A method, system, device and readable storage medium for automatic control of a frame loader

Abstract

The present invention relates to the technical field of frame framing machines, and provides a frame framing machine automatic control method, system, device and readable storage medium, including the following steps: extracting characteristic values ​​of a profile to be framed, and judging the type of the profile to be framed according to the characteristic values; obtaining a key detection area and a general detection area according to the type of the profile to be framed, and performing image acquisition and analysis on the images of the profile to be framed in the key detection area and the general detection area according to different detection strategies to judge whether there are preset defects; if there are preset defects, the profile to be framed is marked as a defective product, otherwise, the profile to be framed is marked as a good product and a framing operation is performed. The present invention performs image acquisition and analysis on the images of the profile to be framed in the key detection area and the general detection area through different detection strategies, provides a new idea for defect detection of the profile to be framed, can perform defect detection on different detection areas of the profile to be framed in a targeted manner, and improves the degree of intelligence.

Description

Technical Field

[0001] The present invention relates to the technical field of frame loader, and in particular to an automatic control method, system, device and readable storage medium of a frame loader. Background Art

[0002] The framing equipment (including automatic framing equipment and semi-automatic framing equipment) consists of a material frame, a pad frame, pads, a profile conveyor, etc. After the products that have been cut to size are conveyed out through the conveying equipment, pads must be placed under the products before they are placed in the material frame to avoid scratching or damaging the products.

[0003] Chinese invention patents with announcement numbers CN110040522B, CN117226486B and CN112520378B all disclose an automatic aluminum profile framing device, which can automatically frame aluminum profiles to improve work efficiency. However, such automatic aluminum profile framing devices do not perform defect detection on the aluminum profiles to be framed before framing them. The aluminum profiles may have defects after framing, and the aluminum profile framing effect needs to be improved. Summary of the invention

[0004] In order to solve the problem that the existing automatic framing equipment for aluminum profiles does not perform defect detection on the aluminum profiles to be framed before framing, and the aluminum profiles after framing may have defects, the present invention provides an automatic control method, system, equipment and readable storage medium for a framing machine.

[0005] The automatic control method of the framing machine comprises the following steps:

[0006] Extracting characteristic values ​​of the profile to be framed, and determining the type of the profile to be framed according to the characteristic values;

[0007] Acquire key inspection areas and general inspection areas according to the type of the profile to be framed, and perform image acquisition and analysis on the images of the profile to be framed in the key inspection areas and the general inspection areas according to different inspection strategies to determine whether there are preset defects;

[0008] If the preset defects exist, the profile to be framed is marked as a defective product; otherwise, the profile to be framed is marked as a good product and the framing operation is performed.

[0009] The automatic control method for the framing machine obtains the key detection area and the general detection area according to the type of the profile to be framed, and performs image acquisition and analysis on the images of the profile to be framed in the key detection area and the general detection area according to different detection strategies, determines whether there are preset defects, and performs a framing operation on the profile to be framed when the profile to be framed is calibrated as a good product, which can reduce the problem of defects in the profile after framing. In addition, by obtaining the key detection area and the general detection area, and performing image acquisition and analysis on the images of the profile to be framed in the key detection area and the general detection area according to different detection strategies, a new idea is also provided for defect detection of the profile to be framed, which can perform defect detection on different detection areas of the profile to be framed in a targeted manner, thereby improving the degree of intelligence.

[0010] Furthermore, the specific method for obtaining the key detection area and the general detection area according to the type of the profile to be framed includes the following steps:

[0011] Obtain defect detection areas of different batches of profiles;

[0012] Obtaining the correlation between different batches of profiles and the profile to be framed;

[0013] Calculate the weight index of different defect detection areas according to the detection weights of the defect detection areas of different batches of profiles and the correlation degree;

[0014] The key detection area and the general detection area are obtained according to the weight index.

