Material intelligent processing method and system based on image recognition

Abstract

The invention discloses an intelligent material processing method and system based on image recognition, and relates to the technical field of image recognition, and the method comprises the steps: configuring a qualified material data set; capturing a material image on the conveying belt, and performing comparative analysis on the material image and a preset material image library; establishing an unqualified material data set; when the unqualified material is non-metal and the size is greater than a preset threshold value, the propelling device is driven to push the non-metal material into the crushing device; when the unqualified materials are metal, an iron removal mechanism is called to grab the metal materials away from the conveying belt; and the intelligent control system sends out an alarm signal when detecting equipment abnormity. The technical problems that due to the fact that existing materials are too large in size, a feeding port of production equipment is blocked, production efficiency is affected, meanwhile, metal impurities enter the production equipment, the equipment is damaged, and the service life of the equipment is shortened are solved, unqualified materials are intelligently recognized, and the purposes of reducing loss of the production equipment and improving production efficiency are achieved. And the technical effect of intelligent material identification and processing is improved.

Description

Technical Field

[0001] This application relates to the field of image recognition technology, and particularly to an intelligent material processing method and system based on image recognition. Background Art

[0002] With the rapid development of modern industrial production, the scale of the cement processing industry has gradually expanded. The pretreatment of materials required for cement production, such as limestone, clay, iron powder, etc., is particularly important. Materials with too large particle size will cause blockage of the feeding port during the feeding process. If metal blocks enter the production system, it will wear the equipment. The problems of material identification and pretreatment are becoming increasingly prominent. If unqualified materials are not properly processed, it will not only cause blockage of the feeding port and affect production efficiency, but also the unqualified metal materials will cause equipment wear. How to identify materials efficiently and accurately has become an important problem that needs to be solved in the current cement production industry. The intelligent identification and processing of unqualified materials are particularly important. However, the traditional identification methods of unqualified materials are mostly manual monitoring and manual processing, which are cumbersome and inaccurate, and it is difficult to meet the requirements of modern cement processing industry for high-efficiency and precise control.

[0003] Therefore, in the related technologies of identifying and processing unqualified materials in the current stage of cement processing, there are technical problems that it is difficult to accurately identify unqualified materials during the material feeding process, and it is impossible to dynamically process the identified unqualified materials according to their actual situations, thus resulting in the inability to accurately identify and process unqualified materials. Summary of the Invention

[0004] The purpose of this application is to provide an intelligent material processing method and system based on image recognition, so as to solve the technical problems that existing materials often cause blockage of the feeding port of production equipment due to their too large volume, affecting production efficiency. At the same time, metal impurities entering the production equipment will damage the equipment and shorten the service life of the equipment.

[0005] In view of the above technical problems, this application provides an intelligent material processing method and system based on image recognition.

[0006] In the first aspect of the embodiments of this application, an intelligent material processing method based on image recognition is provided. The method includes: Establish a digital connection between the intelligent control system and the image acquisition device, the propulsion device, and the iron removal mechanism; Configure a qualified material data set, which includes a material picture library, material types, material sizes and thresholds, and material numbers; By capturing the material image on the conveyor belt in real time and comparing and analyzing it with the preset material picture library, identify the material type and material size, where the material size includes the length, width, and height of the material; Compare the recognized material types and sizes with the types and size thresholds of preset qualified materials to identify materials with sizes exceeding the thresholds, and establish a dataset of unqualified materials. The unqualified materials include non-metallic materials with sizes exceeding the thresholds and materials with the type of metal. The dataset of unqualified materials includes the numbers, types, and sizes of unqualified materials. Calculate the real-time position of the unqualified material during transportation on the conveyor belt based on the moving speed of the material on the conveyor belt and the recognized position information of the unqualified material. Obtain the dataset of unqualified materials. When the type of the unqualified material is non-metallic, compare the size of the non-metallic material. When the size is greater than the preset threshold, read the real-time position of the non-metallic material, and drive the propulsion device to push the non-metallic material into the crushing device according to the real-time position. The crushing device is configured on the side of the conveyor belt to crush the material and transport the crushed material back to the starting position of the conveyor belt. When it is determined that the type of the unqualified material is metal, obtain the type and size of the metal material, obtain the real-time position of the metal material, and call the iron removal mechanism to grab the metal material off the conveyor belt according to the preset grabbing scheme. Monitor the operating states of the image acquisition device, the propulsion device, and the iron removal mechanism in real time. When the intelligent control system detects equipment abnormalities or reaches the preset alarm threshold, immediately send out an alarm signal.

