defect images are obtained, and the second defect identification model is retrained. And sending the area image corresponding to the second defect to each process control terminal of each casting processing device.
On the other hand, the embodiment of the application also provides a casting process control device based on the artificial intelligence technology, and the device comprises:
at least one processor; and a memory communicatively coupled to the at least one processor. Wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to:
a number of first casting images are acquired of the first defect. And inputting the plurality of first casting images serving as samples into a denoising model so as to train the denoising model. And receiving the surface image of the casting, and inputting the surface image of the casting into a denoising model. And determining a demarcating area of the denoising model for the casting surface image. The demarcated area includes a first defect. And determining a second defect in the non-demarcated area of the casting surface image through a pre-trained second defect identification model. And according to the defect type of the second defect and the position information of the second defect in the casting, the second defect and the plurality of casting processing devices respectively correspond to first matching degrees. And correspondingly storing the defect type, the position information of the second defect in the casting and the casting processing equipment information in a defect process comparison table. And generating a process control instruction under the condition that at least one first matching degree meets a preset condition, and sending the process control instruction and the information of each casting processing device corresponding to each first matching degree to a corresponding process control terminal.
By the scheme, the casting problems caused by the casting processing equipment can be accurately identified, the casting processing equipment which generates the casting problems can be timely adjusted and controlled according to the casting problems generated historically, and the casting benefit and the casting quality are guaranteed. Simultaneously, this application can avoid foundry goods processing equipment to take place the problem after, problem before subsequent production casting product continues produces batched inferior product.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
FIG. 1 is a schematic flow chart of a casting process control method based on artificial intelligence technology in an embodiment of the present application;
FIG. 2 is another schematic flow chart of a casting process control method based on artificial intelligence technology in the embodiment of the application;
fig. 3 is a schematic structural diagram of a casting process control device based on artificial intelligence technology in an embodiment of the application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The embodiment of the application provides a casting process control method and equipment based on an artificial intelligence technology, which are used for adjusting and controlling casting processing equipment in time according to casting problems generated historically, and ensuring casting benefit and casting quality.
Various embodiments of the present application are described in detail below with reference to the accompanying drawings.
The embodiment of the application provides a casting process control method based on an artificial intelligence technology, and as shown in fig. 1, the method may include steps S101 to S107:
s101, the server acquires a plurality of first casting images with first defects.
In an embodiment of the present application, before the server obtains a plurality of first casting images with a first defect, the method includes:
firstly, the server inputs a plurality of second casting images into the second defect identification model to determine whether second defects exist in the second casting images.
The second defect identification model is a convolution neural network model, is formed by training a plurality of process defect images and is used for identifying a second defect.
The second defect is a defect generated by casting processing equipment, the second defect and the first defect are obtained by comparing the defect positions and the defect types of a plurality of casting images, and for the comparison difference, the server can compare the first defect and the second defect through a neural network model, and can also obtain the characteristic of the second defect and the characteristic of the first defect through a manual characteristic marking mode.
Then, the server removes all second casting images with second defects from the second casting images to obtain a plurality of first sample images.
When the second defect identification model identifies that the second casting image has a second defect, the server can remove the second casting image from the second casting images, and the remaining second casting images from which the second casting images with the second curve are removed are used as the first sample images.
Then, the server screens the first sample images, and sends the screened preset number of first sample images to the user terminal so as to determine the noise retention result of each first sample image based on the operation of the user at the user terminal.
Wherein the noise retention result is a result of whether the first defect is included in the first sample image.
The number of the first sample images may be too many, and the server may select a preset number of first sample images from the first sample images in a screening manner, where the preset number is set in an actual use process, and this is not specifically limited in this application. The user may select each first sample image displayed on the display interface of the user terminal as a result including the first defect or a result not including the first defect by clicking the first sample image.
And finally, under the condition that the noise retention results are matched with the preset results, the server takes the first sample images as first casting images.
The preset result may be a result including the first defect, and if the noise retention results of the screened first sample images are all the results including the first defect, the server takes each first sample image as the first casting image.