[0015] Furthermore, a specific method for obtaining the correlation between different batches of profiles and the profile to be framed includes the following steps:

[0016] Acquire multi-dimensional feature information that causes defects in different batches of profiles and the profiles to be framed;

[0017] Calculate the correlation between different batches of profiles and the profile to be framed according to the multi-dimensional feature information ;

[0018] in , Indicates the number of feature information dimensions between the batch profile and the profile to be framed, Indicates the number of feature information of a certain dimension and , Indicates the first dimension of a batch profile feature information, Indicates the first dimension of the frame profile to be installed feature information, Indicates The correlation coefficient between the batch profiles corresponding to the dimension feature information and the profiles to be framed, Represents the proportional adjustment coefficient.

[0019] Furthermore, the specific formula for calculating the weight index of different detection areas according to the detection weight of the defect detection area and the correlation degree is: ;

[0020] in, Indicates The weight index of the defect detection area, Indicates The detection weight of each defect detection area.

[0021] Furthermore, the specific method for performing image acquisition and analysis on the images of the profiles to be framed in the key detection area and the general detection area according to different detection strategies is as follows:

[0022] Preset basic testing resources;

[0023] For key inspection areas, additional corresponding inspection resources are added based on the basic inspection resources according to the weight index, and image acquisition and analysis is performed on the images of the profiles to be framed in the key inspection areas;

[0024] For the image of the profile to be framed in the general detection area, basic detection resources are used to perform image acquisition and analysis.

[0025] Furthermore, the multi-dimensional feature information includes feature information of profile raw material quality, mold quality, processing equipment accuracy, environmental parameters and process parameters.

[0026] Furthermore, the characteristic information of the mold quality includes the mold surface finish grade.

[0027] The present invention also provides an automatic control system for a framing machine, which comprises:

[0028] A feature extraction module is used to extract feature values ​​of the profile to be framed, and determine the type of the profile to be framed according to the feature values;

[0029] An image analysis module is used to obtain a key detection area and a general detection area according to the type of the profile to be framed, perform image acquisition and analysis on the images of the profile to be framed in the key detection area and the general detection area according to different detection strategies, and feed back the results of the image acquisition and analysis to the judgment execution module;

[0030] The judgment execution module is used to judge whether the profile to be framed has a preset defect based on the result of image acquisition and analysis. If the preset defect exists, the profile to be framed is marked as a defective product. Otherwise, the profile to be framed is marked as a good product and a framing operation is performed.

[0031] The present invention also provides an automatic control device for a frame assembling machine, comprising:

[0032] Controller;

[0033] A memory storing executable instructions;

[0034] The executable instructions can be run on the controller and implement the automatic control method of the frame assembling machine.

[0035] The present invention also provides a computer-readable storage medium storing a computer program, wherein the computer program implements the automatic control method of the frame assembling machine when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The present invention can be further understood from the following description in conjunction with the accompanying drawings. The components in the figures are not necessarily drawn to scale, but the emphasis is placed on illustrating the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.

[0037] Figure 1 It is a schematic diagram of the overall flow of an automatic control method for a framing machine in one embodiment of the present invention.

[0038] Figure 2 It is a flowchart of a specific method for obtaining a key detection area and a general detection area in another embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with its embodiments. It should be understood that the specific implementation methods described herein are only used to explain the present invention and do not limit the protection scope of the present invention.

[0040] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only and do not represent the only implementation method.

[0041] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0042] The “first” and “second” mentioned in the present invention do not represent specific quantities and orders, but are merely used to distinguish names.

[0043] like Figure 1 As shown, an automatic control method of the frame assembling machine in one embodiment of the present invention includes the following steps:

[0044] S1, extracting characteristic values ​​of the profile to be framed, and judging the type of the profile to be framed according to the characteristic values. The characteristic values ​​include but are not limited to the size, image profile, strength parameter, wall thickness and aperture of the profile to be framed. Different types of profiles have different attribute parameters and characteristic values. By extracting the characteristic values ​​of the profile to be framed, the type of the profile to be framed can be judged according to the difference in characteristic values.

[0045] S2, obtaining a key inspection area and a general inspection area according to the type of the profile to be framed, and performing image acquisition and analysis on the images of the profile to be framed in the key inspection area and the general inspection area according to different inspection strategies to determine whether there are preset defects.

