Processing method and system for injection molding raw materials

Abstract

The invention provides an injection molding raw material processing method and system, and relates to the field of injection molding raw material processing, and the method comprises the steps: inputting an original image to a classification prediction model, and obtaining a current classification result; determining a processing priority according to the current hardness and the size of the raw material; conveying the raw materials corresponding to the first priority to a crushing container, and processing the raw materials corresponding to the first priority by utilizing a crushing device corresponding to the first priority; identifying the first current granularity of the raw materials, conveying the raw materials corresponding to the second priority to the crushing container, and processing the raw materials corresponding to the second priority and the raw materials corresponding to the first current granularity by utilizing a crushing device corresponding to the second priority; and identifying a second current granularity of the raw material, and when the second current granularity is smaller than a second preset granularity, stopping crushing to obtain a target injection molding raw material. The processing priority can be determined according to the current classification result of the raw materials, and the raw materials are crushed according to the processing priority, so that the crushing efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of injection molding raw material processing, and particularly to a method and system for processing injection molding raw materials. Background Art

[0002] The preprocessing and processing method of injection molding raw materials refers to the method of preprocessing and processing plastic raw materials before injection molding. The following problems exist when the existing injection molding raw materials are crushed: First, when the traditional crushing device for injection molding raw materials crushes the injection molding raw materials, due to the difference in the hardness of different raw materials, in the crushing process of transforming the injection molding raw materials from high particle size to low particle size, the existing crushing method of mixing and processing raw materials will lead to a reduction in the crushing efficiency of the injection molding raw materials, and it is easy to cause excessive wear of the crushing rollers. However, adjusting the gap of the meshing components inside the crusher is too large to ensure the crushing accuracy of the raw materials with lower hardness or smaller particle size; Second, because some raw materials are of large size, after the raw materials are poured along the feed hopper, although the raw materials will contact the crushing rollers, it is difficult to crush the raw materials of large size during the meshing rotation of the two crushing rollers, thus reducing the crushing efficiency; Third, plastic containers such as plastic bottles are common plastic products used to form injection molding raw materials. After the plastic containers such as plastic bottles are poured along the feed hopper, the plastic containers will contact the crushing rollers. In the case of less single feeding of the feed hopper, due to the small extrusion force from above on the plastic containers at the lower position and the effect of their own bulging shape, it is difficult to crush the arc surface of the plastic containers during the meshing rotation of the two crushing rollers, causing the bulging plastic containers to move back and forth in the feed hopper under the extrusion of the crushing rollers, affecting the crushing efficiency of plastic products and resulting in low preprocessing and processing efficiency of injection molding raw materials.

[0003] However, there is no technical solution in the prior art that can solve the above technical problems, and there is no method and system for processing injection molding raw materials. Summary of the Invention

[0004] The present invention provides a method and system for processing injection molding raw materials, which can determine the processing priority according to the current classification result of the raw materials, and perform raw material crushing according to the processing priority, thereby improving the crushing efficiency.

[0005] In a first aspect, the present invention provides a method for processing injection molding raw materials, including:

[0006] For the original image corresponding to each raw material, input the original image into the classification prediction model, and obtain the current classification result of the raw material output by the classification prediction model. The classification prediction model is determined after being trained according to the sample image and the classification result corresponding to the sample image;

[0007] Determine the current hardness and softness of the raw material according to the current classification result, and determine the processing priority of the raw material according to the current hardness and softness and the size of the raw material in the original image. The processing priority includes a first priority and a second priority;

[0008] Traverse all the original images to generate a first indication instruction, which is used to indicate conveying the raw material corresponding to the first priority to the crushing container to process the raw material corresponding to the first priority by using the crushing device corresponding to the first priority;

[0009] Identify the first current particle size of the raw material in the crushing container. When it is determined that the first current particle size is less than the first preset particle size, generate a second indication instruction, which is used to indicate conveying the raw material corresponding to the second priority to the crushing container to process the raw material corresponding to the second priority and the raw material corresponding to the first current particle size;

[0010] Identify the second current particle size of the raw material in the crushing container. When it is determined that the second current particle size is less than the second preset particle size, generate a third indication instruction, which is used to indicate stopping the crushing to obtain the target injection molding raw material.

[0011] According to the processing method of the injection molding raw material provided by the present invention, the determining the current hardness and softness of the raw material according to the current classification result includes:

[0012] Determine the raw material product composition according to the current classification result;

[0013] Determine the current hardness and softness of the raw material from the preset mapping relationship according to the raw material product composition;

[0014] The preset mapping relationship is constructed according to different raw material product compositions and the corresponding hardness and softness of different raw material product compositions.

[0015] According to the processing method of the injection molding raw material provided by the present invention, the determining the raw material product composition according to the current classification result includes:

[0016] When the current classification result is a plastic bag, a plastic film or a plastic bottle, determine that the raw material product composition is a polyethylene product;

[0017] When the current classification result is chopsticks, spoons, bowls, plastic buckets or plastic boxes, determine that the raw material product composition is a polypropylene product;

[0018] When the current classification result is a water pipe, a floor covering or an electrical wire insulation layer, determine that the raw material product composition is a polystyrene product;

[0019] When the current classification result is an electrical appliance housing, an automotive component, or a toy, it is determined that the raw material product composition is an acrylonitrile-butadiene-styrene product.

