CN110310260B - Material distribution decision method, equipment and storage medium based on machine learning model - Google Patents
Material distribution decision method, equipment and storage medium based on machine learning model Download PDFInfo
- Publication number
- CN110310260B CN110310260B CN201910530692.3A CN201910530692A CN110310260B CN 110310260 B CN110310260 B CN 110310260B CN 201910530692 A CN201910530692 A CN 201910530692A CN 110310260 B CN110310260 B CN 110310260B
- Authority
- CN
- China
- Prior art keywords
- distributed
- defect
- defect type
- machine learning
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- 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
- G06T2207/30164—Workpiece; Machine component
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention provides a material distribution decision method, material distribution decision equipment and a storage medium based on a machine learning model, and relates to the field of cloud computing. The method comprises the following steps: shooting a component to be distributed at multiple angles to obtain multiple images of the component to be distributed; determining at least one defect type of the material to be distributed and confidence degrees corresponding to the at least one defect type according to the plurality of images; and taking the at least one defect type of the part to be distributed and the confidence degree corresponding to the at least one defect type as input parameters of the machine learning model to determine a distribution decision result corresponding to the part to be distributed. The embodiment of the invention improves the detection and distribution efficiency and can comprehensively detect the product parts.
Description
Technical Field
The invention relates to the technical field of quality detection, in particular to a material distribution decision method, material distribution decision equipment and a storage medium based on a machine learning model.
Background
In a traditional industrial manufacturing production scenario, for example, 3C product parts are manufactured, so called "3C product", which is a combination of Computer (Computer), Communication (Communication) and Consumer Electronics (Consumer Electronics), and is also called "information appliance". Surface condition detection is often required for multiple angles of the product. The surface state detection of the product is an important link for controlling the shipment quality of manufacturers.
In the related technology, quality inspection is generally performed through manual visual inspection, namely each production line trains a certain number of quality inspectors, and produced parts are subjected to full-angle defect detection and classification through naked eyes, so that the detection efficiency is low; and the quality inspector often divides the material after detecting a fault of a certain face of the part, and can not detect the part comprehensively.
Disclosure of Invention
The invention provides a material distribution decision-making method, material distribution decision-making equipment and a storage medium based on a machine learning model, which are used for improving the detection efficiency and comprehensively detecting product parts.
In a first aspect, the present invention provides a material distribution decision method based on a machine learning model, including:
shooting a component to be distributed at multiple angles to obtain multiple images of the component to be distributed;
determining at least one defect type of the material to be distributed and confidence degrees corresponding to the at least one defect type according to the plurality of images;
and taking the at least one defect type of the part to be distributed and the confidence degree corresponding to the at least one defect type as input parameters of the machine learning model to determine a distribution decision result corresponding to the part to be distributed.
In a possible implementation manner, the determining, according to the plurality of images, at least one defect type of the component to be separated and a confidence corresponding to each of the at least one defect type includes:
acquiring a plurality of pre-established defect classification models;
and determining at least one image for each defect classification model, and taking the at least one image as an input parameter of the defect classification model to obtain a defect type corresponding to the material to be distributed and a confidence coefficient corresponding to the defect type.
In one possible implementation, the establishing a plurality of defect classification models includes:
acquiring training data corresponding to each defect classification model; the training data comprises a plurality of images acquired under different image acquisition conditions and defect types corresponding to the images;
and training each defect classification model according to the training data corresponding to each defect classification model.
In a possible implementation manner, before the determining, by using the at least one defect type existing in the component to be separated and the confidence corresponding to the at least one defect type as input parameters of the machine learning model, a separation decision result corresponding to the component to be separated, further includes:
acquiring a plurality of training data corresponding to the machine learning model; the training data includes: the defect type output by each defect classification model, the confidence corresponding to the defect type and a material distribution decision result;
training the machine learning model with the plurality of training data.
In one possible implementation manner, the method further includes:
and distributing the material to be distributed to a material box corresponding to the material distribution decision result according to the material distribution decision result.