[0046] For a certain type of profile, due to different process parameters, processing equipment accuracy, mold surface smoothness and raw material quality, the finished product will have more or less defects, such as bulges and distortions, sand holes and burrs, scratches, corner bottoms, orange peels, leaks, jets, paint bubbles, pits, variegated colors, dirty spots, cracks, etc. In many cases, for the same batch of profiles, the locations of different types of defects are often different. For example, uneven defects often appear at the edges or corners of the profile, and burrs are more likely to appear at the edges of the profile.

[0047] Specifically, the quality of raw materials is the key to determining the quality of finished aluminum profiles. If the raw materials have defects such as inclusions, pores, cracks, or improper operation during the smelting and casting process, it may lead to uneven internal structure of the profile and defects in the finished product; if the processing equipment for profiles, such as extruders, cutting machines, punching machines, etc., has unstable performance, low precision, or improper maintenance, it may lead to errors in the profile processing process, such as dimensional deviation, irregular shape, etc.

[0048] As for the process parameters, their setting is crucial to the processing quality of aluminum profiles. If the process parameters such as extrusion temperature, extrusion speed, cooling speed, etc. are not properly controlled, the internal structure of the aluminum profile may be loose, the grains may be coarse, and cracks and pockmarks may appear on the surface. For example, if the temperature of the extrusion barrel is too high, the surface of the metal material will be burned, an oxide layer will appear on the surface, and defects will appear on the surface of the profile, affecting the service life of the profile. If the temperature of the extrusion barrel is too low, the metal material will lack fluidity, scratches will appear on the surface of the profile, and the hardness of the profile surface will be insufficient, affecting the quality of the profile.

[0049] During the processing of profiles, environmental factors such as temperature, humidity, and cleanliness may affect the processing quality. For example, too high a temperature may cause thermal deformation of the aluminum profile during processing; too high a humidity may cause rust on the surface of the aluminum profile; insufficient environmental cleanliness may cause the surface of the aluminum profile to be contaminated, affecting its appearance and performance.

[0050] According to different detection strategies, image acquisition and analysis are performed on the images of the profile to be framed in the key detection area and the general detection area, and the specific method for judging whether there are preset defects includes: obtaining multiple preset defect images of profiles from different batches and different defect types, and performing similarity calculation and analysis on the images of the profile to be framed in the key detection area and the general detection area and the preset defect images; if the similarity between the image of the profile to be framed and the preset defect image is greater than a preset similarity threshold, it is judged that the profile to be framed has defects; otherwise, it is judged that the profile to be framed does not have defects.

[0051] Of course, different preset defect images can also be collected first to train the neural network model, and the trained neural network model can be used to identify the collected images of the profile to be framed in the key inspection area and the general inspection area to determine whether the profile to be framed has defects.

[0052] Since the visual inspection method for profile surface defects based on image processing technology is a conventional technical means in this field, it will not be described in detail here.

[0053] In the existing profile defect detection methods based on visual images, generally, the profile is subjected to all-round image acquisition, detection and analysis regardless of the area, and the weight of the detection area is not differentiated according to the defect type, that is, the image acquisition and analysis of the profile to be framed in the key detection area and the general detection area is not performed according to different detection strategies. In this way, since there is no targeted defect detection for different detection areas of the profile to be framed, there is a problem of uneven distribution of detection resources or waste of detection resources, which undoubtedly reduces the intelligence level of profile defect detection and analysis and the utilization efficiency of detection resources.

[0054] The automatic control method for the framing machine obtains key inspection areas and general inspection areas, and performs image acquisition and analysis on the images of the profiles to be framed in the key inspection areas and the general inspection areas according to different inspection strategies to determine whether there are preset defects. This can, on the one hand, reduce the problem of defects in profiles after framing, and on the other hand, can also allocate defect detection resources in a targeted manner, thereby improving the utilization efficiency of detection resources and the intelligence level of profile defect detection and analysis, and avoiding the problem of uneven distribution of detection resources or waste of detection resources.

[0055] S3, if there is a preset defect, the profile to be framed is marked as a defective product, otherwise, the profile to be framed is marked as a good product and a framing operation is performed.

