Multi-hole-site plug insulator injection molding method and system based on feature recognition

Abstract

The invention relates to the technical field of plug injection molding, in particular to a multi-hole-site plug insulator injection molding method and system based on feature recognition, and the method comprises the steps: carrying out the thread end recognition of a thread end image to be subjected to injection molding, and obtaining a thread end type set and a template style; line object recognition is conducted on the to-be-injection-molded line end image, a line recognition result is obtained, and according to the line recognition result, strand splitting operation is conducted on the to-be-injection-molded plug line ends, and regularly-arranged line ends are obtained; according to the template style, performing position arrangement on each thread end in the regularly arranged thread ends to obtain distributed thread ends; the distributed thread ends are inserted into a dynamic compensation injection mold, the dynamic compensation injection mold is utilized to conduct self-adaptive attaching and fixing on the distributed thread ends, and a standard thread end distribution mold is obtained; and performing injection molding operation on the standard thread end distribution mold by using a pre-constructed injection molding machine to obtain an injection-molded plug. According to the invention, the insulator injection molding efficiency of the multi-hole-site plug can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of plug injection molding, and particularly to an injection molding method and system for a multi-hole plug insulator based on feature recognition. Background Art

[0002] A plug is the main structure for connecting a device to a power supply. To ensure the safety during plug insertion and removal, insulators are now added to plugs through injection molding. Most of the current plug insulator injection molding methods are as follows: Through manual operation, each lead of the plug is placed into the corresponding mold, and then the mold with the leads is placed on the injection molding table to complete the plug injection molding process.

[0003] However, in the above injection molding method, in order to avoid plastic overflow, the constructed mold and the leads fit closely. Employees rely on vision to insert the leads into the corresponding holes of the mold, resulting in low installation efficiency. In addition, with the continuous update of plug types, the number and distribution positions of the leads also change, and fixed molds are difficult to adapt to the production of diverse plugs. Summary of the Invention

[0004] The present invention provides an injection molding method for a multi-hole plug insulator based on feature recognition, and its main purpose is to improve the injection molding efficiency of the multi-hole plug insulator.

[0005] To achieve the above purpose, an injection molding method for a multi-hole plug insulator based on feature recognition provided by the present invention includes:

[0006] Using a pre-constructed camera device to photograph a pre-constructed plug lead to be injection molded, obtaining an image of the lead to be injection molded;

[0007] Performing lead type recognition on the image of the lead to be injection molded, obtaining a set of lead types, and using a pre-constructed lead style database to query the set of lead types to obtain the template style of the plug lead to be injection molded;

[0008] Performing line object recognition on the image of the lead to be injection molded, obtaining a line recognition result, and using a pre-constructed rotatable multi-stage comb to perform strand separation on the plug lead to be injection molded according to the line recognition result, obtaining regularly arranged leads;

[0009] Performing position arrangement on each lead in the regularly arranged leads according to the template style, obtaining distributed leads;

[0010] Obtaining an initialized dynamic compensation injection mold, inserting the distributed leads into the dynamic compensation injection mold, and using the dynamic compensation injection mold to perform adaptive fitting and fixing on the distributed leads, obtaining a standard lead distribution mold;

[0011] Using a pre-built injection molding machine, perform an injection molding operation on the standard wire head distribution mold to obtain an injection molded plug.

[0012] Optionally, perform wire head type recognition on the wire head image to be injection molded to obtain a wire head type set, including:

[0013] Perform Gaussian filtering on the wire head image to be injection molded to obtain a noise-reduced wire head image, and perform grayscale processing on the noise-reduced wire head image to obtain a grayscale wire head image;

[0014] Use a pre-built sobel operator to perform edge detection on the grayscale wire head image to obtain a wire head contour image;

[0015] Perform image feature extraction on the wire head contour image to obtain a contour feature set, and perform wire head recognition on the contour feature set to obtain a wire head object set;

[0016] According to the contour feature set, perform wire head type classification on the wire head object set to obtain a wire head type set.

[0017] Optionally, perform line object recognition on the wire head image to be injection molded to obtain a line recognition result, including:

[0018] Perform wire line feature extraction on the wire head image to be injection molded to obtain a line feature set, and perform wire axis recognition on the line feature set to obtain a line segment set;

[0019] Configure the line segment set in a pre-built 3D space to obtain line segment distribution data, and obtain the head and tail extension lines of each line segment in the line segment distribution data to obtain a head and tail trend feature direction set;

[0020] According to the head and tail trend feature direction set, perform splicing on each line segment belonging to the same wire in the line segment set to obtain a line recognition result.

[0021] Optionally, use a pre-built rotatable multi-stage comb to perform wire splitting on the wire heads of the plug to be injection molded according to the line recognition result to obtain regularly arranged wire heads, including:

[0022] Sequentially extract one wire from the line recognition result as the target wire, and use the wires other than the target wire in the line recognition result as a wire group;

[0023] Identify the intersection points between the target wire and the wire group to obtain the intersection points and intersection point information, and judge whether rotating the target wire and the wire group can separate the intersection points according to the intersection point information;

[0024] When it is determined that rotating the target wire and the wire group cannot separate the intersection point, return to the step of arbitrarily extracting a wire from the line recognition result as the target wire;

[0025] When it is determined that rotating the target wire and the wire group can separate the intersection point, use a pre-constructed rotatable multi-stage comb to rotate and separate the target wire and the wire group to obtain a separation result;

[0026] If the separation result is that there is a preset new intersection point, return to the step of identifying the intersection point between the target wire and the wire group to obtain intersection point information;

[0027] If the separation result is that all the wire ends in the wire ends to be injection-molded plug are separated, obtain the regularly arranged wire ends.

[0028] Optionally, the judgment of whether rotating the target wire and the wire group can separate the intersection point includes:

[0029] Use a pre-trained intersection point solving model to perform feature extraction operations on the intersection point information to obtain an intersection point convolution feature set;

[0030] Perform average pooling operations on the intersection point convolution feature set to obtain an intersection point dimensionality reduction feature set;

[0031] Perform decision tree classification and judgment operations on the intersection point dimensionality reduction feature set to obtain an intersection point recognition result;

[0032] According to the intersection point recognition result, judge whether rotating the target wire and the wire group can separate the intersection point.