In one possible implementation manner, the distributing the parts to be distributed to the material boxes corresponding to the distribution decision result according to the distribution decision result includes:
generating a control instruction according to the material distribution decision result;
and controlling the mechanical arm to distribute the material to be distributed to the material box corresponding to the material distribution decision result according to the control instruction.
In one possible implementation, the machine learning model is any one of the following models: adaboost model, neural network model, support vector machine model and decision tree model.
In a second aspect, the present invention provides a material distribution decision device based on a machine learning model, including:
the image acquisition module is used for shooting a material component to be distributed at multiple angles to obtain a plurality of images of the material component to be distributed;
the defect classification module is used for determining at least one defect type of the component to be distributed and confidence degrees corresponding to the at least one defect type according to the plurality of images;
and the processing module is used for taking at least one defect type of the component to be distributed and the confidence corresponding to the at least one defect type as input parameters of the machine learning model so as to determine a distribution decision result corresponding to the component to be distributed.
In a possible implementation manner, the defect classification module is specifically configured to:
establishing a plurality of defect classification models;
and determining at least one image for each defect classification model, and taking the at least one image as an input parameter of the defect classification model to obtain a defect type corresponding to the material to be distributed and a confidence coefficient corresponding to the defect type.
In one possible implementation manner, the method further includes: the model training module is used for:
acquiring training data corresponding to each defect classification model; the training data comprises a plurality of images acquired under different image acquisition conditions and defect types corresponding to the images;
and training each defect classification model according to the training data corresponding to each defect classification model.
In one possible implementation, the model training module is further configured to:
acquiring a plurality of training data corresponding to the machine learning model; the training data includes: the defect type output by each defect classification model, the confidence corresponding to the defect type and the final material distribution decision result;
training the machine learning model with the plurality of training data.
In one possible implementation manner, the method further includes: the material distributing module is used for:
and distributing the material to be distributed to a material box corresponding to the material distribution decision result according to the material distribution decision result.
In one possible implementation, the material distribution module is specifically configured to:
generating a control instruction according to the material distribution decision result;
and controlling the mechanical arm to distribute the material to be distributed to the material box corresponding to the material distribution decision result according to the control instruction.
In one possible implementation, the machine learning model is any one of the following models: adaboost model, neural network model, support vector machine model and decision tree model.
In a third aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the computer program implements the method described in any one of the first aspect.
In a fourth aspect, an embodiment of the present invention provides an electronic device, including:
a processor; and
a memory for storing executable instructions of the processor;
wherein the processor is configured to perform the method of any of the first aspects via execution of the executable instructions.
According to the material distribution decision method, the equipment and the storage medium based on the machine learning model, which are provided by the embodiment of the invention, the component to be distributed is shot in multiple angles to obtain multiple images of the component to be distributed; determining at least one defect type of the material to be distributed and confidence degrees corresponding to the at least one defect type according to the plurality of images; the method comprises the steps of taking at least one defect type of a part to be distributed and confidence degrees corresponding to the at least one defect type as input parameters of a machine learning model to determine a distribution decision result corresponding to the part to be distributed, so that the detection efficiency is improved, and the machine learning model is utilized according to the at least one defect type of the part to be distributed and the confidence degrees corresponding to the at least one defect type, so that the defect detection can be carried out on the part to be distributed more comprehensively, the distribution can be carried out according to the final decision result, and the distribution accuracy is higher.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and together with the description, serve to explain the principles of the disclosure.
FIG. 1 is a schematic flow chart diagram of an embodiment of a material distribution decision method based on a machine learning model according to the present invention;
FIG. 2 is a schematic diagram of an embodiment of a method provided by the present invention;
FIG. 3 is a schematic diagram illustrating an implementation process of an embodiment of the method provided by the present invention;
FIG. 4 is a schematic structural diagram of a material distribution decision-making device based on a machine learning model according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an embodiment of an electronic device provided in the present invention.
With the foregoing drawings in mind, certain embodiments of the disclosure have been shown and described in more detail below. These drawings and written description are not intended to limit the scope of the disclosed concepts in any way, but rather to illustrate the concepts of the disclosure to those skilled in the art by reference to specific embodiments.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.