[0020] According to the method for processing injection molding raw materials provided by the present invention, determining the processing priority of the raw materials according to the current hardness and softness and the size of the raw materials in the original image includes:

[0021] P = w H ·H + w S ·S

[0022] Wherein, P is the priority tendency result of the raw materials, H is the current hardness and softness, with a value range of 0 to 1, S is the size ratio, and the size ratio is determined according to the pixel ratio of the size of the raw materials in the original image in the original image. w H is the first weight coefficient, w S is the second weight coefficient, wherein, w H is greater than 0, w S is greater than 0, w H + w S = 1;

[0023] Wherein, when the value of the priority tendency result P of the raw materials is greater than 0.5, it is determined that the processing priority of the raw materials is the first priority, and when the value of the priority tendency result P of the raw materials is less than or equal to 0.5, it is determined that the processing priority of the raw materials is the second priority.

[0024] According to the method for processing injection molding raw materials provided by the present invention, the crushing device corresponding to the first priority is two crushing rollers and a crushing knife that mesh and rotate with each other in the crushing container. The first instruction is used to instruct the conveyance of the raw materials corresponding to the first priority to the crushing container, crush the raw materials corresponding to the first priority with the crushing knife, and process the raw materials corresponding to the first priority with two crushing rollers spaced apart by a first preset distance;

[0025] The crushing device corresponding to the second priority is two crushing rollers and a cutting knife that mesh and rotate with each other in the crushing container. The second instruction is used to instruct the conveyance of the raw materials corresponding to the second priority to the crushing container, cut the raw materials corresponding to the second priority with the cutting knife, and process the raw materials corresponding to the second priority with two crushing rollers spaced apart by a second preset distance;

[0026] The first preset distance is greater than the second preset distance.

[0027] According to the method for processing injection molding raw materials provided by the present invention, the identification of the first current particle size of the raw materials in the crushing container includes:

[0028] Continuously obtain a first image of the crushing container;

[0029] For each first image, input the first image into a preset particle size recognition model to obtain the first current particle size output by the preset particle size recognition model;

[0030] The preset particle size recognition model is determined after being trained according to all particle size sample images and the average particle size corresponding to each particle size sample image.

[0031] According to the method for processing injection molding raw materials provided by the present invention, the identification of the second current particle size of the raw materials in the crushing container includes:

[0032] Continuously obtain a second image of the crushing container;

[0033] For each second image, input the second image into the preset particle size recognition model to obtain the first current particle size output by the preset particle size recognition model.

[0034] According to the method for processing injection molding raw materials provided by the present invention, after obtaining the target injection molding raw materials, the method further includes:

[0035] Use a preset cleaning device to clean the target injection molding raw materials, remove impurities, dirt, and grease, and obtain the cleaned raw materials;

[0036] Use a preset drying device to dry the cleaned raw materials to obtain the dried raw materials, and generate an injection molding instruction, where the injection molding instruction is used to achieve injection molding according to the dried raw materials.

[0037] According to the method for processing injection molding raw materials provided by the present invention, after inputting the original image into the classification prediction model and obtaining the current classification result of the raw materials output by the classification prediction model, the method further includes:

[0038] In the case where the current classification result of the raw materials cannot be determined, determine that the processing priority of the raw materials is the second priority.

[0039] In a second aspect, a system for processing injection molding raw materials is provided, including:

[0040] An input unit, which is used to input the original image of each raw material into a classification prediction model for each original image corresponding to the raw material, and obtain the current classification result of the raw material output by the classification prediction model. The classification prediction model is determined after being trained according to the sample image and the classification result corresponding to the sample image;

[0041] A determining unit, configured to determine the current hardness of the raw material according to the current classification result, and determine the processing priority of the raw material according to the current hardness and the size of the raw material in the original image, where the processing priority includes a first priority and a second priority;

[0042] A generating unit, configured to traverse all the original images and generate a first instruction for instructing to convey the raw material corresponding to the first priority to a crushing container, so as to process the raw material corresponding to the first priority by using a crushing device corresponding to the first priority;

[0043] A first identifying unit, configured to identify the first current particle size of the raw material in the crushing container, and generate a second instruction when it is determined that the first current particle size is smaller than a first preset particle size, where the second instruction is used to instruct to convey the raw material corresponding to the second priority to the crushing container, so as to process the raw material corresponding to the second priority and the raw material corresponding to the first current particle size by using a crushing device corresponding to the second priority;

[0044] A second identifying unit, configured to identify the second current particle size of the raw material in the crushing container, and generate a third instruction when it is determined that the second current particle size is smaller than a second preset particle size, where the third instruction is used to instruct to stop crushing to obtain a target injection molding raw material.