[0056] Since the framing operation for the profiles to be framed that are calibrated as good products is a conventional technical means in the field, it will not be described in detail here.

[0057] In summary, the automatic control method of the framing machine obtains the key inspection area and the general inspection area according to the type of the profile to be framed, and performs image acquisition and analysis on the images of the profile to be framed in the key inspection area and the general inspection area according to different detection strategies, determines whether there are preset defects, and performs framing operations on the profile to be framed when it is calibrated as a good product, which can reduce the problem of defects in the profile after framing. In addition, by obtaining the key inspection area and the general inspection area, and performing image acquisition and analysis on the images of the profile to be framed in the key inspection area and the general inspection area according to different detection strategies, it also provides a new idea for defect detection of the profile to be framed, and can perform defect detection on different inspection areas of the profile to be framed in a targeted manner, thereby improving the degree of intelligence.

[0058] In one embodiment, in step S2, as Figure 2 As shown, the specific method for obtaining the key detection area and the general detection area according to the type of the profile to be framed includes the following steps:

[0059] S21, obtaining defect detection areas of different batches of profiles. Different batches of profiles here refer to profiles of the same type produced at different times. Preferably, profiles with process similarities in raw material sources, processing equipment, process flow, parameter settings, etc. or produced within a continuous period of time can be divided into different batches of profiles according to preset standards or conditions.

[0060] The defect detection area can be divided into areas by existing technical personnel for different batches of profiles, different areas are assigned different serial numbers, and the area where the defect is located is detected, and then the area where the defect is located is marked as the defect detection area.

[0061] S22, obtaining the correlation between different batches of profiles and the profile to be framed.

[0062] Preferably, the specific method for obtaining the correlation between different batches of profiles and the profile to be framed comprises the following steps:

[0063] S221, obtaining multi-dimensional feature information that causes defects in different batches of profiles and the profile to be framed. The multi-dimensional feature information includes but is not limited to feature information such as raw material quality (such as strength parameters, etc.), process parameters, mold quality (mold surface smoothness), processing equipment accuracy, and environmental parameters.

[0064] Preferably, the characteristic information of the mold quality includes the mold surface finish grade.

[0065] S222, calculating the correlation between different batches of profiles and the profile to be framed according to the multi-dimensional feature information .

[0066] in , Indicates the number of feature information dimensions between the batch profile and the profile to be framed, Indicates the number of feature information of a certain dimension and , Indicates the first dimension of a batch profile feature information, Indicates the first dimension of the frame profile to be installed feature information, Indicates The correlation coefficient between the batch profiles corresponding to the dimension feature information and the profiles to be framed, Indicates the proportional adjustment coefficient, which can be set by technicians.

[0067] Assuming that in the same batch, the feature information of a certain dimension is relatively stable and almost unchanged, the number of feature information of a certain dimension can be understood as the number of profiles in the same batch.

[0068] S23, calculating weight indexes of different defect detection areas according to the detection weights of the defect detection areas of different batches of profiles and the correlation degree.

[0069] Specifically, the specific formula for calculating the weight index of different detection areas according to the detection weight of the defect detection area and the correlation degree is: ;in, Indicates The weight index of the defect detection area, Indicates The detection weight of each defect detection area.

[0070] S24, obtaining the key detection area and the general detection area according to the weight index. Specifically, a preset weight threshold can be set first, and the defect detection area with a weight index greater than the preset weight threshold is used as the key detection area, and the defect detection area with a weight index not greater than the preset weight threshold is used as the general detection area, and the corresponding key detection area and general detection area are marked on the frame profile.

[0071] The correlation between different batches of profiles and the profiles to be framed is calculated respectively through the multi-dimensional feature information, and the weight index of different defect detection areas is calculated according to the correlation and the detection weights of the defect detection areas of different batches of profiles. The multi-dimensional feature information that may affect the defects of different batches of profiles and the profiles to be framed is associated, and the multi-dimensional feature information can be integrated to calculate a relatively reliable weight index, thereby improving the accuracy of the division of key detection areas and general detection areas.