The terms "comprising" and "having," and any variations thereof, in the description and claims of this invention and the drawings described herein are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.
Firstly, the application scene related to the invention is introduced:
the material distribution method based on the machine learning model provided by the embodiment of the invention is applied to the image data based on the surface of the product part, the quality of the product part is detected, and the product part is distributed in a material distribution scene, namely, the product part is distributed according to the finally obtained defect type of the quality detection.
The product component targeted in the embodiment of the invention is, for example, a 3C product component, such as a component of a mobile phone, a tablet computer, and a wearable smart device, for example, a charging port component of a mobile phone.
The technical solution of the present invention will be described in detail below with specific examples. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments.
Fig. 1 is a schematic flow chart of a material distribution decision method based on a machine learning model according to an embodiment of the present invention. As shown in fig. 1, the method provided by this embodiment includes:
Specifically, each surface of the material to be distributed component is shot through an image acquisition device, such as a camera, and a plurality of images are acquired. For example, multiple images are acquired at different camera angles, under lighting conditions.
Training data acquired under different image acquisition conditions in advance according to a deep learning algorithm to obtain a plurality of defect classification models for predicting different image data; the training data includes the defect types obtained by manual quality inspection. The image acquisition conditions include, for example, at least one of: shooting angle, illumination condition.
Specifically, the plurality of images may be input into a defect classification model established in advance, so as to obtain the confidence levels corresponding to at least one defect type and various defect types of the component to be separated, and the higher the confidence level is, the higher the possibility that the component to be separated has the defect of the type is.
103, taking at least one defect type of the component to be distributed and the confidence corresponding to the at least one defect type as input parameters of the machine learning model to determine a distribution decision result corresponding to the component to be distributed.
Specifically, at least one defect type and a confidence degree corresponding to the at least one defect type are input into a pre-established machine learning model, and a material distribution decision result output by the machine learning model includes the defect type of the component to be distributed, and further includes a final confidence degree corresponding to the defect type, or includes a confidence degree corresponding to the component to be distributed as a qualified product and the qualified product.
Further, after obtaining the material distribution decision result, the method further includes:
and distributing the material to be distributed to a material box corresponding to the material distribution decision result according to the material distribution decision result.
The material distribution can be realized in the following way:
generating a control instruction according to the material distribution decision result;
and controlling the mechanical arm to distribute the material to be distributed to the material box corresponding to the material distribution decision result according to the control instruction.
In the method of the embodiment, a component to be distributed is shot in multiple angles to obtain multiple images of the component to be distributed; determining at least one defect type of the material to be distributed and confidence degrees corresponding to the at least one defect type according to the plurality of images; the method comprises the steps of taking at least one defect type of a part to be distributed and confidence degrees corresponding to the at least one defect type as input parameters of a machine learning model to determine a distribution decision result corresponding to the part to be distributed, so that the detection efficiency is improved, and the machine learning model is utilized according to the at least one defect type of the part to be distributed and the confidence degrees corresponding to the at least one defect type, so that the defect detection can be carried out on the part to be distributed more comprehensively, the distribution can be carried out according to the final decision result, and the distribution accuracy is higher.
On the basis of the foregoing embodiment, further, step 102 may specifically be implemented by:
acquiring a plurality of pre-established defect classification models;
and determining at least one image for each defect classification model, and taking the at least one image as an input parameter of the defect classification model to obtain a defect type corresponding to the material to be distributed and a confidence coefficient corresponding to the defect type.
Further, establishing a plurality of defect classification models can be specifically realized by the following method:
acquiring training data corresponding to each defect classification model; the training data comprises a plurality of images acquired under different image acquisition conditions and defect types corresponding to the images;
and training each defect classification model according to the training data corresponding to each defect classification model.
Specifically, a plurality of defect classification models can be established in advance by using a deep learning algorithm according to training data. The training data comprises training data collected under different shooting angles, lighting conditions and the like.