[0033] Optionally, before using the pre-trained intersection point solving model, the method further includes:

[0034] Obtain an initialized intersection point solving model and obtain a knot sample set, where the knot sample set includes knot samples and true solving labels;

[0035] Successively extract a knot sample from the knot sample set, and use the intersection point solving model to solve and predict the knot sample to obtain a predicted solving result;

[0036] Judge the loss value between the predicted solving result and the true solving label of the knot sample, minimize the loss value, and obtain the network model parameters when the loss value is the smallest;

[0037] According to the network model parameters, perform reverse network parameter update on the intersection point solving model to obtain an updated intersection point solving model;

[0038] Determine whether there is a knot sample in the set of knot samples;

[0039] When there is a knot sample in the set of knot samples, return the step of sequentially extracting a knot sample from the set of knot samples;

[0040] When there is no knot sample in the set of knot samples, obtain the trained intersection solving model.

[0041] Optionally, the arranging the positions of each wire end in the regularly arranged wire ends according to the template style to obtain distributed wire ends includes:

[0042] Obtain the wire end arrangement rule, the line arrangement rule, and the reserved length of the wire end in the template style;

[0043] According to the line arrangement rule, wind the regularly arranged wire ends until the distance between the wire and the wire end is the reserved length of the wire end, obtaining the wire ends to be distributed;

[0044] According to the wire end arrangement rule, move the positions of the wire ends to be distributed to obtain distributed wire ends.

[0045] Optionally, the adaptively fitting and fixing the distributed wire ends by using the dynamic compensation injection mold to obtain a standard wire end distribution mold includes:

[0046] Use the dynamic compensation injection mold to extrude each wire end in the wire ends to be distributed according to a preset pressure standard to obtain a fixed wire end distribution;

[0047] According to the preset style specifications in the template style, perform fine-tuning operations on each fixed wire end in the fixed wire end distribution based on the position distance and angle to obtain a standard wire end distribution mold.

[0048] Optionally, the injection molding operation on the standard wire end distribution mold by using a pre-built injection molding machine to obtain an injection molded plug includes:

[0049] During the process of performing the injection molding operation on the standard wire end distribution mold by using a pre-built injection molding machine, obtain the pressure information of the plastic on the dynamic compensation injection mold to obtain the injection pressure;

[0050] When the injection pressure is greater than a preset qualified threshold, stop the process of performing the injection molding operation on the standard wire end distribution mold to obtain a primary injection molded plug;

[0051] After cooling the primary injection molded plug, use the dynamic compensation injection mold to push out the cooled primary injection molded plug from the dynamic compensation injection mold according to a preset output pressure to obtain an injection molded plug.

[0052] To achieve the above object, the present invention further provides a multi-hole plug insulator injection molding system based on feature recognition, including:

[0053] An injection molding style recognition module, configured to use a pre-built camera device to photograph a pre-built plug wire head to be injection molded to obtain an image of the wire head to be injection molded, and perform wire head type recognition on the image of the wire head to be injection molded to obtain a set of wire head types, and use a pre-built wire style database to query the set of wire head types to obtain the template style of the plug wire head to be injection molded;

[0054] A wire head combing module, configured to perform line object recognition on the image of the wire head to be injection molded to obtain a line recognition result, and use a pre-built rotatable multi-stage comb to perform strand separation operation on the plug wire head to be injection molded according to the line recognition result to obtain a regularly arranged wire head;

[0055] A plug wire head stabilization module, configured to arrange the positions of each wire head in the regularly arranged wire head according to the template style to obtain a distributed wire head, and obtain an initialized dynamic compensation injection mold, insert the distributed wire head into the dynamic compensation injection mold, and use the dynamic compensation injection mold to perform adaptive fitting and fixing on the distributed wire head to obtain a standard wire head distribution mold;

[0056] An injection molding module, configured to use a pre-built injection molding machine to perform injection molding operation on the standard wire head distribution mold to obtain an injection molded plug.

[0057] To solve the above problems, the present invention further provides an electronic device, which includes:

[0058] A memory, storing at least one instruction;

[0059] A processor, executing the instruction stored in the memory to implement the above-mentioned multi-hole plug insulator injection molding method based on feature recognition.

[0060] To solve the above problems, the present invention further provides a computer-readable storage medium, in which at least one instruction is stored, and the at least one instruction is executed by a processor in an electronic device to implement the above-mentioned multi-hole plug insulator injection molding method based on feature recognition.

[0061] To solve the problems described in the background art, the present invention first takes an image of the plug wire head to be injection-molded to obtain an image of the wire head to be injection-molded, and then uses an identification algorithm to identify the wire head type and wire distribution in the image of the wire head to be injection-molded, respectively obtaining a wire head type set and a wire recognition result. Among them, the wire head type set can query the wire head style database to know the template style of the current plug wire head to be injection-molded, and the wire recognition result can know whether there are phenomena such as winding of each wire head, so as to provide a suitable regular arrangement of wire heads for subsequent injection molding and improve the product qualification rate. Then, the present invention uses a dynamic compensation injection mold to fix and fine-tune the regularly arranged wire heads to obtain a standard wire head distribution mold, and then performs injection molding to obtain an injection-molded plug. Among them, the standard wire head distribution mold has an embedded micro electric cylinder that can change the depth, width and position of the wire head fixing groove. Therefore, the present invention can improve the injection molding efficiency of the insulator of the multi-hole plug. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 FIG. is a schematic flow chart of a method for injecting an insulator of a multi-hole plug based on feature recognition provided by an embodiment of the present invention;

[0063] Figure 2 FIG. is a functional module diagram of an injection molding system for an insulator of a multi-hole plug based on feature recognition provided by an embodiment of the present invention;

[0064] Figure 3 FIG. is a schematic structural diagram of an electronic device for implementing the method for injecting an insulator of a multi-hole plug based on feature recognition provided by an embodiment of the present invention.

[0065] DESCRIPTION OF THE REFERENCE NUMERALS:

[0066] 1. Electronic device; 10. Processor; 11. Memory; 12. Bus.

[0067] The realization, functional features and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0068] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0069] An embodiment of the present application provides a method for injection molding a multi - hole plug insulator based on feature recognition. The execution subject of the method for injection molding a multi - hole plug insulator based on feature recognition includes, but is not limited to, at least one of electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for injection molding a multi - hole plug insulator based on feature recognition can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to: a single server, a server cluster, a cloud server, or a cloud server cluster, etc.