According to the scheme, the first sample image only containing the first defect can be obtained, so that the existence of the second defect is avoided when the denoising model is trained, the training result of the denoising model is influenced, and the accuracy of recognizing the first defect by the denoising model is guaranteed.
It should be noted that the server is only exemplary and the execution subject of the casting process control method based on the artificial intelligence technology is not limited to the server, and the present application is not limited to this.
S102, the server inputs the first casting images into a denoising model as samples so as to train the denoising model.
In this embodiment of the present application, the denoising model can be normally used only after being trained, and therefore, the server inputs a plurality of first casting images into the denoising model as a sample, so as to train the denoising model, further including:
first, the server determines the casting type corresponding to each first casting image.
The casting types at least comprise a pump casting, a valve casting, a hydraulic casting, a pneumatic element casting, a metallurgical machinery casting and a mining machinery casting … …, and the server obtains the casting types through image recognition or a mode determined by a sending terminal.
The server then determines whether the casting types are consistent.
The sending terminal may send a plurality of first casting images, and casting types corresponding to the first casting images may be inconsistent, so that the server needs to determine the consistency of the casting types in order to ensure the accuracy of defect identification and the accuracy of regulation and control of casting processing equipment.
And then, under the condition that all casting types are consistent, the server determines a denoising model matched with the names of the casting types.
And under the condition that the casting types are inconsistent, the server classifies the casting types according to the names of the casting types.
And finally, the server determines the denoising models matched with the names of the classified casting types.
By the scheme, the denoising models corresponding to the casting types can be determined, and the first defects of the casting surface images can be identified more accurately in a classification mode.
S103, the server receives the casting surface image and inputs the casting surface image into a denoising model.
In this embodiment of the present application, before the server inputs the casting surface image into the denoising model, as shown in fig. 2, the method further includes the following steps:
s201, the server determines the casting type corresponding to the casting surface image through the image recognition model. Or the server sends a casting type acquisition instruction to a sending terminal of the casting surface image so as to determine the casting type of the casting surface image.
The server can obtain the casting type of the casting surface image through an image recognition model or a mode of obtaining the casting type from the sending terminal. The casting surface image can be an image acquired by image acquisition equipment in the casting machining equipment process, or an image acquired by the image acquisition equipment after the casting is produced. Among them, the image recognition models include but are not limited to: AlexNet, VGG19, ResNet _152, inclusion v4, DenseNet. The image capturing device may be a mobile phone, a camera, and the like, which is not particularly limited in this application.
S202, the server determines a denoising model corresponding to the casting type of the casting surface image according to the casting type.
After obtaining the casting type, the server selects a denoising model matched with the name of the casting type from the plurality of denoising models according to the name of the casting type. For example, the name of the casting type is: and (4) a hub, wherein the name of the denoising model is the hub denoising model.
S203, the server determines the edge characteristic point of the first defect in the casting surface image through the denoising model.
After the server inputs the casting surface image into the denoising model, the denoising model determines a region of a first defect in the casting surface image, and labels edge feature points of the region of the first defect, wherein the region may be a regular graph, but the edge feature points may label the edge points of the first defect in the regular image.
And S204, connecting the edge characteristic points by the server according to a preset mode to obtain an edge line of the first defect.
The server may sequentially connect the edge feature points to each adjacent edge feature point in a clockwise or counterclockwise manner, where the connected edge feature points form an edge line, and the edge line may form a regular pattern or an irregular pattern.
S205, the server determines a defect area located inside the edge line as a demarcated area according to the denoising model and the edge line.
And the server determines a defect region of the first defect according to the denoising model, and determines the area of the defect region. The closed area of the edge line is determined. And taking the area of the closed region corresponding to the maximum value of the area coverage rate of the area of the defect region as a demarcated region.
The server obtains a region of the first defect and an obtained edge line of the first defect according to the denoising model, and takes the region with the largest area coverage ratio of the area of the closed region of the edge line and the area coverage of the region of the first defect as a demarcated region.
S206, the server uploads the casting surface image to a preset database, and casting parameters of the corresponding casting of the casting surface image stored in the preset database are determined.