And respectively determining input parameters of each defect classification model, namely determining at least one image from the plurality of images as the input parameters of each defect classification model, and obtaining a defect classification result output by each model, namely the defect type of the part to be distributed and the confidence corresponding to each defect type.
Different defect classification models may correspond to different image acquisition conditions, such as shooting angle, lighting conditions.
And determining input parameters of different defect classification models according to image acquisition conditions, namely determining the defect classification model to be input by each image according to the image acquisition conditions of the images.
On the basis of the above embodiment, further, before step 103, a model may be established and learned, which may specifically be implemented as follows:
acquiring a plurality of training data corresponding to the machine learning model; the training data includes: the defect type output by each defect classification model, the confidence corresponding to the defect type and the final material distribution decision result;
training the machine learning model with the plurality of training data.
Further, the machine learning model is any one of the following models: adaboost model, neural network model, support vector machine model and decision tree model.
Inputting images of a plurality of parts into the defect classification models to obtain the defect types of different parts output by each defect classification model and the confidence degrees corresponding to the defect types, determining the final material distribution decision result corresponding to each part according to manual experience, and training the machine learning model by taking the data as training data.
As shown in fig. 2, the trained machine learning model takes the defect type corresponding to the component to be separated by 3 and the confidence level corresponding to each defect type as input, so as to obtain a final separation decision result, that is, the defect type with the highest probability exists in the component to be separated, or the component to be separated is a qualified product (further, the confidence level corresponding to the qualified product is included), and a separation decision is performed according to the separation decision result, that is, the component to be separated is placed in a material box corresponding to the separation decision result; the input to the machine learning model may be a vector of confidence levels for each defect type.
According to the scheme, the defect types of the parts to be distributed are predicted through the defect classification models, the predicted defect types and the confidence degrees corresponding to the defect types are obtained, the final distribution decision result is judged through the machine learning model according to the defect types and the confidence degrees corresponding to the defect types, distribution is carried out according to the distribution decision result, and the accuracy is high.
The material distribution decision method can be executed through one or more electronic devices, a processor of the electronic device controls an image acquisition device to acquire images of a material to be distributed, the image acquisition device can be a part of the electronic device or is arranged separately from the electronic device, the image acquisition device generates a prediction request for the acquired images and information of the material to be distributed and sends the prediction request to the processor, and the processor outputs defect types corresponding to the material to be distributed and confidence degrees corresponding to the defect types according to the prediction request through a defect classification model. The processor finally uses the defect type corresponding to the material to be distributed and the confidence degree corresponding to each defect type as the input of the machine learning model, and operates the machine learning model to obtain an output result, i.e. a material distribution decision result, such as the final defect type corresponding to the material to be distributed (further including the final confidence degree corresponding to the defect type), or the material to be distributed is a qualified product. And finally, distributing the material to be distributed according to a material distribution decision result, namely putting the material into a corresponding material box, such as a material box for a fault product or a material box for a qualified product, wherein a plurality of material boxes for the fault product can be arranged according to different defect types. The electronic device can also output or store the intermediate result, for example, the defect type corresponding to the material to be distributed and the confidence degree corresponding to each defect type are output or stored, so that the subsequent analysis and use are facilitated. The electronic device can also send out prompt information such as alarm and the like, for example, the electronic device can directly alarm after the confidence degree corresponding to a certain corresponding defect type is greater than a certain threshold value.
Furthermore, the electronic device can also monitor the use condition of hardware resources, an operating system, a CPU, heterogeneous computing chips and a memory of each computer of the whole product line.
In summary, according to the method provided by the embodiment of the invention, for the prediction results of a plurality of defect classification models which are independent from each other, a machine learning algorithm such as Adaboost is used, so that the decision accuracy of classification based on the prediction results of the defect classification models is greatly improved, the over-killing of the models with small probability can be effectively inhibited, meanwhile, the method can adapt to the production line detection standard in real time through the adjustment of manual supervision data, the high yield is ensured, the manufacturing cost of a factory is greatly reduced, and the method provided by the embodiment of the invention is strong in robustness and can be used for sustainable learning optimization.