[0007] Further, the method of recognizing the material type and size by capturing the material image on the conveyor belt in real time and comparing and analyzing it with the preset material picture library includes: Use a high-resolution industrial camera or camera as the image acquisition device and install it above or on the side of the conveyor belt. Perform denoising, contrast enhancement, and grayscale preprocessing on the collected original image. Use the edge detection algorithm to identify the position and shape of the material in the image. Extract the features of the material. The features include the type and size of the material. The types include gypsum and metal, and the sizes include length, width, and height. Match the extracted material type with the preset material picture library and use a convolutional neural network to identify the material type. Calculate the volume of the material according to the extracted size.

[0008] Further, the method of comparing the recognized material types and sizes with the types and size thresholds of preset qualified materials to identify materials with sizes exceeding the thresholds and establishing a dataset of unqualified materials. The unqualified materials include non-metallic materials with sizes exceeding the thresholds and materials with the type of metal. The dataset of unqualified materials includes the numbers, types, and sizes of unqualified materials, includes: Set an upper threshold for the size based on each qualified material type; Compare the identified material type with the preset qualified material types. If the material type is metal, directly mark this material as unqualified; If the material type is non - metal, compare the size of the non - metal material with the preset size threshold. If the material size exceeds the threshold range, mark it as unqualified; Summarize the information of all materials marked as unqualified. The unqualified material information includes the material number, type, and size; Use the summarized unqualified material information to construct an unqualified material data set. The unqualified material data set contains the unique number, type, and size information of each unqualified material.

[0009] Furthermore, calculating the real - time position of the unqualified material transported on the conveyor belt according to the material moving speed on the conveyor belt and the identified position information of the unqualified material includes: Obtain the moving speed of the material through a belt speed sensor; When the material is identified as unqualified, the intelligent control system records the specific position of the material on the belt and the moment when the material is identified. The specific position is the distance relative to the starting point of the belt; The calculation formula for the real - time position of the unqualified material on the belt is as follows: D d = D i + V×(T c -T i ) ; Where, D d is the real - time position of the unqualified material on the belt, D i is the initial position of the material, T c is the current moment, T i is the moment when the material is identified, V is the moving speed of the material, T c -T i is the time difference from the moment when the material is identified to the current moment, V×(T c -T i ) is to calculate the moving distance of the material on the belt within the time difference by the product of the material moving speed and the time difference, D i + V×(T c -T i ) is to calculate the real - time position of the unqualified material on the belt by adding its initial position and the moving distance.

[0010] Further, when it is determined that the type of unqualified material is metal, obtain the type and size of the metal material, obtain the real-time position of the metal material, and call the iron removal mechanism to grab the metal material away from the conveyor belt according to a preset grabbing scheme, including: According to the type and size of the metal material, preset the grabbing scheme of the iron removal mechanism; According to the real-time obtained material position, the control system issues an instruction to start the iron removal mechanism for grabbing operation.

[0011] Further, the presetting of the grabbing scheme of the iron removal mechanism according to the type and size of the metal material includes: According to the size and shape of the metal material, configure the grabbing scheme and establish a grabbing scheme library; When it is determined that the type of the metal to be grabbed is a magnetic metal, the iron removal mechanism uses the magnetic field generated by the permanent magnet to absorb the magnetic metal and grabs the magnetic metal away from the conveyor belt; When the type of the metal to be grabbed is a non-magnetic metal, the iron removal mechanism uses the clamping mechanism to clamp and grab the non-magnetic metal away from the conveyor belt, and the clamping mechanism realizes the grabbing of the material through the opening and closing of the mechanical claw.