Wherein the casting parameters include: casting size, casting density.
The server can send the casting surface images from the image acquisition equipment to a preset database, and the preset database summarizes and stores casting parameters of all castings corresponding to the casting surface images, such as casting parameters of A vehicle hubs.
In addition, the preset database can actively identify the surface image of the casting to obtain the name of the casting. Or the server identifies the casting surface image to obtain the name of the casting corresponding to the casting surface image, and sends the casting surface image and the casting name to a preset database, the preset database compares the casting surface image with the image in the database, compares the casting name with the name in the preset database, and determines the casting parameters of the matched casting after comparison.
And S207, the server determines the ratio of the area of the defect region inside the edge line to the surface area of the casting according to the size of the casting.
The server can calculate the area of the defect area inside the edge line of the casting corresponding to the casting surface image through the casting size, and calculate the ratio of the area to the surface area of the casting. For example, the area of the defect region inside the edge line of the casting is S1, the surface area of the casting is S2, and the ratio is S1/S2.
S208, the server determines whether the ratio of the area of the defect region inside the edge line to the surface area of the casting is larger than a preset threshold value.
In the embodiment of the present application, the preset threshold may be set according to an actual use process, for example, 0.5 and 0.4, and specific data of the preset threshold is not limited.
S209, the server sends the image corresponding to the defect area to the production terminal under the condition that the ratio of the area of the defect area in the edge line to the surface area of the casting is larger than a preset threshold value.
In the embodiment of the application, if the area of the first defect on the surface of the casting is too large, although the first defect is not caused by the casting processing equipment, the defect affects the appearance or the use experience, and therefore, in the case that the area of the first defect is too large, the image corresponding to the defect area is collected and sent to the production terminal, so that the production terminal selects to perform a production operation on the casting corresponding to the first defect, for example, re-casting, or otherwise trimming the first defect.
Through the scheme, the area corresponding to the first defect can be accurately identified, so that the production terminal can find the problem of the existing first defect, the casting is trimmed or recast in time, the qualification of the cast product is ensured, the casting production quality is improved, and meanwhile, under the condition that the first defect is identified between the casting processing equipment, the next casting processing equipment is not influenced by the existence of the first defect, and the normal production of the casting processing equipment is ensured.
And S104, the server determines a defined region of the denoising model for the casting surface image.
Wherein the demarcated areas include a first defect.
The embodiment of determining the demarcated area is, for example, the step S205, which is not described herein again.
And S105, determining a second defect in the non-demarcated area of the casting surface image by the server through a pre-trained second defect identification model.
In this embodiment, the server, before determining the second defect in the non-demarcated area of the casting surface image through the pre-trained second defect identification model, further includes:
firstly, a server acquires a plurality of process defect images from a preset defect image database.
Wherein the process defect image comprises a second defect generated by the casting processing equipment.
The server can be connected with a preset defect image database, and the first defect identification model is trained through the process defect image in the defect image database. The plurality of process defect images correspond to different defect information, the defect information at least comprising: the defect type of the defect position a is convex, and the defect type of the defect position b is a crack.
Secondly, the server inputs each process defect image into the convolution layer of the second defect identification model through the model input layer so as to carry out convolution processing on each process defect image and obtain the characteristic image of each process defect image.
Wherein the feature image includes features of the second defect.
And thirdly, inputting each feature image into the pooling layer by the server for pooling.
Then, the server outputs the model identification image, the defect type and the position information of the second defect in the casting through the output layer of the second defect identification model.
Wherein the model identifies an image including the second defect in the image.
And then, the server uploads the identification images, the defect types and the position information of the second defects in the casting to an image comparison database so as to determine whether the second defects in the model images are matched with the corresponding defect types and the position information of the second defects in the casting or not, and accumulates the matching times.
The server may match the defect type and defect location with a second defect of the model image in the image comparison database in a manner that determines whether the defect type and defect location are consistent. And accumulates the total number of matches.
And then, the server calculates a second matching degree according to the accumulated matching times and the corresponding matching result.