[0070] Referring to Figure 1 As shown, it is a schematic flowchart of a method for injection molding a multi - hole plug insulator based on feature recognition provided by an embodiment of the present invention. In this embodiment, the method for injection molding a multi - hole plug insulator based on feature recognition includes:

[0071] S1. Use a pre - constructed camera device to take a picture of a pre - constructed plug wire head to be injection - molded, and obtain an image of the wire head to be injection - molded.

[0072] Among them, the camera device is an electronic device that captures static images or dynamic videos, and generally includes a lens, an image sensor, and a processing unit using optics.

[0073] Among them, the plug wire head to be injection - molded is composed of a wire and a wire head. The first half area of the wire has been injection - molded, and the second half area of the wire is connected to the wire head. The wire head is a metal piece on the socket for contacting the power supply.

[0074] Among them, the image of the wire head to be injection - molded is the shooting result of the plug wire head to be injection - molded.

[0075] Specifically, in an embodiment of the present invention, an optical CCD camera is used to take a picture of the plug wire head to be injection - molded placed on a fixed bayonet to obtain an image of the wire head to be injection - molded, and then the image of the wire head to be injection - molded is sent to the processor for injection - molding control to cooperate with subsequent injection - molding work.

[0076] S2. Perform wire - head type recognition on the image of the wire head to be injection - molded to obtain a set of wire - head types, and use a pre - constructed wire - head style database to query the set of wire - head types to obtain the template style of the plug wire head to be injection - molded.

[0077] Among them, the wire - head type recognition refers to the operation of recognizing the type of the wire head in the image through a neural network. The set of wire - head types is the recognition result of the wire - head type recognition process.

[0078] Among them, the wire - head style database is a database that stores template information of various plugs. By simply updating the wire - head style database, adaptive injection - molding of new types of plugs can be achieved.

[0079] Among them, the template style is some key information of the plug, such as the size of the plug, the number of wire ends of the plug (such as 2-wire plug, 3-wire plug, USB plug, etc.), the arrangement order of each wire end, and the type of each wire end (sheet-like, convex, cylindrical, etc.).

[0080] Specifically, in the embodiment of the present invention, the identification of the wire end type of the to-be-injected wire end image to obtain a wire end type set includes:

[0081] Perform Gaussian filtering on the to-be-injected wire end image to obtain a noise-reduced wire end image, and perform grayscale processing on the noise-reduced wire end image to obtain a grayscale wire end image;

[0082] Use a pre-constructed sobel operator to perform edge detection on the grayscale wire end image to obtain a wire end contour image;

[0083] Perform image feature extraction on the wire end contour image to obtain a contour feature set, and perform wire end recognition on the contour feature set to obtain a wire end object set;

[0084] According to the contour feature set, perform wire end type classification on the wire end object set to obtain a wire end type set.

[0085] Among them, the Gaussian filtering process refers to the process of convolving an image with a Gaussian function to smooth the image and reduce noise, which is used for image denoising and blurring. The noise-reduced wire end image is the denoising result of the to-be-injected wire end image.

[0086] Among them, the grayscale processing refers to the process of converting a color image into a grayscale image. Among them, the grayscale wire end image is the grayscale result of the noise-reduced wire end image, which only contains luminance information and does not contain color information.

[0087] Among them, the sobel operator is used to quickly locate regions of sudden grayscale change in an image (such as object contours, texture boundaries, etc.). Its core function is to identify edge information in the image by calculating the gradient intensity of pixel points. The edge detection operation refers to the operation of helping to analyze and understand the image content by identifying significant luminance changes in the image. The wire end contour image is the edge detection result of the grayscale wire end image.

[0088] Among them, the image feature extraction operation refers to the operation of performing convolution and dimensionality reduction through the convolutional layer, pooling layer, and flattening layer of a neural network. The contour feature set is the feature extraction result of the wire end contour image.

[0089] Among them, the wire-end recognition operation refers to the process of classification prediction through the fully connected layer of the neural network. The wire-end object set is the output result of the wire-end recognition operation.

[0090] Among them, the wire-end type classification operation refers to: extracting the features corresponding to each wire-end object from the contour feature set, and then performing the operation of neural network classification judgment. The wire-end type set represents the output result of the wire-end type classification operation.

[0091] Specifically, in the embodiment of the present invention, first, a filter with a Gaussian filtering algorithm is called to denoise the to-be-injected wire-end image to obtain a denoised wire-end image. Then, according to the gray formula, the RGB three-channel color values in each pixel region of the denoised wire-end image are converted into brightness values between 0 and 1 to obtain a gray wire-end image. Then, using a Canny edge detection device, according to the calculation rule of the sobel operator, a wire-end contour image is obtained. Among them, the wire-end contour image contains both object information and object type information.

[0092] Specifically, in the embodiment of the present invention, first, according to the contour feature set of the wire-end contour image, each wire-end object is found to obtain a wire-end object set, excluding the influence of the wire. Then, the contour features of each wire-end object are used for wire-end type recognition to obtain a wire-end type set, for example, sheet type, column type, and semi-column type, etc.

[0093] Further, in the embodiment of the present invention, the wire-end type set (for example, "two sheet-shaped wire-ends, one column-shaped wire-end" or "three sheet-shaped wire-ends") is queried through a wire-end style database to obtain a template style, for example, a certain model of automotive cable plug.

[0094] S3. Perform line object recognition on the to-be-injected wire-end image to obtain a line recognition result, and use a pre-constructed rotatable multi-stage comb to perform a wire splitting operation on the to-be-injected plug wire-end according to the line recognition result to obtain a regularly arranged wire-end.

[0095] Among them, the line object recognition refers to the operation of predicting line objects through the mapping relationship learned by the neural network. The line recognition result is the output result of the line object recognition. When multiple wires interact, it can be observed from the picture that some wires are blocked by other wires at the intersection. Therefore, even the same wire will be divided into many line segments through image recognition.

[0096] Among them, the rotatable multi-stage comb is a precision mechanical device for wire splitting and sorting. Through the coordinated rotational movement of the multi-stage adjustable comb teeth, the randomly wound wires (such as multi-core cables, ribbon cables, etc.) are quickly separated into regularly arranged independent wire bundles. The wire splitting operation refers to the process of separating the wires that are wound together.

[0097] Among them, the regularly arranged wire ends represent the respective wire ends arranged in sequence in a row.