[0012] In the second aspect of the embodiments of the present application, a material intelligent processing system based on image recognition is provided, and the system includes: A digital connection establishment module, which is used to establish digital connections between the intelligent control system and the image acquisition device, the propulsion device, and the iron removal mechanism; A qualified material data set configuration module, which is used to configure a qualified material data set, and the qualified material data set includes a material picture library, material types, material sizes and thresholds, and material numbers; An identification module, which is used to capture the material image on the conveyor belt in real time and perform comparative analysis with the preset material picture library to identify the material type and material size, and the material size includes the length, width, and height of the material; A data set establishment module, which is used to compare the identified material type and material size with the type and size thresholds of the preset qualified materials, identify the materials with sizes exceeding the thresholds, and establish an unqualified material data set. The unqualified materials include those with non-metal material sizes exceeding the thresholds and materials with the material type being metal. The unqualified material data set includes the numbers, types, and sizes of the unqualified materials; A real-time position calculation module, which is used to calculate the real-time position of the unqualified material transported on the conveyor belt according to the material moving speed on the conveyor belt and the position information of the identified unqualified material; Material propulsion module, which is used to obtain a dataset of unqualified materials. When it is determined that the type of unqualified material is non-metal, the sizes of the non-metal materials are compared. When the size is greater than a preset threshold, the real-time positions of the non-metal materials are read, and a propulsion device is driven to push the non-metal materials into a crushing device according to the real-time positions. The crushing device is arranged on the side of a conveyor belt to crush the materials and transport the crushed materials back to the starting position of the conveyor belt; Metal grasping module, which is used to determine when the type of unqualified material is metal, obtain the type and size of the metal material, obtain the real-time position of the metal material, and call a magnetic separation mechanism to grasp the metal material away from the conveyor belt according to a preset grasping scheme; Intelligent monitoring module, which is used to monitor the operating states of an image acquisition device, a propulsion device and a magnetic separation mechanism in real time. When the intelligent control system detects equipment anomalies or reaches a preset alarm threshold, an alarm signal is immediately issued.

[0013] One or more technical solutions provided in this application have at least the following technical effects or advantages: Establish digital connections between the intelligent control system, the image acquisition device, the propulsion device, and the iron removal mechanism; configure a qualified material dataset, which includes a material picture library, material types, material dimensions and thresholds, and material numbers; by capturing the material images on the conveyor belt in real time and comparing and analyzing them with the preset material picture library, identify the material types and material dimensions, where the material dimensions include the length, width, and height of the material; compare the identified material types and material dimensions with the types and dimension thresholds of the preset qualified materials to identify materials with dimensions exceeding the thresholds, and establish an unqualified material dataset. The unqualified materials include non-metal materials with dimensions exceeding the thresholds and materials with the material type being metal. The unqualified material dataset includes the numbers, types, and dimensions of the unqualified materials; calculate the real-time position of the unqualified materials transported on the conveyor belt based on the moving speed of the materials on the conveyor belt and the identified position information of the unqualified materials; obtain the unqualified material dataset, and when the type of the unqualified material is non-metal, compare the dimensions of the non-metal material. When the dimension is greater than the preset threshold, read the real-time position of the non-metal material, and drive the propulsion device to push the non-metal material into the crushing device according to the real-time position. The crushing device is configured on the side of the conveyor belt to crush the material and transport the crushed material back to the starting position of the conveyor belt; when the type of the unqualified material is metal, obtain the type and dimensions of the metal material, obtain the real-time position of the metal material, and call the iron removal mechanism to grab the metal material off the conveyor belt according to the preset grabbing scheme; monitor the operating states of the image acquisition device, the propulsion device, and the iron removal mechanism in real time. When the intelligent control system detects equipment anomalies or reaches the preset alarm threshold, immediately send an alarm signal. This solves the technical problems in the prior art that materials cause blockages at the inlets of production equipment due to their excessive volume, affecting production efficiency. At the same time, metal impurities enter the production equipment and cause damage to the equipment, shortening the service life of the equipment, realizing the intelligent identification of unqualified materials, and achieving the technical effects of reducing the loss of production equipment and improving the intelligent identification and processing of materials.