[0045] In the present invention, when processing raw materials with large hardness or large particle size, a way of having a larger gap between the meshing components inside the crusher and cooperating with a crushing knife is adopted, which effectively alleviates the excessive wear of the crushing rollers. When processing raw materials with small hardness or small particle size, a way of having a smaller gap between the meshing components inside the crusher and cooperating with a chopping knife is adopted, which effectively ensures the crushing accuracy of the raw materials; by decomposing the raw materials with large particle size into raw materials with small particle size and then performing crushing, the technical problem that it is difficult to crush the raw materials with large size during the meshing rotation of the two crushing rollers is effectively solved, thereby improving the crushing efficiency; by chopping the plastic bottle with a bulging shape, it is easier to crush the arc surface of the plastic container during the meshing rotation of the two crushing rollers, improving the crushing efficiency of plastic products;

[0046] The present invention realizes the automatic classification of injection molding raw materials through image classification technology, so as to determine the processing priority according to the characteristics and dimensions of the raw materials. The application of particle size recognition technology ensures that the finally obtained injection molding raw materials meet the particle size requirements, improving the stability and consistency of the products. The present invention realizes the automatic, efficient and precise processing of injection molding raw materials by comprehensively applying technical means such as image classification, particle size recognition, priority determination and differential crushing. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0048] Figure 1 is a schematic flow chart of the processing method of injection molding raw materials provided by the present invention;

[0049] Figure 2 is a schematic structural diagram of the processing system of injection molding raw materials provided by the present invention;

[0050] Figure 3 is a schematic structural diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] To make the purpose, technical solutions and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention in conjunction with the drawings in the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the protection scope of the present invention.

[0052] Figure 1 is a schematic flow chart of the processing method of injection molding raw materials provided by the present invention. The processing method of the injection molding raw materials includes:

[0053] Step 101: For the original image corresponding to each raw material, input the original image into the classification prediction model, and obtain the current classification result of the raw material output by the classification prediction model. The classification prediction model is determined after being trained according to the sample image and the classification result corresponding to the sample image;

[0054] Step 102: Determine the current hardness of the raw material according to the current classification result, and determine the processing priority of the raw material according to the current hardness and the size of the raw material in the original image. The processing priority includes the first priority and the second priority;

[0055] Step 103: Traverse all the original images and generate a first instruction, which is used to instruct to convey the raw material corresponding to the first priority to the crushing container, so as to process the raw material corresponding to the first priority by using the crushing device corresponding to the first priority;

[0056] Step 104: Identify the first current particle size of the raw material in the crushing container. When it is determined that the first current particle size is smaller than the first preset particle size, generate a second instruction, which is used to instruct to convey the raw material corresponding to the second priority to the crushing container, so as to process the raw material corresponding to the second priority and the raw material corresponding to the first current particle size by using the crushing device corresponding to the second priority;

[0057] Step 105: Identify the second current particle size of the raw material in the crushing container. When it is determined that the second current particle size is smaller than the second preset particle size, generate a third instruction, which is used to instruct to stop crushing to obtain the target injection molding raw material.

[0058] In step 101, the original image of each raw material in the feed hopper or on the conveyor belt is obtained through a camera or an image scanning device. A classification prediction model is constructed and trained using a deep learning framework. The classification prediction model receives the original image as input and outputs the current classification result of the raw material. The training of the model is based on a large number of sample images with classification labels. The model outputs the current classification result of the raw material. The current classification result can be a plastic bag, a plastic film or a plastic bottle, or it can be an electrical appliance shell, an automotive component or a toy. The recognition efficiency is improved through automatic classification, and manual intervention is reduced.

[0059] In step 102, according to the output result of the classification prediction model, the current hardness of the raw material is determined. For example, hard plastics have a higher hardness, while soft plastics have a lower hardness. Combining the current hardness of the raw material and the size information (such as length, width, height) in the original image, the processing priority of the raw material is determined through a preset rule or algorithm. For example, raw materials with high hardness and large size may have the first priority, while raw materials with low hardness and small size may have the second priority.

[0060] Optionally, the determining the current hardness of the raw material according to the current classification result includes:

[0061] Determine the composition of the raw material product according to the current classification result;

[0062] Determine the current hardness and softness of the raw material from the preset mapping relationship according to the composition of the raw material product.

[0063] The preset mapping relationship is constructed according to different raw material product compositions and the corresponding hardness and softness of different raw material product compositions.

[0064] Optionally, the determination of the raw material product composition according to the current classification result includes:

[0065] When the current classification result is a plastic bag, plastic film or plastic bottle, determine that the raw material product composition is a polyethylene product;

[0066] When the current classification result is chopsticks, spoons, bowls, plastic buckets or plastic boxes, determine that the raw material product composition is a polypropylene product;

[0067] When the current classification result is a water pipe, floor covering or wire insulation layer, determine that the raw material product composition is a polystyrene product;

[0068] When the current classification result is an electrical appliance housing, automotive component or toy, determine that the raw material product composition is an acrylonitrile-butadiene-styrene product.

[0069] Optionally, first use a trained classification prediction model to identify the original image of the raw material and output the current classification result of the raw material. The classification result represents the possible product types of the raw material, such as plastic bags, plastic bottles, plastic buckets, etc. Then, based on these classification results, the system needs to establish a mapping from the classification results to the raw material product composition, which is usually achieved through a preset mapping table or database that records the raw material compositions corresponding to different product types. For example, plastic bags usually correspond to polyethylene (PE), and plastic buckets may correspond to polypropylene (PP) or polyethylene (PE). After determining the raw material product composition, the system needs to determine the current hardness and softness of the raw material based on these compositions. This is also achieved through a preset mapping relationship that records the hardness and softness corresponding to different raw material compositions. For example, polyethylene (PE) may be considered a soft plastic, while polypropylene (PP) may be considered a hard plastic. According to the preset mapping relationship, the system can automatically judge the current hardness and softness of the raw material, providing a basis for subsequent processing steps. For example, the current hardness and softness value of polyethylene is 0.2, that of polypropylene products is 0.3, that of polyvinyl chloride products is 0.6, and that of acrylonitrile-butadiene-styrene products is 0.8.