[0098] Specifically, in the embodiment of the present invention, the recognition of line objects for the image of the wire ends to be injection-molded to obtain a line recognition result includes:

[0099] Performing an operation of extracting wire line features on the image of the wire ends to be injection-molded to obtain a set of line features, and performing an operation of recognizing the wire axis on the set of line features to obtain a set of line segments;

[0100] Configuring the set of line segments in a pre-constructed 3D space to obtain line segment distribution data, and obtaining a set of head and tail trend feature directions by obtaining the head and tail extension lines of each line segment in the line segment distribution data;

[0101] According to the set of head and tail trend feature directions, performing a splicing operation on the line segments belonging to the same wire in the set of line segments to obtain a line recognition result.

[0102] Among them, the operation of extracting wire line features is similar to the above-mentioned operation of extracting image features, except that in the convolution process, objects related to lines are extracted. The set of line features is the output result of the operation of extracting wire line features.

[0103] Among them, the operation of recognizing the wire axis refers to transforming the set of long-strip line features into linear features. The set of line segments is the axis result of the set of line features.

[0104] Among them, the configuring of the set of line segments in a pre-constructed 3D space refers to the process of mapping the two-dimensional set of line segments into a three-dimensional result. The line segment distribution data is the three-dimensional mapping result of the set of line segments.

[0105] Among them, the head and tail extension lines refer to the tangents of the head and tail points of each line segment. The set of head and tail trend feature directions is the set of the head and tail extension lines of each line segment.

[0106] Among them, the splicing operation refers to the operation of connecting the line segments with a connection trend. The line recognition result is the output result of the splicing operation. For example, when the wire end of the plug to be injection-molded is a three-pole plug, the line recognition result is three lines.

[0107] Specifically, in the embodiments of the present invention, first, a configured feature extraction network is used to extract features of linear objects in the image of the to-be-injected plug head to obtain a set of line features. Assuming that each wire has a uniform thickness, the wire can be represented by the axis of the wire to obtain a set of line segments. Since the set of line segments is in a two-dimensional form, in order to more clearly understand which wire each line segment belongs to, the present invention performs a three-dimensional mapping on the set of line segments through a preset 3D space to obtain line segment distribution data. Then, by checking whether the included angle between the head and tail extension lines of each line segment is less than a preset trend threshold, such as 0.01, it is determined whether there is a connection trend between each line segment, and thus each line segment with a connection trend is spliced to obtain a line recognition result.

[0108] Specifically, in the embodiments of the present invention, the step of using a pre-constructed rotatable multi-level comb to split the to-be-injected plug head according to the line recognition result to obtain a regularly arranged plug head includes:

[0109] Extract one wire as the target wire from the line recognition result in sequence, and use the wires other than the target wire in the line recognition result as a wire group;

[0110] Identify the intersection points between the target wire and the wire group to obtain the intersection points and intersection information, and judge whether rotating the target wire and the wire group can separate the intersection points according to the intersection information;

[0111] When it is determined that rotating the target wire and the wire group cannot separate the intersection points, return to the step of arbitrarily extracting one wire from the line recognition result as the target wire;

[0112] When it is determined that rotating the target wire and the wire group can separate the intersection points, use the pre-constructed rotatable multi-level comb to perform rotational separation on the target wire and the wire group to obtain a separation result;

[0113] If the separation result is that there are preset new intersection points, return to the step of identifying the intersection points between the target wire and the wire group to obtain the intersection information;

[0114] If the separation result is that all the plug head wires in the to-be-injected plug head are separated, a regularly arranged plug head is obtained.

[0115] Wherein, the intersection point refers to a point where the wires cannot be separated. The intersection information is the positional relationship of each wire at the intersection point.

[0116] Among them, the rotation separation is close to the splitting operation and is the changing operation of each comb tooth in the rotatable multi-stage comb teeth. The separation results include successful separation without intersections, successful separation but with new intersections, and failed separation.

[0117] Specifically, in the embodiment of the present invention, by configuring the target wire and the wire group, the influence of only one wire on the intersection is adjusted each time. Through the traversal process, the target wire is continuously changed, so as to untie one intersection after another until there is no intersection (the wires of each wire head do not intersect each other), and a neatly arranged wire head is obtained.

[0118] In detail, in the embodiment of the present invention, the judgment of whether rotating the target wire and the wire group can separate the intersection includes:

[0119] Using a pre-trained intersection solution model, perform feature extraction operations on the intersection information to obtain an intersection convolution feature set;

[0120] Perform average pooling operations on the intersection convolution feature set to obtain an intersection dimensionality reduction feature set;

[0121] Perform decision tree classification and judgment operations on the intersection dimensionality reduction feature set to obtain an intersection recognition result;

[0122] According to the intersection recognition result, judge whether rotating the target wire and the wire group can separate the intersection.

[0123] Among them, the intersection solution model is a regression network model for learning the mapping relationship between each rope intersection and the solution method.

[0124] Among them, the feature extraction operation is the same as the above-mentioned feature extraction operation, except that the extraction object is intersection information this time. The intersection convolution feature set is the feature extraction result of the intersection information.

[0125] Among them, the average pooling operation is a downsampling algorithm in the pooling layer, which is used to reduce the magnitude of features while retaining important feature information. The intersection dimensionality reduction feature set is the output result of the average pooling operation.

[0126] Among them, the decision tree classification and judgment operation refers to the operation of continuously classifying the feature extraction results through a decision tree forest to achieve the final classification task. The intersection recognition result is the judgment result of whether the target area is an intersection.

[0127] Specifically, in the embodiments of the present invention, a pre-trained intersection solving model is used to obtain the intersection recognition result through the processes of feature extraction, dimensionality reduction, and classification recognition of the neural network. When the intersection recognition result is an intersection type, it is determined that rotating the target wire and the wire group cannot separate the intersection. When the intersection recognition result is a non-intersection type, it is determined that rotating the target wire and the wire group cannot separate the intersection.