[0014] The above description is only an overview of the technical solution of this application. In order to be able to more clearly clarify the technical means of this application, it can be implemented in accordance with the content of the description. And in order to make the above and other purposes, features, and advantages of this application more obvious and understandable, the following specifically illustrates the embodiments of this application. Brief Description of the Drawings

[0015] To more clearly illustrate the technical solutions of the embodiments of the present disclosure, the accompanying drawings of the embodiments of the present disclosure will be briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the operations described above or below do not necessarily need to be executed precisely in order. On the contrary, according to the need, various steps can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0016] Figure 1 It is a schematic flowchart of the intelligent material processing method based on image recognition provided by the embodiments of the present application; Figure 2 It is a schematic structural diagram of the intelligent material processing system based on image recognition provided by the embodiments of the present application.

[0017] Explanation of reference numerals: Digital connection establishment module 10, qualified material dataset configuration module 20, recognition module 30, dataset establishment module 40, real-time position calculation module 50, material propulsion module 60, metal grasping module 70, intelligent monitoring module 80. Detailed implementation manners

[0018] By providing an intelligent material processing method and system based on image recognition, the present application solves the technical problems that the existing materials cause blockage of the feeding port of the production equipment due to their too large volume, affecting production efficiency. At the same time, metal impurities enter the production equipment, causing damage to the equipment and shortening the service life of the equipment.

[0019] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the scope of protection of the present application.

[0020] It should be noted that the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0021] Embodiment 1, as Figure 1 shown, the present application provides an intelligent material processing method based on image recognition, wherein the method includes: Establish a digital connection between the intelligent control system and the image acquisition device, the propulsion device, and the iron removal mechanism; Specifically, as the core of the entire system, the intelligent control system is responsible for coordination, monitoring, and management, and has data processing capabilities, logical judgment capabilities, and control capabilities for external devices. It can receive data from different devices in real time, process it according to preset algorithms and rules, and then issue corresponding instructions. Digital connections are established between the intelligent control system, the image acquisition device, the propulsion device, and the iron removal mechanism to achieve efficient data interaction, precise control, and coordinated operation of the overall system. Through this connection, the intelligent control system can obtain the image data of the image acquisition device for analysis and judgment, control the propulsion device, and manage the iron removal mechanism to perform metal removal operations.

[0022] Configure a qualified material dataset, where the qualified material dataset includes a material picture library, material types, material dimensions and thresholds, and material numbers. Specifically, configure a dataset of qualified materials, which includes a material picture library, material types, material dimensions and thresholds, and material numbers. Among them, the material pictures in the material picture library are stored in a unified format; for small materials, the resolution reaches above 1000×1000 pixels; for large materials, it can be adjusted according to the actual situation, but it is also necessary to ensure that the key features of the materials can be clearly identified. The material types include at least metals, gypsum, etc., and size thresholds are set according to the usage requirements and production process requirements of the materials. The material numbers should follow certain rules to facilitate identification, management, and data processing. The material numbers should be associated with data such as the material picture library, material types, material dimensions and thresholds, to facilitate quick query and acquisition of relevant material information. For example, when querying a certain material type, detailed information such as the number, picture, size, and threshold of the material can be obtained at the same time.

[0023] By capturing the material images on the conveyor belt in real time and comparing and analyzing them with the preset material picture library, identify the material types and material dimensions, where the material dimensions include the length, width, and height of the material. Specifically, select an industrial camera or camera with high resolution, high frame rate, and low latency for shooting to ensure that the material images on the conveyor belt can be clearly captured. According to the width, speed of the conveyor belt, and the characteristics of the materials, reasonably arrange the position and angle of the camera to ensure that the image coverage is comprehensive and there are no blind spots. Perform preprocessing operations on the captured original images to improve the image quality for subsequent analysis. Use algorithms to extract material features from the preprocessed images and perform comparative analysis in combination with the preset material picture library to identify the types and dimensions of the materials, where the dimensions include the length, width, and height of the materials.