[0072] In one embodiment, the specific method for performing image acquisition and analysis on the image of the profile to be framed in the key detection area and the general detection area according to different detection strategies is:

[0073] S25, preset basic detection resources.

[0074] The preset basic detection resources can be set by the technicians, which are defined as the minimum detection resources that meet the general profile surface defect detection and analysis requirements. The detection resources include but are not limited to cameras for collecting profile surface images, memory for image analysis, storage space for storing images of common profile surface defects, and neural network models for profile surface defect recognition.

[0075] The preset basic detection resources can be determined by obtaining in advance the average of the minimum detection resources required for multiple general detection areas.

[0076] S26, for the key inspection area, based on the basic inspection resources, additional corresponding inspection resources are added according to the weight index, and image acquisition and analysis are performed on the image of the profile to be framed in the key inspection area.

[0077] The detection resources are not limited to one category. Accordingly, when the basic detection resources include multiple different categories of detection resources such as cameras, memory and storage, each category of basic detection resources corresponds to one additional corresponding detection resource.

[0078] S27, for the image of the profile to be framed in the general detection area, perform image acquisition and analysis using basic detection resources.

[0079] Regarding the addition of corresponding detection resources on the basis of basic detection resources, it can be understood as: assuming that the basic detection resources are a certain type such as cameras, memory resources or neural network models, and the corresponding unit of measurement is N, the addition of corresponding detection resources can be expressed as K×N; wherein K is greater than 0, which means that the multiple of the corresponding detection resources is additionally increased according to the weight index.

[0080] For example, the basic detection resources are N=5 cameras, and when K=1, the corresponding additional detection resources are 1×5=5 cameras. At this time, the detection resources for image acquisition and analysis of the images of the profiles to be framed in the key detection area include the basic detection resources and the corresponding additional detection resources according to the weight index. The total number of detection resources is (1+K)×N=(1+1)×5=10 cameras, that is, the detection resources of 10 cameras are used to perform image acquisition and analysis on the images of the profiles to be framed in the key detection area, and the detection resources of 5 cameras are used to perform image acquisition and analysis on the images of the profiles to be framed in the general detection area. In this way, for the images of the profiles to be framed in the key detection area, the speed and efficiency of image acquisition can be improved by using the detection resources of 10 cameras, and because the angular positions of multiple cameras relative to the key detection area are often different, it is also possible to obtain rich acquired images, which is convenient for subsequent image analysis.

[0081] If the basic detection resource is N=200MB of memory resources, when K=0.3, the corresponding increase in detection resources is 0.3×200=60MB of memory resources. At this time, the detection resources for image acquisition and analysis of the images of the profiles to be framed in the key detection area include the basic detection resources and the corresponding detection resources added according to the weight index. The total number of detection resources is (1+K)×N=(1+0.3)×200=260MB of memory resources, that is, the detection resources of 260MB of memory are used to perform image acquisition and analysis on the images of the profiles to be framed in the key detection area, and the detection resources of 200MB of memory are used to perform image acquisition and analysis on the images of the profiles to be framed in the general detection area. In this way, for the images of the profiles to be framed in the key detection area, the computing power of the image analysis can be increased by the detection resources of 260MB of memory, thereby improving the image analysis efficiency of the images of the profiles to be framed in the key detection area.

[0082] That is to say, allocating more detection resources to conduct targeted surface defect detection and analysis in key monitoring areas can effectively improve the efficiency of image acquisition and analysis.

[0083] Preferably, the total number of a certain type of detection resources in the key detection area .in, Indicates the weight index corresponding to the key detection area, Represents the corresponding resource adjustment coefficient, which is set by the technical staff and , Indicates the amount of a certain type of preset basic detection resources. represents a natural constant. It can be seen that for the key inspection area, the larger the weight index is, the larger the total number of corresponding inspection resources of a certain type is. In this way, more inspection resources can be allocated to the key inspection area for image acquisition and analysis, such as adding K×N cameras on the basis of N cameras and / or adding K×N MB of memory on the basis of N MB of memory to perform image acquisition and analysis on the key inspection area, so as to more effectively and accurately detect the defects on the profile surface.