[0070] Optionally, accurate determination of the raw material composition helps to select appropriate processing technologies and equipment, avoiding resource waste and cost increase caused by incorrect judgment. By using a classification prediction model and a preset mapping relationship, this technical solution can automatically determine the current hardness and softness of the raw material without manual intervention, greatly improving production efficiency. Since the classification prediction model is trained based on a large number of sample images, its recognition accuracy is relatively high. At the same time, the preset mapping relationship is also constructed based on professional knowledge and practical experience, so it can accurately reflect the relationship between the raw material composition and the hardness and softness. The present invention directly maps the classification result to the raw material composition through a preset mapping table, realizing automatic, accurate and efficient raw material processing, providing strong support for subsequent crushing processing.

[0071] Optionally, determining the processing priority of the raw material according to the current hardness and softness and the size of the raw material in the original image includes:

[0072] P = w H ·H + w S ·S

[0073] where P is the priority tendency result of the raw material, H is the current hardness and softness, with a value ranging from 0 to 1, S is the size ratio, and the size ratio is determined according to the pixel ratio of the size of the raw material in the original image in the original image. w H is the first weight coefficient, w S is the second weight coefficient, where w H is greater than 0, w S is greater than 0, w H + w S = 1;

[0074] Among them, when the value of the priority tendency result P of the raw material is greater than 0.5, it is determined that the processing priority of the raw material is the first priority; when the value of the priority tendency result P of the raw material is less than or equal to 0.5, it is determined that the processing priority of the raw material is the second priority.

[0075] Optionally, the priority preference result of the raw material is calculated according to the above formula. The priority preference result of the raw material is a value between 0 and 1, which reflects the preference of the raw material in terms of processing priority. When the value of the priority preference result is greater than 0.5, the processing priority of the raw material is determined as the first priority, which means that the raw material has a higher processing priority and subsequent processing steps need to be carried out preferentially. When the value of the priority preference result is less than or equal to 0.5, the processing priority of the raw material is determined as the second priority, which means that the processing priority of the raw material is relatively low and it can be processed after the raw materials with the first priority are processed. By comprehensively considering the current hardness and softness degree and size proportion of the raw material, the present invention determines the processing priority, making the processing process more scientific and reasonable. Both the hardness and softness degree and the size are important influencing factors affecting the subsequent crushing efficiency of the raw material.

[0076] In step 103, the original images of all raw materials are traversed. According to the processing priority, a first instruction is generated. The first instruction is sent to the conveying device through the control system to instruct it to convey the raw material corresponding to the first priority to the crushing container. The crushing device corresponding to the first priority is configured according to the characteristics of the raw material to ensure the best crushing effect.

[0077] Optionally, the crushing device corresponding to the first priority is two crushing rolls and a crushing knife that mesh and rotate with each other in the crushing container. The first instruction is used to instruct the conveyance of the raw material corresponding to the first priority to the crushing container, and the raw material corresponding to the first priority is crushed by the crushing knife, and the raw material corresponding to the first priority is processed by two crushing rolls spaced at a first preset distance.

[0078] Optionally, the crushing device corresponding to the first priority in the present invention is two crushing rolls with parameters set at a first preset distance. Since the raw material corresponding to the first priority is plastic with a relatively large particle size and high hardness, the present invention can first crush the raw material corresponding to the first priority with the crushing knife to make it into a raw material with a relatively smaller particle size, so that the raw material corresponding to the first priority can more easily enter the crushing rolls, avoiding being stuck due to too large a size or being unable to enter the crushing rolls due to too large a size. After being processed by the crushing knife, a rough crushing method with a relatively large gap between the meshing components inside the crusher is adopted for rough crushing, effectively alleviating the excessive wear of the crushing rolls and improving the service life of the crushing rolls.

[0079] In step 104, when the present invention determines that the first current granularity is smaller than the first preset granularity, a second instruction is generated. The second instruction instructs to convey the raw material corresponding to the second priority to the crushing container and process it using the crushing device corresponding to the second priority. The crushing device corresponding to the second priority is two crushing rollers and a cutting knife that are meshed and rotated with each other in the crushing container. The second instruction is used to instruct to convey the raw material corresponding to the second priority to the crushing container, cut the raw material corresponding to the second priority using the cutting knife, and process the raw material corresponding to the second priority using the two crushing rollers with a second preset distance therebetween. The first preset distance is greater than the second preset distance.

[0080] Optionally, at this time, the raw material corresponding to the first priority has been processed to be below the first preset granularity. At this time, the raw material corresponding to the first current granularity can be subjected to a more refined crushing process. Also, since the raw material corresponding to the second priority is plastic with a smaller and softer granularity, the present invention can simultaneously perform fine processing on the raw material corresponding to the first current granularity and the raw material corresponding to the second priority. Specifically, by adjusting the first preset distance to the second preset distance, the present invention can also first cut the raw material corresponding to the second priority using the cutting knife. For example, when cutting a plastic bottle with a bulging shape, the cut plastic bottle will no longer maintain its bulging shape, making it easier to crush the arc surface of the plastic container during the meshing rotation of the two crushing rollers, improving the crushing efficiency of plastic products. After the cutting knife finishes processing, a crushing method with a smaller gap between the meshing components inside the crusher is adopted, effectively ensuring the crushing accuracy of the raw material.