Fig. 4 is a structural diagram of an embodiment of a material distribution decision-making device based on a machine learning model, as shown in fig. 4, the material distribution decision-making device based on the machine learning model of the embodiment includes:
the image acquisition module 401 is configured to perform multi-angle shooting on a component to be distributed to obtain multiple images of the component to be distributed;
a defect classification module 402, configured to determine, according to the multiple images, at least one defect type of the component to be separated and a confidence level corresponding to each of the at least one defect type;
the processing module 403 is configured to use at least one defect type of the component to be separated and a confidence corresponding to the at least one defect type as input parameters of the machine learning model to determine a separation decision result corresponding to the component to be separated.
In a possible implementation manner, the defect classification module 402 is specifically configured to:
establishing a plurality of defect classification models;
and determining at least one image for each defect classification model, and taking the at least one image as an input parameter of the defect classification model to obtain a defect type corresponding to the material to be distributed and a confidence coefficient corresponding to the defect type.
In one possible implementation manner, the method further includes: a model training module to:
acquiring training data corresponding to each defect classification model; the training data comprises a plurality of images acquired under different image acquisition conditions and defect types corresponding to the images;
and training each defect classification model according to the training data corresponding to each defect classification model.
In one possible implementation, the model training module is further configured to:
acquiring a plurality of training data corresponding to the machine learning model; the training data includes: the defect type output by each defect classification model, the confidence corresponding to the defect type and the final material distribution decision result;
training the machine learning model with the plurality of training data.
In one possible implementation manner, the method further includes: a material distribution module 404 for:
and distributing the material to be distributed to a material box corresponding to the material distribution decision result according to the material distribution decision result.
In one possible implementation, the material distributing module 404 is specifically configured to:
generating a control instruction according to the material distribution decision result;
and controlling the mechanical arm to distribute the material to be distributed to the material box corresponding to the material distribution decision result according to the control instruction.
In one possible implementation, the machine learning model is any one of the following models: adaboost model, neural network model, support vector machine model and decision tree model.
As shown in fig. 3, the image acquisition module generates a prediction request from the acquired multiple images and the information of the material to be distributed, and sends the prediction request to the processor, and the processing module outputs a defect classification result corresponding to the material to be distributed, that is, the defect types of the material to be distributed and the confidence degrees corresponding to the defect types, according to the prediction request through the defect classification model. The processing module takes the defect type corresponding to the component to be distributed and the confidence degree corresponding to each defect type as the input of the machine learning model, and operates the machine learning model to obtain an output result, i.e. a distribution decision result, such as the final defect type corresponding to the component to be distributed (further including the final confidence degree corresponding to the defect type), or the component to be distributed is a qualified product. The material distribution module distributes the material to the parts to be distributed according to the material distribution decision result, namely the parts are placed into corresponding material boxes, for example, a material box of a fault product or a material box of a qualified product is placed, and a plurality of material boxes of the fault product can be arranged according to different defect types.
The apparatus of this embodiment may be configured to implement the technical solutions of the above method embodiments, and the implementation principles and technical effects are similar, which are not described herein again.
Fig. 5 is a structural diagram of an embodiment of an electronic device provided in the present invention, and as shown in fig. 5, the electronic device includes:
a processor 501, and a memory 302 for storing executable instructions for the processor 501.
Optionally, the method may further include: and the image acquisition part 503 is used for acquiring images of the material to be distributed.
The above components may communicate over one or more buses.
The processor 501 is configured to execute the corresponding method in the foregoing method embodiment by executing the executable instruction, and the specific implementation process of the method may refer to the foregoing method embodiment, which is not described herein again.
The embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the method in the foregoing method embodiment is implemented.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (9)
1. A material distribution decision method based on a machine learning model is characterized by comprising the following steps:
shooting a component to be distributed at multiple angles to obtain multiple images of the component to be distributed;
acquiring a plurality of pre-established defect classification models;
determining at least one image for each defect classification model, and taking the at least one image as an input parameter of the defect classification model to obtain a defect type corresponding to the material to be distributed and a confidence coefficient corresponding to the defect type output by each defect classification model;
and taking the at least one defect type of the part to be distributed and the confidence degree corresponding to the at least one defect type as input parameters of the machine learning model to determine a distribution decision result corresponding to the part to be distributed.