[0084] For a frame loader, the total number of various detection resources is often unchanged. Allocating different detection resources to key detection areas and general detection areas can effectively allocate the overall detection resources of the frame loader. While meeting the basic detection resource needs of general detection areas for surface defects, more detection resources are allocated to conduct targeted surface defect detection and analysis in key monitoring areas, thereby improving the utilization efficiency of detection resources and the intelligence level of profile defect detection and analysis, and avoiding the problem of uneven distribution of detection resources or waste of detection resources.

[0085] In one embodiment, the present invention further provides an automatic control system for a frame assembling machine, which includes a feature extraction module, an image analysis module and a judgment execution module.

[0086] The feature extraction module is used to extract the characteristic values ​​of the profile to be framed, and determine the type of the profile to be framed based on the characteristic values; the image analysis module is used to obtain the key detection area and the general detection area according to the type of the profile to be framed, and perform image acquisition and analysis on the images of the profile to be framed in the key detection area and the general detection area according to different detection strategies, and feed back the results of the image acquisition and analysis to the judgment execution module.

[0087] The judgment execution module is used to judge whether the profile to be framed has a preset defect according to the result of image acquisition and analysis. If the preset defect exists, the profile to be framed is marked as a defective product. Otherwise, the profile to be framed is marked as a good product and the framing operation is performed. For example, when the result of image acquisition and analysis is the similarity between the image of the profile to be framed and the preset defect image, the judgment execution module compares the similarity between the image of the profile to be framed and the preset defect image with the size relationship of the preset similarity threshold, and judges whether the profile to be framed has the preset defect.

[0088] After the profile to be framed is marked as a defective product, it can be clamped by a robot and transferred to a defective product area.

[0089] Since it is a conventional technical means in the art to perform framing operations on the profiles to be framed by using tools such as robots after detecting that the profiles to be framed are good products, and to transfer the profiles to be framed to the defective product area or warehouse by using tools such as robots and carts after detecting that the profiles to be framed are defective products, they are not described in detail here.

[0090] The automatic control system of the framing machine obtains the key detection area and the general detection area according to the type of the profile to be framed, and performs image acquisition and analysis on the images of the profile to be framed in the key detection area and the general detection area according to different detection strategies, determines whether there are preset defects, and performs framing operations on the profile to be framed when the profile to be framed is calibrated as a good product, which can reduce the problem of defects in the profile after framing. In addition, by obtaining the key detection area and the general detection area, and performing image acquisition and analysis on the images of the profile to be framed in the key detection area and the general detection area according to different detection strategies, it also provides a new idea for defect detection of the profile to be framed, and can perform defect detection on different detection areas of the profile to be framed in a targeted manner, thereby improving the degree of intelligence.

[0091] In one embodiment, the present invention further provides an automatic control device for a frame assembling machine, comprising:

[0092] Controller;

[0093] A memory storing executable instructions;

[0094] The executable instructions can be run on the controller and implement the automatic control method of the frame assembling machine.

[0095] In one of the embodiments, the present invention further provides a computer-readable storage medium storing a computer program, wherein the computer program implements the automatic control method of the framing machine when executed by a processor.