[0081] Optionally, the identifying the first current granularity of the raw material in the crushing container includes:

[0082] Continuously obtain the first image in the crushing container;

[0083] For each first image, input the first image into a preset granularity recognition model to obtain the first current granularity output by the preset granularity recognition model;

[0084] The preset granularity recognition model is determined after being trained according to all granularity sample images and the average granularity corresponding to each granularity sample image.

[0085] Optionally, the present invention uses a camera or other image acquisition device to capture images of the raw materials in the crushing container in real time or at regular intervals. The image acquisition device should have good lighting conditions, clear imaging quality, and appropriate shooting angles to ensure that the captured images can accurately reflect the particle size of the raw materials. The preset particle size recognition model is a pre-trained machine learning model used to identify the particle size of the raw materials in the input image. It can be a deep learning model (such as a convolutional neural network CNN), a support vector machine (SVM), or other models suitable for image classification. The preset particle size recognition model processes the input image and outputs the first current particle size of the raw materials. The first current particle size can be a specific value (such as the particle diameter size) or a classification label (such as fine particles, medium particles, coarse particles, etc.). According to the output of the model, the particle size of the raw materials in the crushing container can be understood in real time.

[0086] Optionally, the preset particle size recognition model is determined after being trained based on all particle size sample images and the average particle size corresponding to each particle size sample image. This means that during the model training stage, it is necessary to collect a large number of raw material images with different particle sizes and label the average particle size corresponding to each image. Use these labeled image data to train the model so that the model can learn the mapping relationship between the raw material particle size and the image features. After training, the model can be used to identify the particle size of new images. By using a machine learning model for particle size recognition, the accuracy and precision of recognition can be significantly improved. Compared with traditional manual recognition methods, the machine learning model can process a large amount of image data more quickly and automatically learn the complex relationship between the raw material particle size and the image features. Real-time monitoring of the raw material particle size helps to adjust the parameters and operation processes of the crushing equipment in a timely manner. For example, in the case where it is determined that the first current particle size is smaller than the first preset particle size, it is instructed to convey the raw materials corresponding to the second priority to the crushing container and process them using the crushing device corresponding to the second priority. The parameters such as the rotation speed and pressure of the crushing equipment can also be adjusted to achieve fine processing of the raw materials.

[0087] In step 105, the present invention uses the same preset particle size recognition model to identify the current particle size of the raw materials in the crushing container. In the case where it is determined that the second current particle size is smaller than the second preset particle size, a third instruction is generated, and the third instruction instructs to stop the crushing process to obtain the target injection molding raw materials. Optionally, the identifying the second current particle size of the raw materials in the crushing container includes: continuously acquiring the second image in the crushing container; for each second image, inputting the second image into the preset particle size recognition model to obtain the first current particle size output by the preset particle size recognition model.

[0088] Optionally, after obtaining the target injection molding raw materials, the method further includes:

[0089] Clean the target injection molding raw materials using a preset cleaning device, and after removing impurities, dirt, and grease, obtain the cleaned raw materials;

[0090] Dry the cleaned raw materials using a preset drying device to obtain the dried raw materials, and generate an injection molding instruction, where the injection molding instruction is used to achieve injection molding according to the dried raw materials.

[0091] Optionally, after obtaining the target injection molding raw materials, place the target injection molding raw materials in a cleaning device. The type, concentration, temperature, and cleaning time of the cleaning agent can be adjusted according to the pollution degree and material of the raw materials. Start the cleaning device to remove impurities, dirt, and grease on the surface of the raw materials through physical or chemical effects. After cleaning, rinse the raw materials with clean water or a solvent to remove the residual cleaning agent. Place the cleaned raw materials in the tray of the drying device. The temperature, wind speed, and drying time of the drying device can be adjusted according to the material and thickness of the raw materials. Start the drying device to evaporate the moisture and solvent on the surface of the raw materials through hot air, infrared radiation, or vacuum dehumidification, etc. After drying, put the dried raw materials into the hopper of the injection molding machine, start the injection molding machine, and perform injection molding according to the injection molding instruction.

[0092] Optionally, after inputting the original image into the classification prediction model and obtaining the current classification result of the raw materials output by the classification prediction model, the method further includes:

[0093] In the case where the current classification result of the raw materials cannot be determined, determine that the processing priority of the raw materials is the second priority.

[0094] Optionally, check the result output by the classification prediction model. If the confidence level of the classification result output by the model is lower than a preset threshold, or the output result is of the "unknown" category, it is considered that the current classification result of the raw materials cannot be determined. In the case where the classification result cannot be determined, according to a preset rule or strategy, set the processing priority of the raw materials to the second priority. By setting the second priority, it can be ensured that the raw materials can be properly processed in the case of uncertain classification results, which helps to improve the flexibility and accuracy of raw material processing, and at the same time reduces the potential risks caused by classification errors.