[0128] Specifically, in the embodiments of the present invention, before using the pre-trained intersection solving model, the method further includes:

[0129] Obtain an initialized intersection solving model and obtain a knot sample set, where the knot sample set includes knot samples and true solving labels;

[0130] Extract one knot sample from the knot sample set in sequence, and use the intersection solving model to solve and predict the knot sample to obtain a predicted solving result;

[0131] Judge the loss value between the predicted solving result and the true solving label of the knot sample, minimize the loss value, and obtain the network model parameters when the loss value is minimized;

[0132] Reverse update the network parameters of the intersection solving model according to the network model parameters to obtain an updated intersection solving model;

[0133] Judge whether there is a knot sample in the knot sample set;

[0134] When there is a knot sample in the knot sample set, return to the step of extracting one knot sample from the knot sample set in sequence;

[0135] When there is no knot sample in the knot sample set, obtain the trained intersection solving model.

[0136] Among them, the knot sample set is: knot samples obtained by manually winding different numbers (2 to 6) of ropes and then taking pictures, and the true solving labels are obtained through actual tests on the knot samples.

[0137] Among them, the solving prediction is the forward execution process of the intersection solving model. The predicted solving result is the output result of the forward execution process.

[0138] Among them, the process of judging the loss value between the predicted solving result and the true solving label of the knot sample refers to the process of calculating through the cross-entropy loss algorithm. The loss value is the result of the cross-entropy loss algorithm.

[0139] Among them, minimizing the loss value refers to a method of controlling the training direction of the model through the gradient descent algorithm. The network model parameters are the numerical values of each parameter term when minimizing the loss value.

[0140] Among them, the reverse update operation of the network parameters refers to the process of solving the intersection model according to the network model parameters. The updated intersection model is the updated result of the intersection model.

[0141] Specifically, in the embodiment of the present invention, first, an initialized regression function equation is constructed to obtain an intersection solving model. Then, through the cross-entropy loss algorithm and the gradient descent algorithm, the loss value of the knot samples is continuously calculated to obtain the network model parameters. Then, using the network model parameters, the number of terms and the weight of each term in the regression function equation are updated. Finally, an updated regression function equation with an accuracy of over 90% is obtained, thereby completing the training process of the intersection solving model.

[0142] S4. According to the template style, arrange the positions of each wire head in the regularly arranged wire heads to obtain distributed wire heads.

[0143] Among them, the position arrangement refers to changing the positions of each wire head in the regularly arranged wire heads. The distributed wire heads are the positions of each wire head after arrangement.

[0144] Specifically, in the embodiment of the present invention, after the regularly arranged wire heads are obtained, they are arranged horizontally one by one. However, since the plug has requirements for the positions of the wire heads, for example, two sheet-shaped wire heads are located on one side of the plug, and one columnar wire head is on the perpendicular bisector of the two sheet-shaped wire heads, and the three form an isosceles triangle. Therefore, the present invention divides the positions in advance before injection molding to obtain distributed wire heads.

[0145] In detail, in the embodiment of the present invention, arranging the positions of each wire head in the regularly arranged wire heads according to the template style to obtain distributed wire heads includes:

[0146] Obtain the wire head arrangement rule, line arrangement rule, and wire head reserved length in the template style;

[0147] According to the line arrangement rule, wind the regularly arranged wire heads until the distance between the wire and the wire head is the wire head reserved length to obtain wire heads to be distributed;

[0148] According to the wire head arrangement rule, move the positions of the wire heads to be distributed to obtain distributed wire heads.

[0149] Among them, the wire head arrangement rule refers to the distance between each other and the inclination angle of each wire head.

[0150] Among them, the wire arrangement rule refers to how each wire is arranged before entering the plug, such as parallel arrangement, circular arrangement, three-strand braiding, etc.

[0151] Among them, the reserved length of the wire end refers to the area of the wire that can move freely and is not restricted by the wire arrangement rule. It is configured to be 2 cm.

[0152] Among them, the winding is an operation of implementing the wire arrangement rule by using a rotatable multi-stage comb. The wire ends to be distributed refer to the distribution situation of the remaining wires and wire ends in the plug after the wires are wound.

[0153] Among them, the position movement includes movement and rotation, which are used to change the positional relationship between the wire ends in the wire ends to be distributed.

[0154] Specifically, in the embodiment of the present invention, first, according to the wire arrangement rule in the template style, the wires before 2 cm from the wire-wire end junction are arranged, and then according to the wire end arrangement rule in the template style, each wire end in the wire ends to be distributed is moved and rotated to obtain the distributed wire ends.

[0155] S5. Obtain an initialized dynamic compensation injection mold, insert the distributed wire ends into the dynamic compensation injection mold, and use the dynamic compensation injection mold to adaptively fit and fix the distributed wire ends to obtain a standard wire end distribution mold.

[0156] Among them, the dynamic compensation injection mold is a mold with an embedded micro electric cylinder that can change the depth, width, position, and angle of the wire end fixing groove. The initialized dynamic compensation injection mold refers to configuring the slot positions in the mold to preset numerical values so that the distributed wire ends can be inserted into the slots in a lower calibration scenario.

[0157] Among them, the adaptive fitting and fixing refers to the process of compressing each wire end from all around and in depth until the wire end is clamped. The standard wire end distribution mold refers to a mold-wire end integrated structure in which the distributed wire ends are fixed according to the template style.

[0158] Specifically, in the embodiment of the present invention, the distributed wire ends are inserted into the dynamic compensation injection mold through a manipulator or a push-pull structure.

[0159] In detail, in the embodiment of the present invention, the use of the dynamic compensation injection mold to adaptively fit and fix the distributed wire ends to obtain a standard wire end distribution mold includes:

[0160] Using the dynamic compensation injection mold to extrude each wire end in the wire ends to be distributed according to a preset pressure standard to obtain a fixed wire end distribution;

[0161] According to the preset style specifications in the template style, perform fine-tuning operations on each fixed wire head in the fixed wire head distribution based on position distance and angle to obtain a standard wire head distribution mold.

[0162] Among them, the pressure standard is related to the pressure in the injection molding process, so that each wire head will not move during the subsequent injection molding process.

[0163] Among them, the extrusion is the above-mentioned process of adaptive fitting and fixing.

[0164] Among them, the style specification is a refined version of the wire head arrangement rule.

[0165] Among them, the fine-tuning operation refers to the operation of adjusting the position and angle between each wire head.

[0166] Specifically, in the embodiment of the present invention, during the process of the dynamic compensation injection mold extruding the wire heads to be distributed, due to the different contact times of each extrusion position, the final fixed wire head distribution deviates from the distributed wire heads. Therefore, re-perform fine-tuning according to the preset style specifications in the template style to obtain a standard wire head distribution mold.