[0096] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0097] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for automatically controlling a framing machine, characterized in that: The automatic control method of the frame assembling machine comprises the following steps: Extracting characteristic values ​​of the profile to be framed, and determining the type of the profile to be framed according to the characteristic values; Acquire key inspection areas and general inspection areas according to the type of the profile to be framed, and perform image acquisition and analysis on the images of the profile to be framed in the key inspection areas and the general inspection areas according to different inspection strategies to determine whether there are preset defects; If there is a preset defect, the profile to be framed is marked as a defective product; otherwise, the profile to be framed is marked as a good product and the framing operation is performed; The specific method for obtaining the key detection area and the general detection area according to the type of the frame profile to be installed includes the following steps: Obtain defect detection areas of different batches of profiles; Obtaining the correlation between different batches of profiles and the profile to be framed; Calculate the weight index of different defect detection areas according to the detection weights of the defect detection areas of different batches of profiles and the correlation degree; Obtaining a key detection area and a general detection area according to the weight index; The specific method for obtaining the correlation between different batches of profiles and the profile to be framed comprises the following steps: Acquire multi-dimensional feature information that causes defects in different batches of profiles and the profiles to be framed; Calculate the correlation between different batches of profiles and the profile to be framed according to the multi-dimensional feature information ; in , Indicates the number of feature information dimensions between the batch profile and the profile to be framed, Indicates the number of feature information of a certain dimension and , Indicates the first dimension of a batch profile feature information, Indicates the first dimension of the frame profile to be installed feature information, Indicates The correlation coefficient between the batch profiles corresponding to the dimension feature information and the profiles to be framed, represents the proportional adjustment coefficient; The specific formula for calculating the weight index of different detection areas according to the detection weight of the defect detection area and the correlation degree is: ; in, Indicates The weight index of the defect detection area, Indicates The detection weight of each defect detection area.

2. The automatic control method of a framing machine as claimed in claim 1, characterized in that: The specific method for performing image acquisition and analysis on the images of the profiles to be framed in the key detection area and the general detection area according to different detection strategies is as follows: Preset basic testing resources; For key inspection areas, additional corresponding inspection resources are added based on the basic inspection resources according to the weight index, and image acquisition and analysis is performed on the images of the profiles to be framed in the key inspection areas; For the image of the profile to be framed in the general detection area, basic detection resources are used to perform image acquisition and analysis.

3. The automatic control method of a frame assembling machine as claimed in claim 2, characterized in that: The multi-dimensional feature information includes feature information of profile raw material quality, mold quality, processing equipment accuracy, environmental parameters and process parameters.

4. The automatic control method of a frame assembling machine as claimed in claim 3, characterized in that: The characteristic information of the mold quality includes the mold surface finish grade.

5. An automatic control system for a framing machine, characterized in that: The automatic control system of the framing machine comprises: A feature extraction module is used to extract feature values ​​of the profile to be framed, and determine the type of the profile to be framed according to the feature values; An image analysis module is used to obtain a key detection area and a general detection area according to the type of the profile to be framed, perform image acquisition and analysis on the images of the profile to be framed in the key detection area and the general detection area according to different detection strategies, and feed back the results of the image acquisition and analysis to the judgment execution module; A judgment execution module, used for judging whether the profile to be framed has a preset defect according to the result of image acquisition and analysis, and if so, marking the profile to be framed as a defective product; otherwise, marking the profile to be framed as a good product and performing a framing operation; The specific method for obtaining the key detection area and the general detection area according to the type of the frame profile to be installed includes the following steps: Obtain defect detection areas of different batches of profiles; Obtaining the correlation between different batches of profiles and the profile to be framed; Calculate the weight index of different defect detection areas according to the detection weights of the defect detection areas of different batches of profiles and the correlation degree; Obtaining a key detection area and a general detection area according to the weight index; The specific method for obtaining the correlation between different batches of profiles and the profile to be framed comprises the following steps: Acquire multi-dimensional feature information that causes defects in different batches of profiles and the profiles to be framed; Calculate the correlation between different batches of profiles and the profile to be framed according to the multi-dimensional feature information ; in , Indicates the number of feature information dimensions between the batch profile and the profile to be framed, Indicates the number of feature information of a certain dimension and , Indicates the first dimension of a batch profile feature information, Indicates the first dimension of the profile to be framed feature information, Indicates The correlation coefficient between the batch profiles corresponding to the dimension feature information and the profiles to be framed, represents the proportional adjustment coefficient; The specific formula for calculating the weight index of different detection areas according to the detection weight of the defect detection area and the correlation degree is: ; in, Indicates The weight index of the defect detection area, Indicates The detection weight of each defect detection area.

6. An automatic control device for a framing machine, characterized in that: The automatic control device of the frame assembling machine comprises: Controller; A memory storing executable instructions; The executable instructions can be run on the controller and implement the automatic control method for the framing machine as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the automatic control method for a framing machine according to any one of claims 1 to 4 is implemented.

Images