[0095] In the present invention, when processing raw materials with relatively high hardness or relatively large particle size, a method of having a relatively large gap between the meshing components inside the crusher and cooperating with crushing knives is adopted, effectively alleviating the excessive wear of the crushing rollers. When processing raw materials with relatively low hardness or relatively small particle size, a method of having a relatively small gap between the meshing components inside the crusher and cooperating with chopping knives is adopted, effectively ensuring the crushing accuracy of the raw materials. By decomposing raw materials with large particle size into raw materials with small particle size before crushing, the technical problem that it is difficult to crush raw materials with relatively large size during the meshing rotation of the two crushing rollers is effectively solved, thereby improving the crushing efficiency. By chopping the plastic bottles with a bulging shape, it is easier to crush the arc surface of the plastic container during the meshing rotation of the two crushing rollers, improving the crushing efficiency of plastic products.

[0096] The present invention realizes the automatic classification of injection molding raw materials through image classification technology, so as to determine the processing priority according to the characteristics and size of the raw materials. The application of particle size recognition technology ensures that the finally obtained injection molding raw materials meet the particle size requirements, improving the stability and consistency of the products. The present invention realizes the automatic, efficient and precise processing of injection molding raw materials by comprehensively applying technical means such as image classification, particle size recognition, priority determination and differential crushing.

[0097] Figure 2 It is a schematic structural diagram of a processing system for injection molding raw materials provided by the present invention. The processing system for injection molding raw materials includes an input unit 1. The input unit is used to input the original image corresponding to each raw material into the classification prediction model and obtain the current classification result of the raw material output by the classification prediction model. The classification prediction model is determined after being trained according to the sample image and the classification result corresponding to the sample image. The working principle of the input unit 1 can refer to the foregoing step 101 and will not be elaborated here.

[0098] The processing system for injection molding raw materials further includes a determination unit 2. The determination unit is used to determine the current hardness of the raw material according to the current classification result, and determine the processing priority of the raw material according to the current hardness and the size of the raw material in the original image. The processing priority includes a first priority and a second priority. The working principle of the determination unit 2 can refer to the foregoing step 102 and will not be elaborated here.

[0099] The processing system for injection molding raw materials further includes a generation unit 3. The generation unit is used to traverse all the original images and generate a first instruction command. The first instruction command is used to instruct to convey the raw materials corresponding to the first priority to the crushing container to process the raw materials corresponding to the first priority by using the crushing device corresponding to the first priority. The working principle of the generation unit 3 can refer to the foregoing step 103 and will not be elaborated here.

[0100] The processing system for the injection molding raw materials further includes a first identification unit 4. The first identification unit is used to identify the first current particle size of the raw materials in the crushing container. When it is determined that the first current particle size is smaller than the first preset particle size, a second instruction is generated. The second instruction is used to instruct the conveyance of the raw materials corresponding to the second priority level to the crushing container, so as to use the crushing device corresponding to the second priority level to process the raw materials corresponding to the second priority level. The working principle of the first identification unit 4 can refer to the aforementioned step 104 and will not be elaborated here.

[0101] The processing system for the injection molding raw materials further includes a second identification unit 5. The second identification unit is used to identify the second current particle size of the raw materials in the crushing container. When it is determined that the second current particle size is smaller than the second preset particle size, a third instruction is generated. The third instruction is used to instruct the stopping of crushing to obtain the target injection molding raw materials. The working principle of the second identification unit 5 can refer to the aforementioned step 105 and will not be elaborated here.

[0102] In the present invention, when processing raw materials with a relatively large hardness or a relatively large particle size, a way of having a relatively large gap between the meshing components inside the crusher and cooperating with the crushing knives is adopted, effectively alleviating the excessive wear of the crushing rollers. When processing raw materials with a relatively small hardness or a relatively small particle size, a way of having a relatively small gap between the meshing components inside the crusher and cooperating with the chopping knives is adopted, effectively ensuring the crushing accuracy of the raw materials; by decomposing the raw materials with a large particle size into raw materials with a small particle size and then performing crushing, the technical problem that it is difficult to crush the raw materials with a relatively large size during the meshing rotation of the two crushing rollers is effectively solved, thereby improving the crushing efficiency; by performing chopping treatment on the plastic bottles with a bulging shape of their own, it is easier to crush the arc surface of the plastic container during the meshing rotation of the two crushing rollers, improving the crushing efficiency of plastic products;

[0103] The present invention realizes the automatic classification of injection molding raw materials through image classification technology, so as to be able to determine the processing priority according to the characteristics and dimensions of the raw materials. The application of the particle size identification technology ensures that the finally obtained injection molding raw materials meet the particle size requirements, improving the stability and consistency of the products; the present invention realizes the automatic, efficient and precise processing of injection molding raw materials through the comprehensive application of technologies such as image classification, particle size identification, priority determination and differential crushing.