The second matching degree is calculated by determining the cumulative matching times X, and if the matching times is T, the second matching degree Y is T/X.
And finally, the server acquires a plurality of process defect images under the condition that the second matching degree is smaller than a second preset threshold, retrains the second defect identification model until the second matching degree is larger than or equal to the second preset threshold, and finishes the training of the second defect identification model.
When the second matching degree is smaller than a second preset threshold, the server determines that the accuracy of the second defect identification model is not enough, and in the actual use process, the setting of the second preset threshold affects the accuracy of the second defect identification model, so that the second preset threshold can be set according to actual needs.
And S106, determining first matching degrees respectively corresponding to the second defects and the plurality of casting machining devices by the server according to the defect types of the second defects and the position information of the second defects in the castings.
And correspondingly storing the defect type, the position information of the second defect in the casting and the casting processing equipment information in a defect process comparison table.
In this embodiment of the present application, the determining, by the server, a plurality of first matching degrees according to the defect information of the second defect specifically includes:
firstly, the server determines a defect process comparison table of the last time period in a preset database.
The defect process comparison table is generated according to the history records of the casting processing equipment, the defect types and the position information of the second defects in the castings.
The defect process comparison table is generated before the current casting surface image is sent, the defect process comparison table has historical data recorded after defect information occurs in casting processing equipment, and the defect information comprises defect types and position information of a second defect in the casting. For example, the casting processing device N1 corresponds to the casting information m1, the matching degree is N1, and the casting processing device N2 corresponds to the casting information m1, the matching degree is N2. The defect process control list records the history that a second defect identical with the surface image of the casting occurs in the past in the casting processing equipment of the casting corresponding to the surface image of the casting of the enterprise or manufacturer.
And then, the server compares the defect type in the defect information and the position information of the second defect in the casting with the first sequence group in the defect process comparison table to determine whether the defect type and the position information of the second defect in the casting have the defect process comparison table.
The first sequence group comprises a plurality of defect types and defect position information.
And then, under the condition that the defect process comparison table exists in the position information of the defect type and the second defect in the casting, the server determines the matching sequence in the second sequence group corresponding to the position information of the defect type and the second defect in the casting, so as to determine a plurality of first matching degrees according to the matching sequence.
The second sequence group comprises casting machining equipment and corresponding matching degree, wherein the casting machining equipment is matched with the defect types in the first sequence group and the position information of the second defects in the casting.
And the second sequence group records the matching degree of each casting processing device and the defect information corresponding to each casting processing device.
And under the condition that the defect information does not have a defect process comparison table, the server sends the image corresponding to the defect information to the management terminal, and the expert of the management terminal identifies the matching degree of the defect information corresponding to each casting processing device.
Through the scheme, the casting processing equipment for the casting problems can be generated in time according to the casting problems generated in history, so that the casting processing equipment can be adjusted and controlled in time.
S107, the server generates a process control instruction under the condition that at least one first matching degree meets a preset condition, and sends the process control instruction and each casting processing device corresponding to each first matching degree to a corresponding process control terminal.
Wherein the preset condition is used for determining at least one first matching degree ratio.
The process control terminals may correspond to casting processing equipment, such as one process control terminal for each casting processing equipment, or one process control terminal for multiple casting processing equipment. The process control terminal may be a computer, a mobile phone, a server, or other devices, which is not specifically limited in this application.
In this embodiment of the application, the generating, by the server, the process control instruction when the at least one first matching degree meets the preset condition specifically includes:
the server determines at least one first matching degree corresponding to the maximum value in the first matching degrees and determines whether the at least one first matching degree is larger than a preset value.
And the server determines that the ratio of the at least one first matching degree meets a preset condition and generates a process control instruction under the condition that the at least one first matching degree is larger than a preset value.
And the process control instruction is used for adjusting equipment operation of casting processing equipment corresponding to the process control terminal. For example: stop running, continue running, etc.
In another embodiment of the application, the server obtains a plurality of process defect images and retrains the second defect identification model when the first matching degrees which meet the preset conditions do not exist in the first matching degrees. And sending the area image corresponding to the second defect to each process control terminal of each casting processing device.