[0167] S6. Use a pre-built injection molding machine to perform injection molding operations on the standard wire head distribution mold to obtain an injection molded plug.

[0168] Among them, the injection molding machine is a device for plastic processing, mainly used for manufacturing various plastic products. The injection molding operation refers to the process of heating and melting plastic raw materials and then injecting them into the mold to cool and form, so as to produce the required plastic plug.

[0169] In detail, in the embodiment of the present invention, the use of a pre-built injection molding machine to perform injection molding operations on the standard wire head distribution mold to obtain an injection molded plug includes:

[0170] During the process of using a pre-built injection molding machine to perform injection molding operations on the standard wire head distribution mold, obtain the pressure information of the plastic on the dynamic compensation injection mold to obtain the injection molding pressure;

[0171] When the injection molding pressure is greater than the preset qualified threshold, stop the process of performing injection molding operations on the standard wire head distribution mold to obtain a primary injection molded plug;

[0172] After cooling the primary injection molded plug, use the dynamic compensation injection mold to push out the cooled primary injection molded plug from the dynamic compensation injection mold according to the preset output pressure to obtain an injection molded plug.

[0173] Among them, the injection pressure refers to the pressure exerted by the liquid plastic on the dynamic compensation injection mold during the injection process. The density of the plastic can be changed through the injection pressure.

[0174] Among them, the qualified threshold is adjusted according to the type of the plug and the working environment.

[0175] Among them, the primary injection plug refers to the plug when the plastic is still in a liquid or semi-solid state.

[0176] Among them, the cooling can be achieved through heat conduction, air cooling and other means.

[0177] Among them, the output pressure is used to push the cooled plug out of the dynamic compensation injection mold. The injection plug is the final plug product.

[0178] Specifically, in the embodiment of the present invention, during the injection process, by regulating the injection pressure, the density of the plastic can be changed, and then the properties of the plastic can be changed, so that the product is suitable for various environments. Therefore, the injection progress can be controlled by adjusting the preset qualified threshold to obtain the primary injection plug. However, the primary injection plug at this time is semi-solid and needs to be cooled before it can be used. When the primary injection plug is cooled, it can be regarded as the final injection plug, but it is still in the dynamic compensation injection mold and needs to be extracted. At this time, the motor in the dynamic compensation injection mold can be used to achieve automatic ejection, thereby further improving the production efficiency of the injection plug.

[0179] To solve the problems described in the background art, the present invention first takes an image of the wire head of the plug to be injected to obtain an image of the wire head to be injected, and then through an identification algorithm, identifies the type of the wire head and the line distribution in the image of the wire head to be injected, and respectively obtains a set of wire head types and a line identification result. Among them, the set of wire head types can query the wire head style database to know the template style of the current wire head of the plug to be injected, and the line identification result can know whether there are phenomena such as winding of each wire head, so as to provide a suitable regular arrangement of wire heads for subsequent injection and improve the product qualification rate; then, the present invention uses a dynamic compensation injection mold to fix and fine-tune the regularly arranged wire heads to obtain a standard wire head distribution mold, and then injects to obtain an injection plug, wherein the standard wire head distribution mold has an embedded micro electric cylinder that can change the depth, width and position of the wire head fixing groove. Therefore, the present invention can improve the injection efficiency of the insulator of the multi-hole plug.

[0180] As Figure 2 shown, it is a functional module diagram of a multi-hole plug insulator injection system based on feature recognition provided by an embodiment of the present invention.

[0181] The multi - hole plug insulator injection - molding system 100 based on feature recognition according to the present invention can be installed in an electronic device. According to the functions achieved, the multi - hole plug insulator injection - molding system 100 based on feature recognition can include an injection - molding style recognition module 101, a wire - end combing module 102, a plug wire - end stabilization module 103, and an injection - molding module 104. The modules in the present invention can also be referred to as units, which refer to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, and are stored in the memory of the electronic device.

[0182] The injection - molding style recognition module 101 is used to use a pre - constructed camera device to photograph a pre - constructed plug wire - end to be injection - molded, obtain an image of the wire - end to be injection - molded, and perform wire - end type recognition on the image of the wire - end to be injection - molded to obtain a set of wire - end types, and use a pre - constructed wire - end style database to query the set of wire - end types to obtain the template style of the plug wire - end to be injection - molded;

[0183] The wire - end combing module 102 is used to perform line - object recognition on the image of the wire - end to be injection - molded, obtain a line - recognition result, and use a pre - constructed rotatable multi - stage comb to perform strand - splitting operation on the plug wire - end to be injection - molded according to the line - recognition result to obtain regularly arranged wire - ends;

[0184] The plug wire - end stabilization module 103 is used to arrange the positions of each wire - end in the regularly arranged wire - ends according to the template style to obtain distributed wire - ends, and obtain an initialized dynamic - compensation injection - molding die, insert the distributed wire - ends into the dynamic - compensation injection - molding die, and use the dynamic - compensation injection - molding die to perform adaptive fitting and fixing on the distributed wire - ends to obtain a standard wire - end distribution die;

[0185] The injection - molding module 104 is used to use a pre - constructed injection - molding machine to perform injection - molding operation on the standard wire - end distribution die to obtain an injection - molded plug.

[0186] Specifically, each module in the multi - hole plug insulator injection - molding system 100 based on feature recognition in the embodiment of the present invention uses the same technical means as those Figure 1 described in the multi - hole plug insulator injection - molding method based on feature recognition, and can produce the same technical effects, which will not be elaborated here.

[0187] As Figure 3 shown, it is a schematic structural diagram of an electronic device for implementing the multi - hole plug insulator injection - molding method based on feature recognition provided by an embodiment of the present invention.

[0188] The electronic device 1 may include a processor 10, a memory 11, and a bus 12, and may further include a computer program stored in the memory 11 and executable on the processor 10, such as a multi-hole plug insulator injection molding method program based on feature recognition.

[0189] Among them, the memory 11 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 may be an internal storage unit of the electronic device 1, such as the mobile hard disk of the electronic device 1. In some other embodiments, the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the electronic device 1. Further, the memory 11 further includes the internal storage unit of the electronic device 1 and also includes an external storage device. The memory 11 can not only be used to store application software installed in the electronic device 1 and various types of data, such as the code of the multi-hole plug insulator injection molding method program based on feature recognition, etc., but also be used to temporarily store data that has been output or will be output.