2. The method of claim 1, wherein the establishing a plurality of defect classification models comprises:
acquiring training data corresponding to each defect classification model; the training data comprises a plurality of images acquired under different image acquisition conditions and defect types corresponding to the images;
and training each defect classification model according to the training data corresponding to each defect classification model.
3. The method according to claim 1 or 2, wherein before the step of using the at least one defect type of the parts to be distributed and the confidence corresponding to the at least one defect type as input parameters of the machine learning model to determine the distribution decision result corresponding to the parts to be distributed, the method further comprises:
acquiring a plurality of training data corresponding to the machine learning model; the training data includes: the defect type output by each defect classification model, the confidence corresponding to the defect type and a material distribution decision result;
training the machine learning model with the plurality of training data.
4. The method of claim 1 or 2, further comprising:
and distributing the material to be distributed to a material box corresponding to the material distribution decision result according to the material distribution decision result.
5. The method according to claim 4, wherein the distributing the parts to be distributed to the material boxes corresponding to the distribution decision results according to the distribution decision results comprises the following steps:
generating a control instruction according to the material distribution decision result;
and controlling the mechanical arm to distribute the material to be distributed to the material box corresponding to the material distribution decision result according to the control instruction.
6. The method of claim 1 or 2, wherein the machine learning model is any one of: adaboost model, neural network model, support vector machine model and decision tree model.
7. A material distribution decision-making device based on a machine learning model is characterized by comprising:
the image acquisition module is used for shooting a material component to be distributed at multiple angles to obtain a plurality of images of the material component to be distributed;
the defect classification module is used for determining at least one defect type of the component to be distributed and confidence degrees corresponding to the at least one defect type according to the plurality of images;
the processing module is used for taking at least one defect type of the component to be distributed and the confidence corresponding to the at least one defect type as input parameters of the machine learning model so as to determine a distribution decision result corresponding to the component to be distributed;
the defect classification module is specifically configured to:
acquiring a plurality of pre-established defect classification models;
and determining at least one image for each defect classification model, and taking the at least one image as an input parameter of the defect classification model to obtain the defect type corresponding to the material to be distributed and the confidence corresponding to the defect type output by each defect classification model.
8. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method of any one of claims 1-6.
9. An electronic device, comprising:
a processor; and
a memory for storing executable instructions of the processor;
wherein the processor is configured to perform the method of any of claims 1-6 via execution of the executable instructions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910530692.3A CN110310260B (en) | 2019-06-19 | 2019-06-19 | Material distribution decision method, equipment and storage medium based on machine learning model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910530692.3A CN110310260B (en) | 2019-06-19 | 2019-06-19 | Material distribution decision method, equipment and storage medium based on machine learning model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110310260A CN110310260A (en) | 2019-10-08 |
| CN110310260B true CN110310260B (en) | 2021-10-15 |
Family
ID=68076950
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910530692.3A Active CN110310260B (en) | 2019-06-19 | 2019-06-19 | Material distribution decision method, equipment and storage medium based on machine learning model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110310260B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113469033A (en) * | 2021-06-30 | 2021-10-01 | 北京集创北方科技股份有限公司 | Image recognition method and device, electronic equipment and storage medium |