Through the scheme, the casting problem caused by the casting processing equipment can be accurately identified, the casting problem generated according to history is adjusted and controlled in time to the casting processing equipment generating the casting problem, the casting benefit and the casting quality are guaranteed, and the quantity of produced qualified products can be further guaranteed. The casting quality and the qualified quantity are improved, and meanwhile, a better purchasing and using experience can be provided for consumers.
Fig. 3 is a schematic structural diagram of a casting process control device based on artificial intelligence technology according to an embodiment of the present application, where the device includes:
at least one processor; and a memory communicatively coupled to the at least one processor. Wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to:
a plurality of first casting images are acquired in which a first defect is present. And inputting the plurality of first casting images serving as samples into a denoising model so as to train the denoising model. And receiving the surface image of the casting, and inputting the surface image of the casting into a denoising model. And determining a demarcating area of the denoising model for the casting surface image. The demarcated areas include a first defect. And determining a second defect in the non-demarcated area of the casting surface image through a pre-trained second defect identification model. And determining first matching degrees respectively corresponding to the second defects and the plurality of casting machining devices according to the defect types of the second defects and the position information of the second defects in the castings. And correspondingly storing the defect type, the position information of the second defect in the casting and the casting processing equipment information in a defect process comparison table. And generating a process control instruction under the condition that at least one first matching degree meets a preset condition, and sending the process control instruction and information of each casting processing device corresponding to each first matching degree to a corresponding process control terminal.
The embodiments in the present application are described in a progressive manner, and the same and similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the apparatus embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The devices and the methods provided by the embodiment of the application are in one-to-one correspondence, so the devices also have beneficial technical effects similar to the corresponding methods.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art to which the present application pertains. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A casting process control method based on artificial intelligence technology is characterized by comprising the following steps:
acquiring a plurality of first casting images with first defects;
inputting a plurality of first casting images serving as samples into a denoising model so as to train the denoising model;
receiving a casting surface image, and inputting the casting surface image into the denoising model;
determining a demarcating area of the denoising model for the casting surface image; the demarcated area includes the first defect;
determining a second defect in a non-demarcated area of the casting surface image through a pre-trained second defect identification model;
determining first matching degrees respectively corresponding to the second defects and a plurality of casting processing devices according to the defect types of the second defects and the position information of the second defects in the casting; the defect type, the position information of the second defect in the casting and the casting processing equipment information are correspondingly stored in a defect process comparison table;
and generating a process control instruction under the condition that at least one first matching degree meets a preset condition, and sending the process control instruction and the information of each casting processing device corresponding to each first matching degree to a corresponding process control terminal.
2. The method of claim 1, wherein prior to obtaining the plurality of first casting images having the first defect, the method further comprises:
inputting a plurality of second casting images into the second defect identification model to determine whether the second defects exist in each second casting image;
removing the second casting images with the second defects from the second casting images to obtain a plurality of first sample images;
screening each first sample image, and sending the screened preset number of first sample images to a user terminal so as to determine a noise retention result of each first sample image based on the operation of a user at the user terminal; wherein the noise retention result is a result of whether the first defect is included in the first sample image;
and under the condition that each noise retention result is matched with a preset result, taking each first sample image as the first casting image.
3. The method of claim 1, wherein prior to inputting the plurality of first casting images as a sample into a denoising model for training the denoising model, the method further comprises:
determining the casting type corresponding to each first casting image;
determining whether the types of the castings are consistent;
determining the denoising model corresponding to the casting type under the condition that the casting types are consistent;
under the condition that the casting types are inconsistent, classifying the casting types according to the casting types;
and determining each denoising model matched with each classified casting type.
4. The method of claim 1, wherein prior to inputting the casting surface image into the de-noising model, the method further comprises:
determining the casting type corresponding to the casting surface image through an image recognition model; or alternatively
Sending a casting type acquisition instruction to a sending terminal of the casting surface image so as to determine the casting type of the casting surface image;
determining the denoising model corresponding to the casting type of the casting surface image according to the casting type;
determining a demarcated area of the denoising model for the casting surface image, specifically comprising:
determining edge characteristic points of the first defects in the casting surface image through the denoising model;
connecting the edge feature points according to a preset mode to obtain an edge line of the first defect;
and determining a defect region positioned in the edge line as the demarcated region according to the denoising model and the edge line.