[0190] In some embodiments, the processor 10 may be composed of integrated circuits. For example, it may be composed of a single packaged integrated circuit, or may be composed of multiple integrated circuits with the same or different functions, including a combination of one or more Central Processing Units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control core (Control Unit) of the electronic device, connecting various components of the entire electronic device through various interfaces and lines, and by running or executing programs or modules stored in the memory 11 (such as the multi-hole plug insulator injection molding method program, etc.), and calling data stored in the memory 11, to execute various functions of the electronic device 1 and process data.

[0191] The bus 12 can be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to implement the connection and communication between the memory 11, at least one processor 10, and the like.

[0192] Figure 3 Only an electronic device with components is shown. Those skilled in the art can understand that Figure 3 the shown structure does not constitute a limitation on the electronic device 1, and it may include fewer or more components than shown, or combine certain components, or have a different component arrangement.

[0193] For example, although not shown, the electronic device 1 may further include a power source (such as a battery) for powering each component. Preferably, the power source can be logically connected to the at least one processor 10 through a power management device, so as to implement functions such as charge management, discharge management, and power consumption management through the power management device. The power source may also include any components such as one or more DC or AC power sources, a recharge device, a power failure detection circuit, a power converter or inverter, and a power status indicator. The electronic device 1 may also include various sensors, a Bluetooth module, a Wi-Fi module, etc., which will not be elaborated here.

[0194] Furthermore, the electronic device 1 may further include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0195] Optionally, the electronic device 1 may further include a user interface. The user interface may be a display, an input unit (such as a keyboard), and optionally, the user interface may also be a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch liquid crystal display, and an OLED (Organic Light-Emitting Diode) toucher, etc. Among them, the display may also be appropriately referred to as a display screen or a display unit, which is used to display the information processed in the electronic device 1 and to display a visual user interface.

[0196] The program of the method for injection molding of a multi-hole plug insulator based on feature recognition stored in the memory 11 in the electronic device 1 is a combination of multiple instructions, and when running in the processor 10, it can achieve:

[0197] Using a pre-built camera device, photograph a pre-built plug wire head to be injection molded to obtain an image of the wire head to be injection molded;

[0198] Perform wire head type recognition on the image of the wire head to be injection molded to obtain a set of wire head types, and use a pre-built wire head style database to query the set of wire head types to obtain the template style of the plug wire head to be injection molded;

[0199] Perform line object recognition on the image of the wire head to be injection molded to obtain a line recognition result, and use a pre-built rotatable multi-stage comb to perform strand separation on the plug wire head to be injection molded according to the line recognition result to obtain regularly arranged wire heads;

[0200] Arrange the positions of each wire head in the regularly arranged wire heads according to the template style to obtain distributed wire heads;

[0201] Obtain an initialized dynamic compensation injection mold, insert the distributed wire heads into the dynamic compensation injection mold, and use the dynamic compensation injection mold to perform adaptive fitting and fixing on the distributed wire heads to obtain a standard wire head distribution mold;

[0202] Use a pre-built injection molding machine to perform injection molding on the standard wire head distribution mold to obtain an injection molded plug.

[0203] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figures 1 to 3 The description of the relevant steps in the corresponding embodiment, which will not be elaborated here.

[0204] Furthermore, if the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM, Read-Only Memory).

[0205] The present invention also provides a computer-readable storage medium, and the readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, it can achieve:

[0206] Using a pre-built camera device, photograph a pre-built plug wire head to be injection molded to obtain an image of the wire head to be injection molded;

[0207] Perform wire head type recognition on the image of the wire head to be injection molded to obtain a set of wire head types, and use a pre-built wire head style database to query the set of wire head types to obtain the template style of the plug wire head to be injection molded;

[0208] Perform line object recognition on the image of the wire head to be injection molded to obtain a line recognition result, and use a pre-built rotatable multi-stage comb to perform strand separation on the plug wire head to be injection molded according to the line recognition result to obtain regularly arranged wire heads;

[0209] Arrange the positions of each wire head in the regularly arranged wire heads according to the template style to obtain distributed wire heads;

[0210] Obtain an initialized dynamic compensation injection mold, insert the distributed wire heads into the dynamic compensation injection mold, and use the dynamic compensation injection mold to perform adaptive fitting and fixing on the distributed wire heads to obtain a standard wire head distribution mold;

[0211] Use a pre-built injection molding machine to perform injection molding on the standard wire head distribution mold to obtain an injection molded plug.

[0212] In several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, and there can be other division methods in actual implementation.

[0213] The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0214] In addition, the functional modules in each embodiment of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware, or in the form of a combination of hardware and software functional modules.

[0215] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention.

[0216] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for injection molding a multi-hole plug insulator based on feature recognition, characterized in that: The method comprises: Using a pre-built camera device, photographing the pre-built plug thread end to be injected, to obtain an image of the thread end to be injected; Performing thread head type recognition on the thread head image to be injected to obtain a thread head type set, and using a pre-built thread head style database to query the thread head type set to obtain a template style of the thread head of the plug to be injected; Performing line object recognition on the image of the thread end to be injected, obtaining a line recognition result, and using a pre-built rotatable multi-stage comb, performing a stranding operation on the thread end of the plug to be injected according to the line recognition result, to obtain a regularly arranged thread end; Arranging the positions of the thread ends in the regularly arranged thread ends according to the template pattern to obtain distributed thread ends; Obtaining an initialized dynamic compensation injection mold, inserting the distributed thread heads into the dynamic compensation injection mold, and using the dynamic compensation injection mold to adaptively fit and fix the distributed thread heads to obtain a standard thread head distribution mold; The standard wire head distribution mold is injection molded by using a pre-built injection molding machine to obtain an injection molded plug.

2. The method for injection molding a multi-hole plug insulator based on feature recognition according to claim 1, characterized in that: The step of performing thread head type identification on the thread head image to be injected to obtain a thread head type set includes: Performing Gaussian filtering on the image of the thread end to be injected to obtain a denoised thread end image, and performing grayscale processing on the denoised thread end image to obtain a grayscale thread end image; Using a pre-built Sobel operator, an edge detection operation is performed on the grayscale line head image to obtain a line head contour image; Performing an image feature extraction operation on the thread head contour image to obtain a contour feature set, and performing a thread head recognition operation on the contour feature set to obtain a thread head object set; According to the contour feature set, a line head type classification operation is performed on the line head object set to obtain a line head type set.