[0104] Figure 3 is a schematic structural diagram of the electronic device provided by the present invention. As Figure 3As shown, the electronic device may include: a processor 110, a communications interface 120, a memory 130, and a communication bus 140. Among them, the processor 110, the communications interface 120, and the memory 130 complete communication with each other through the communication bus 140. The processor 110 may call the logical instructions in the memory 130 to execute a method for processing injection molding raw materials. The method includes: for each original image corresponding to a raw material, inputting the original image into a classification prediction model, and obtaining the current classification result of the raw material output by the classification prediction model. The classification prediction model is determined after being trained according to sample images and the classification results corresponding to the sample images; determining the current hardness and softness of the raw material according to the current classification result, and determining the processing priority of the raw material according to the current hardness and softness and the size of the raw material in the original image. The processing priority includes a first priority and a second priority; traversing all the original images to generate a first instruction instruction, where the first instruction instruction is used to instruct to convey the raw material corresponding to the first priority to a crushing container to process the raw material corresponding to the first priority by using a crushing device corresponding to the first priority; identifying the first current particle size of the raw material in the crushing container, and in the case of determining that the first current particle size is smaller than a first preset particle size, generating a second instruction instruction, where the second instruction instruction is used to instruct to convey the raw material corresponding to the second priority to the crushing container to process the raw material corresponding to the second priority and the raw material corresponding to the first current particle size by using a crushing device corresponding to the second priority; identifying the second current particle size of the raw material in the crushing container, and in the case of determining that the second current particle size is smaller than a second preset particle size, generating a third instruction instruction, where the third instruction instruction is used to instruct to stop crushing to obtain the target injection molding raw material.

[0105] In addition, the logical instructions in the above-mentioned memory 130 may be implemented in the form of software function units. When used as an independent product or sold, they may be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or a part of this technical solution may be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc that can store program codes.

[0106] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute a processing method of an injection molding raw material provided by each of the above methods. The method includes: for each original image corresponding to a raw material, input the original image into a classification prediction model, and obtain the current classification result of the raw material output by the classification prediction model. The classification prediction model is determined after being trained according to sample images and the classification results corresponding to the sample images; determine the current hardness and softness of the raw material according to the current classification result, and determine the processing priority of the raw material according to the current hardness and softness and the size of the raw material in the original image. The processing priority includes a first priority and a second priority; traverse all the original images and generate a first instruction instruction, which is used to instruct to convey the raw material corresponding to the first priority to a crushing container to process the raw material corresponding to the first priority by using a crushing device corresponding to the first priority; identify the first current particle size of the raw material in the crushing container. When it is determined that the first current particle size is less than a first preset particle size, generate a second instruction instruction, which is used to instruct to convey the raw material corresponding to the second priority to the crushing container to process the raw material corresponding to the second priority and the raw material corresponding to the first current particle size by using a crushing device corresponding to the second priority; identify the second current particle size of the raw material in the crushing container. When it is determined that the second current particle size is less than a second preset particle size, generate a third instruction instruction, which is used to instruct to stop crushing to obtain a target injection molding raw material.

[0107] In another aspect, the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements a processing method for injection molding raw materials provided by the above-mentioned various methods. The method includes: for the original image corresponding to each raw material, input the original image into a classification prediction model, and obtain the current classification result of the raw material output by the classification prediction model. The classification prediction model is determined after being trained according to sample images and the classification results corresponding to the sample images; determine the current hardness and softness of the raw material according to the current classification result, and determine the processing priority of the raw material according to the current hardness and softness and the size of the raw material in the original image. The processing priority includes a first priority and a second priority; traverse all the original images and generate a first instruction instruction, which is used to instruct to convey the raw material corresponding to the first priority to a crushing container to process the raw material corresponding to the first priority by using a crushing device corresponding to the first priority; identify the first current particle size of the raw material in the crushing container. When it is determined that the first current particle size is smaller than a first preset particle size, generate a second instruction instruction, which is used to instruct to convey the raw material corresponding to the second priority to the crushing container to process the raw material corresponding to the second priority and the raw material corresponding to the first current particle size by using a crushing device corresponding to the second priority; identify the second current particle size of the raw material in the crushing container. When it is determined that the second current particle size is smaller than a second preset particle size, generate a third instruction instruction, which is used to instruct to stop crushing to obtain the target injection molding raw material.

[0108] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or 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. A person of ordinary skill in the art can understand and implement it without creative work.

[0109] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solutions, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disc, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for processing injection molding raw materials, characterized in that, Including: For the original image corresponding to each raw material, input the original image into the classification prediction model to obtain the current classification result of the raw material output by the classification prediction model. The classification prediction model is determined after being trained according to the sample image and the classification result corresponding to the sample image; Determine the current hardness and softness of the raw material according to the current classification result, and determine the processing priority of the raw material according to the current hardness and softness and the size of the raw material in the original image. The processing priority includes the first priority and the second priority; Traverse all the original images to generate a first instruction, which is used to instruct to convey the raw material corresponding to the first priority to the crushing container to process the raw material corresponding to the first priority by using the crushing device corresponding to the first priority; Identify the first current particle size of the raw material in the crushing container. When it is determined that the first current particle size is smaller than the first preset particle size, generate a second instruction, which is used to instruct to convey the raw material corresponding to the second priority to the crushing container to process the raw material corresponding to the second priority and the raw material corresponding to the first current particle size; Identify the second current particle size of the raw material in the crushing container. When it is determined that the second current particle size is smaller than the second preset particle size, generate a third instruction, which is used to instruct to stop crushing to obtain the target injection molding raw material.

2. The processing method of the injection molding raw material according to claim 1, characterized in that, The determining the current hardness and softness of the raw material according to the current classification result includes: Determine the raw material product composition according to the current classification result; Determine the current hardness and softness of the raw material from the preset mapping relationship according to the raw material product composition; The preset mapping relationship is constructed according to different raw material product compositions and the hardness and softness corresponding to different raw material product compositions.