5. The method according to claim 4, wherein determining a defect region located inside the edge line according to the denoising model and the edge line, and specifically includes:
determining a defect region of the first defect according to the denoising model, and determining the area of the defect region;
determining a closure zone area of a closure zone of the edge line;
taking the area of the closed area corresponding to the maximum area coverage rate of the area of the defect area as a demarcated area;
determining a defect region located inside the edge line according to the denoising model and the edge line, and after the region is defined, the method further comprises:
uploading the casting surface image to a preset database, and determining casting parameters of the casting corresponding to the casting surface image stored in the preset database; wherein the casting parameters include: casting size and casting density;
determining the ratio of the area of the defect region inside the edge line to the surface area of the casting according to the size of the casting;
determining whether the ratio of the area of the defect region inside the edge line to the surface area of the casting is greater than a preset threshold;
and if so, sending the image corresponding to the defect area to a production terminal.
6. The method of claim 1, wherein prior to determining a second defect in a non-demarcated region of the casting surface image via a pre-trained second defect identification model, the method further comprises:
acquiring a plurality of process defect images from a preset defect image database; the process defect image comprises the second defects generated by the casting processing equipment;
inputting each process defect image into a convolution layer of the second defect identification model through a model input layer so as to carry out convolution processing on each process defect image and obtain a characteristic image of each process defect image;
inputting each characteristic image into a pooling layer for pooling treatment;
outputting a model identification image, the defect type and the position information of the second defect in the casting through an output layer of the second defect identification model; wherein the model identifies an image in the image that includes the second defect;
uploading the model identification images, the corresponding defect types and the position information of the second defects in the casting to an image comparison database to determine whether the second defects in the model images are matched with the corresponding defect types and the position information of the second defects in the casting or not, and accumulating the matching times;
calculating a second matching degree according to the accumulated matching times and corresponding matching results;
and under the condition that the second matching degree is smaller than a second preset threshold value, acquiring a plurality of process defect images, retraining the second defect identification model until the second matching degree is larger than or equal to the second preset threshold value, and finishing the training of the second defect identification model.
7. The method according to claim 1, wherein determining first matching degrees respectively corresponding to the second defect of the casting processing equipment and a plurality of casting processing equipment according to the defect type of the second defect and the position information of the second defect in the casting, specifically comprises:
determining a defect process comparison table of the last time period in a preset database; the defect process comparison table is generated according to the historical records of the casting processing equipment, the defect types and the position information of the second defects in the castings;
comparing the position information of the defect type and the second defect in the casting with a first sequence group in the defect process comparison table to determine whether the position information of the defect type and the second defect in the casting has the defect process comparison table; wherein the first sequence group comprises a plurality of defect types and defect position information;
if the defect type exists, determining a matching sequence in a second sequence group corresponding to the position information of the defect type and the second defect in the casting, and determining a plurality of first matching degrees according to the matching sequence; the second sequence group comprises the casting machining equipment and the corresponding matching degree, wherein the casting machining equipment matches the defect types in the first sequence group and the position information of the second defects in the casting.
8. The method according to claim 7, wherein generating the process control command when the at least one first matching degree satisfies a preset condition specifically comprises:
determining at least one first matching degree corresponding to the maximum value in the first matching degrees, and determining whether the at least one first matching degree is greater than a preset value;
under the condition that the at least one first matching degree is larger than the preset value, determining that the proportion of the at least one first matching degree meets the preset condition, and generating the process control instruction; the process control instruction is used for adjusting equipment operation of the casting processing equipment corresponding to the process control terminal.
9. The method of claim 1, further comprising:
under the condition that the first matching degrees meeting the preset condition do not exist in the first matching degrees, acquiring a plurality of process defect images, and retraining the second defect identification model;
and sending the area image corresponding to the second defect to each process control terminal of each casting processing device.