3. The method for injection molding a multi-hole plug insulator based on feature recognition according to claim 2, characterized in that: The performing line object recognition on the image of the to-be-injected line head to obtain a line recognition result includes: Performing a wire line feature extraction operation on the image of the wire end to be injected to obtain a line feature set, and performing a wire axis recognition operation on the line feature set to obtain a line segment set; The line segment set is configured in a pre-constructed 3D space to obtain line segment distribution data, and the head and tail extension lines of each line segment in the line segment distribution data are obtained to obtain a head and tail trend feature direction set; According to the head-to-tail trend characteristic direction set, a splicing operation is performed on each line segment in the line segment set that belongs to the same electric wire to obtain a line recognition result.

4. The method for injection molding a multi-hole plug insulator based on feature recognition according to claim 3, characterized in that: The method utilizes the pre-built rotatable multi-stage comb teeth to separate the wire ends of the plug to be injected into strands according to the line recognition result to obtain regularly arranged wire ends, including: Extracting one wire from the line recognition result in sequence as a target wire, and taking the wires in the line recognition result except the target wire as a wire group; Identify the intersection between the target wire and the wire group, obtain the intersection and intersection information, and determine whether rotating the target wire and the wire group can separate the intersection according to the intersection information; When it is determined that the intersection cannot be separated by rotating the target wire and the wire group, returning to the above step of arbitrarily extracting one wire from the line recognition result as the target wire; When it is determined that rotating the target wire and the wire group can separate the intersection, using the pre-constructed rotatable multi-stage comb teeth to rotate and separate the target wire and the wire group to obtain a separation result; If the separation result shows that there is a preset new intersection, then return to the above step of identifying the intersection between the target wire and the wire group to obtain the intersection information; If the separation result is that all the wire ends in the plug wire ends to be injection molded are separated, the wire ends are neatly arranged.

5. The method for injection molding a multi-hole plug insulator based on feature recognition according to claim 4, characterized in that: The determining whether rotating the target wire and the wire group can separate the intersection point comprises: Using a pre-trained intersection solution model, a feature extraction operation is performed on the intersection information to obtain an intersection convolution feature set; Performing an average pooling operation on the intersection convolution feature set to obtain an intersection dimension reduction feature set; Performing a decision tree classification and judgment operation on the intersection dimension reduction feature set to obtain an intersection recognition result; According to the intersection recognition result, it is determined whether rotating the target wire and the wire group can separate the intersection.

6. The method for injection molding a multi-hole plug insulator based on feature recognition according to claim 5, characterized in that: Before using the pre-trained intersection solution model, the method further includes: Obtain an initialized intersection solution model and obtain a knot sample set, wherein the knot sample set includes knot samples and real solution labels; Extracting a knot sample from the knot sample set in turn, and using the intersection solution model to solve and predict the knot sample to obtain a prediction solution result; Determine the loss value between the predicted solution result and the real solution label of the knot sample, minimize the loss value, and obtain the network model parameters when the loss value is minimized; Reversely updating the network parameters of the intersection solving model according to the network model parameters to obtain an updated intersection solving model; Determining whether there is a knot sample in the knot sample set; When there is a knot sample in the knot sample set, return to the above step of extracting one knot sample from the knot sample set in sequence; When there is no knot sample in the knot sample set, a trained intersection solution model is obtained.

7. The method for injection molding a multi-hole plug insulator based on feature recognition according to claim 6, characterized in that: The step of arranging the positions of the respective thread heads in the regularly arranged thread heads according to the template pattern to obtain the distributed thread heads includes: Obtain thread arrangement rules, line arrangement rules and thread reserve length in the template style; According to the line arrangement rule, the regularly arranged wire ends are wound until the distance between the electric wire and the wire end is the reserved length of the wire end, thereby obtaining the wire ends to be distributed; According to the line head arrangement rule, the line heads to be distributed are moved to obtain distributed line heads.

8. The method for injection molding a multi-hole plug insulator based on feature recognition according to claim 7, characterized in that: The method of using the dynamic compensation injection mold to adaptively fit and fix the distributed thread ends to obtain a standard thread end distribution mold includes: Using the dynamic compensation injection mold, according to a preset pressure standard, each of the thread ends to be distributed is squeezed to obtain a fixed thread end distribution; According to the style specifications preset in the template style, each fixed thread end in the fixed thread end distribution is fine-tuned based on position distance and angle to obtain a standard thread end distribution mold.

9. The method for injection molding a multi-hole plug insulator based on feature recognition according to claim 8, characterized in that: The method of using a pre-built injection molding machine to perform an injection molding operation on the standard line head distribution mold to obtain an injection molded plug comprises: In the process of performing an injection molding operation on the standard thread distribution mold using a pre-built injection molding machine, obtaining pressure information of the plastic on the dynamic compensation injection mold to obtain the injection pressure; When the injection molding pressure is greater than a preset qualified threshold, the process of performing the injection molding operation on the standard thread head distribution mold is stopped to obtain a primary injection molding plug; After the primary injection molded plug is cooled, the dynamically compensated injection mold is used to push the cooled primary injection molded plug out of the dynamically compensated injection mold according to a preset output pressure to obtain an injection molded plug.

10. A multi-hole plug insulator injection molding system based on feature recognition, characterized in that: The system comprises: The injection molding style recognition module is used to use a pre-built camera device to shoot a pre-built plug thread to be injected to obtain an image of the thread to be injected, and to identify the thread type of the thread to be injected to obtain a thread type set, and to use a pre-built thread style database to query the thread type set to obtain a template style of the plug thread to be injected; A thread combing module is used to perform line object recognition on the thread image to be injected, obtain line recognition results, and use pre-built rotatable multi-stage comb teeth to perform stranding operation on the thread to be injected according to the line recognition results to obtain neatly arranged thread ends; A plug wire end stabilization module is used to position each wire end in the regularly arranged wire ends according to the template style to obtain a distributed wire end, and obtain an initialized dynamic compensation injection mold, insert the distributed wire end into the dynamic compensation injection mold, and use the dynamic compensation injection mold to adaptively fit and fix the distributed wire end to obtain a standard wire end distribution mold; The injection molding module is used to use a pre-built injection molding machine to perform an injection molding operation on the standard wire head distribution mold to obtain an injection molded plug.

Images