3. The processing method of the injection molding raw material according to claim 2, characterized in that, The determining the raw material product composition according to the current classification result includes: When the current classification result is plastic bag, plastic film or plastic bottle, determine that the raw material product composition is polyethylene product; When the current classification result is chopsticks, spoon, bowl, plastic bucket or plastic box, determine that the raw material product composition is polypropylene product; When the current classification result is water pipe, floor leather or wire insulation layer, determine that the raw material product composition is polystyrene product; When the current classification result is electrical appliance shell, automobile part or toy, determine that the raw material product composition is acrylonitrile-butadiene-styrene product.

4. The processing method of the injection molding raw material according to claim 1, characterized in that, The determining the processing priority of the raw material according to the current hardness and softness and the size of the raw material in the original image includes: P = w H ·H + w S ·S Among them, P is the priority tendency result of the raw material, H is the current hardness and softness, with a value ranging from 0 to 1, S is the size ratio, and the size ratio is determined according to the pixel ratio of the size of the raw material in the original image in the original image, w H is the first weight coefficient, w S is the second weight coefficient, where, w H is greater than 0, w S is greater than 0, w H + w S = 1; Among them, when the value of the priority tendency result P of the raw material is greater than 0.5, determine that the processing priority of the raw material is the first priority. When the value of the priority tendency result P of the raw material is less than or equal to 0.5, determine that the processing priority of the raw material is the second priority.

5. The processing method of the injection molding raw material according to claim 1, characterized in that The crushing device corresponding to the first priority is two crushing rollers and a crushing knife that mesh and rotate with each other in the crushing container. The first instruction is used to instruct the conveyance of the raw material corresponding to the first priority to the crushing container, to crush the raw material corresponding to the first priority by using the crushing knife, and to process the raw material corresponding to the first priority by using the two crushing rollers spaced apart by a first preset distance; The crushing device corresponding to the second priority is two crushing rollers and a cutting knife that mesh and rotate with each other in the crushing container. The second instruction is used to instruct the conveyance of the raw material corresponding to the second priority to the crushing container, to cut the raw material corresponding to the second priority by using the cutting knife, and to process the raw material corresponding to the second priority and the raw material corresponding to the first current particle size by using the two crushing rollers spaced apart by a second preset distance; The first preset distance is greater than the second preset distance.

6. The processing method of the injection molding raw material according to claim 1, characterized in that The identification of the first current particle size of the raw material in the crushing container includes: Continuously acquiring a first image in the crushing container; For each first image, inputting the first image into a preset particle size identification model to obtain the first current particle size output by the preset particle size identification model; The preset particle size identification model is determined after being trained according to all particle size sample images and the average particle size corresponding to each particle size sample image.

7. The processing method of the injection molding raw material according to claim 6, characterized in that, The identification of the second current particle size of the raw material in the crushing container includes: Continuously acquiring a second image in the crushing container; For each second image, inputting the second image into the preset particle size identification model to obtain the first current particle size output by the preset particle size identification model.

8. The processing method of the injection molding raw material according to claim 1, characterized in that After obtaining the target injection molding raw material, the method further includes: Using a preset cleaning device to clean the target injection molding raw material, removing impurities, dirt, and grease to obtain the cleaned raw material; Using a preset drying device to perform a drying process on the cleaned raw material to obtain the dried raw material, and generating an injection molding instruction, where the injection molding instruction is used to achieve injection molding according to the dried raw material.

9. The processing method of the injection molding raw material according to claim 1, characterized in that, After inputting the original image into the classification prediction model and obtaining the current classification result of the raw material output by the classification prediction model, the method further includes: In the case where the current classification result of the raw material cannot be determined, determining the processing priority of the raw material as the second priority.

10. A processing system for injection molding raw materials, characterized in that, Including: An input unit, where the input unit is used to input, for each original image corresponding to a raw material, the original image into the classification prediction model to obtain the current classification result of the raw material output by the classification prediction model, and the classification prediction model is determined after being trained according to the sample image and the classification result corresponding to the sample image; A determination unit, where the determination unit is used to determine the current hardness and softness of the raw material according to the current classification result, and to determine the processing priority of the raw material according to the current hardness and softness and the size of the raw material in the original image, and the processing priority includes a first priority and a second priority; A generating unit, which is used to traverse all the original images and generate a first indication instruction for indicating the conveyance of raw materials corresponding to the first priority to a crushing container so as to process the raw materials corresponding to the first priority by using a crushing device corresponding to the first priority; A first recognition unit, which is used to recognize the first current particle size of the raw materials in the crushing container and generate a second indication instruction when it is determined that the first current particle size is smaller than a first preset particle size. The second indication instruction is used to indicate the conveyance of raw materials corresponding to the second priority to the crushing container so as to process the raw materials corresponding to the second priority and the raw materials corresponding to the first current particle size by using a crushing device corresponding to the second priority; A second recognition unit, which is used to recognize the second current particle size of the raw materials in the crushing container and generate a third indication instruction when it is determined that the second current particle size is smaller than a second preset particle size. The third indication instruction is used to indicate the stoppage of crushing to obtain target injection molding raw materials.

Images