10. A casting process control apparatus based on artificial intelligence technology, the apparatus comprising:
at least one processor; and (c) a second step of,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to:
acquiring a plurality of first casting images with first defects;
inputting a plurality of first casting images serving as samples into a denoising model so as to train the denoising model;
receiving a casting surface image, and inputting the casting surface image into the denoising model;
determining a demarcating area of the denoising model for the casting surface image; the demarcated area includes the first defect;
determining a second defect in a non-demarcated area of the casting surface image through a pre-trained second defect identification model;
determining first matching degrees respectively corresponding to the second defects and a plurality of casting processing devices according to the defect types of the second defects and the position information of the second defects in the casting; the defect type, the position information of the second defect in the casting and the casting processing equipment information are correspondingly stored in a defect process comparison table;
and generating a process control instruction under the condition that at least one first matching degree meets a preset condition, and sending the process control instruction and the information of each casting processing device corresponding to each first matching degree to a corresponding process control terminal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210111350.XA CN114519797A (en) | 2022-01-29 | 2022-01-29 | Casting process control method and equipment based on artificial intelligence technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210111350.XA CN114519797A (en) | 2022-01-29 | 2022-01-29 | Casting process control method and equipment based on artificial intelligence technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114519797A true CN114519797A (en) | 2022-05-20 |
Family
ID=81597699
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210111350.XA Pending CN114519797A (en) | 2022-01-29 | 2022-01-29 | Casting process control method and equipment based on artificial intelligence technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114519797A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114939643A (en) * | 2022-05-24 | 2022-08-26 | 昆山莱捷有色金属有限公司 | Die-casting control method and device |
-
2022
- 2022-01-29 CN CN202210111350.XA patent/CN114519797A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114939643A (en) * | 2022-05-24 | 2022-08-26 | 昆山莱捷有色金属有限公司 | Die-casting control method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111369016B (en) | Image recognition-based on-site operation and maintenance method and system | |
| CN110555819B (en) | Equipment monitoring method, device and equipment based on infrared and visible light image fusion | |
| CN111179249A (en) | Power equipment detection method and device based on deep convolutional neural network | |
| US20200250584A1 (en) | Modeling method for smart prognostics and health management system and computer program product thereof | |
| CN109146847B (en) | Wafer map batch analysis method based on semi-supervised learning | |
| CN111949795A (en) | Work order automatic classification method and device | |
| CN114519797A (en) | Casting process control method and equipment based on artificial intelligence technology | |
| CN116187718B (en) | Intelligent goods identification and sorting method and system based on computer vision | |
| CN109409780B (en) | Change processing method, device, computer equipment and storage medium | |
| CN116740728B (en) | Dynamic acquisition method and system for wafer code reader | |
| CN116050678A (en) | Die-casting product processing test system and method based on cloud computing | |
| CN115880699A (en) | Food packaging bag detection method and system | |
| CN111242924A (en) | Product quality management system | |
| CN117095247B (en) | Numerical control machining-based machining gesture operation optimization method, system and medium | |
| CN114595872A (en) | Method for predicting key operation path of automatic wharf container ship | |
| CN117252851B (en) | Standard quality detection management platform based on image detection and identification | |
| CN111325280A (en) | Label generation method and system | |
| CN116363114A (en) | Ceramic tile surface quality detection method and device, electronic equipment and storage medium | |
| KR102242011B1 (en) | Apparatus for classifying livestock-products image grade based on deep-learning | |
| CN115526859A (en) | Method for identifying production defects, distributed processing platform, equipment and storage medium | |
| CN112581472B (en) | Target surface defect detection method facing human-computer interaction | |
| CN109300025B (en) | Big data and deep learning-based auditing method and robot system | |
| CN115210744A (en) | Systems, methods, and media for manufacturing processes | |
| CN114723748B (en) | Detection method, device and equipment of motor controller and storage medium | |
| CN118095971B (en) | AD calcium milk beverage processing technology assessment method, system and 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 |