| CN113781485B (en) * | 2021-11-12 | 2022-09-09 | 成都数联云算科技有限公司 | Intelligent detection method and device for PCB defect types, electronic equipment and medium |
| EP4361853A1 (en) | 2022-09-06 | 2024-05-01 | Contemporary Amperex Technology Co., Limited | Detection method, detection apparatus and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107167716A (en) * | 2017-07-11 | 2017-09-15 | 国网福建省电力有限公司泉州供电公司 | A kind of shelf depreciation default kind identification method and device |
| CN107657603A (en) * | 2017-08-21 | 2018-02-02 | 北京精密机电控制设备研究所 | A kind of industrial appearance detecting method based on intelligent vision |
| CN108672316A (en) * | 2018-03-27 | 2018-10-19 | 哈尔滨理工大学 | A kind of micro parts quality detecting system based on convolutional neural networks |
| CN109344905A (en) * | 2018-10-22 | 2019-02-15 | 王子蕴 | A kind of transmission facility automatic fault recognition methods based on integrated study |
| CN109408389A (en) * | 2018-10-30 | 2019-03-01 | 北京理工大学 | A kind of aacode defect detection method and device based on deep learning |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109118482B (en) * | 2018-08-07 | 2019-12-31 | 腾讯科技（深圳）有限公司 | Panel defect analysis method and device and storage medium |
-
2019
- 2019-06-19 CN CN201910530692.3A patent/CN110310260B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107167716A (en) * | 2017-07-11 | 2017-09-15 | 国网福建省电力有限公司泉州供电公司 | A kind of shelf depreciation default kind identification method and device |
| CN107657603A (en) * | 2017-08-21 | 2018-02-02 | 北京精密机电控制设备研究所 | A kind of industrial appearance detecting method based on intelligent vision |
| CN108672316A (en) * | 2018-03-27 | 2018-10-19 | 哈尔滨理工大学 | A kind of micro parts quality detecting system based on convolutional neural networks |
| CN109344905A (en) * | 2018-10-22 | 2019-02-15 | 王子蕴 | A kind of transmission facility automatic fault recognition methods based on integrated study |
| CN109408389A (en) * | 2018-10-30 | 2019-03-01 | 北京理工大学 | A kind of aacode defect detection method and device based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 基于多分类器集成的工业品缺陷分析方法研究;符瑞盛;《中国优秀硕士学位论文全文数据库信息科技辑》;20180115;参见摘要、第三章、第四章、第六章 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110310260A (en) | 2019-10-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110310260B (en) | Material distribution decision method, equipment and storage medium based on machine learning model | |
| KR102229594B1 (en) | Display screen quality detection method, device, electronic device and storage medium | |
| WO2018121737A1 (en) | Keypoint prediction, network training, and image processing methods, device, and electronic device | |
| JP6306683B2 (en) | Computer-based method of dynamic category object recognition Computer-based method and system | |
| CN111814850A (en) | Defect detection model training method, defect detection method and related device | |
| CN109685097B (en) | Image detection method and device based on GAN | |
| CN108427927A (en) | Target recognition methods and device, electronic equipment, program and storage medium again | |
| CN105336074A (en) | Alarm method and device | |
| JP2018506046A (en) | Method for detecting defects on the tire surface | |
| US20180109776A1 (en) | Method, system and medium for improving the quality of 2d-to-3d automatic image conversion using machine learning techniques | |
| US20170280130A1 (en) | 2d video analysis for 3d modeling | |
| MX2021012006A (en) | Quantum feature kernel alignment. | |
| CN111080633A (en) | Screen defect detection method and device, terminal equipment and storage medium | |
| CN110781976A (en) | Extension method of training image, training method and related device | |
| CN117152258B (en) | Product positioning method and system for intelligent workshop of pipeline production | |
| US20200082297A1 (en) | Inspection apparatus and machine learning method | |
| CN110427998A (en) | Model training, object detection method and device, electronic equipment, storage medium | |
| CN114529484A (en) | Deep learning sample enhancement method for direct current component change in imaging | |
| CN113139932B (en) | Deep learning defect image identification method and system based on ensemble learning | |
| CN116485796B (en) | Pest detection method, pest detection device, electronic equipment and storage medium | |
| JP7075057B2 (en) | Image judgment device, image judgment method and image judgment program | |
| CN107526438B (en) | Method, device and storage device for tracking recording and broadcasting according to hand raising action | |
| CN117408931A (en) | Image defect detecting system, its producing method and computer readable recording medium | |
| CN113870210A (en) | Image quality evaluation method, device, equipment and storage medium | |
| CN107562050B (en) | Method and system for robot to recognize environment |
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 |