[0024] Furthermore, by capturing the material images on the conveyor belt in real time and comparing and analyzing them with the preset material picture library, identify the material types and material dimensions, including: Use a high-resolution industrial camera or webcam as the image acquisition device, installed above or on the side of the conveyor belt; Perform denoising, contrast enhancement, and grayscale preprocessing on the acquired original image; Use edge detection algorithms to identify the position and shape of the material in the image; Extract the features of the material, where the features include the type and size of the material. The types include gypsum and metal, and the size includes length, width, and height; Match the extracted material type with a preset material image library and use a convolutional neural network for material type recognition; Calculate the volume of the material based on the extracted size.

[0025] Specifically, choosing the right camera is crucial for accurately capturing the material image on the conveyor belt. The selection should be based on the width of the conveyor belt and the required image resolution. A high-resolution industrial camera or webcam is used for image acquisition, and the field of view of the camera should cover the entire belt to ensure that no material image is missed. The camera or webcam is installed above or on the side of the conveyor belt to ensure a comprehensive capture of the material image. The installation position should avoid creating shadows and also consider the size and shape of the material to ensure the integrity of the material in the image; Due to various interference factors that may exist in the transportation environment, such as the vibration of the belt and electromagnetic interference from surrounding equipment, noise may appear in the image. Therefore, denoise the acquired original image. Filtering techniques, such as Gaussian filtering, are used to remove the noise. Secondly, to make the features of the material more obvious, enhance the image. Then perform grayscale preprocessing to convert the color image to a grayscale image to reduce the data volume and improve the efficiency of subsequent processing; Use edge detection algorithms, such as the Sobel algorithm, Prewitt algorithm, Roberts algorithm, and Canny algorithm, etc., to identify the edge contour of the material in the image. The result of edge detection is used to determine the position and shape of the material in the image; Extract the features of the material, including the type and size of the material. The material types include gypsum and metal, and the size includes the length, width, and height of the material; Match the extracted material type with a preset material image library and use a convolutional neural network (CNN) to identify the type of the material; Calculate the volume of the material based on the extracted material size features (length, width, and height). For materials with regular shapes, such as cuboids, the volume can be directly calculated using the product of length, width, and height; for materials with irregular shapes, approximate calculation methods, such as the minimum bounding rectangle, convex hull, etc., can be used to calculate the volume.

[0026] Compare the recognized material types and sizes with the types and size thresholds of the preset qualified materials, identify the materials with sizes exceeding the thresholds, and establish a dataset of unqualified materials. The unqualified materials include those with non-metal material sizes exceeding the thresholds and materials with metal material types. The dataset of unqualified materials includes the numbers, types, and sizes of the unqualified materials. Specifically, first, for the recognized material types and sizes, compare them with the types and size thresholds of the preset qualified materials to identify the materials with sizes exceeding the thresholds and the materials with metal material types. Establish a dataset of unqualified materials, and assign a unique number to each unqualified material for easy tracking and management. The unqualified materials include those with non-metal material sizes exceeding the thresholds and the materials with metal material types. The dataset of unqualified materials records the numbers, types, and sizes of the unqualified materials. Utilize the dataset of unqualified materials to analyze the causes and trends of unqualified materials, providing data support for quality control. By analyzing the dataset of unqualified materials, identify the bottlenecks and problem points in the production process, continuously optimize the production process and flow, and improve production efficiency and product quality.

[0027] Furthermore, compare the recognized material types and sizes with the types and size thresholds of the preset qualified materials, identify the materials with sizes exceeding the thresholds, and establish a dataset of unqualified materials. The unqualified materials include those with non-metal material sizes exceeding the thresholds and the materials with metal material types. The dataset of unqualified materials includes the numbers, types, and sizes of the unqualified materials, including: Based on each qualified material type, set the upper threshold of the size. Compare the recognized material type with the preset qualified material types. If the material type is metal, directly mark the current material as unqualified. If the material type is non-metal, compare the size of the non-metal material with the preset size threshold. If the material size exceeds the threshold range, mark it as unqualified. Summarize the information of all the materials marked as unqualified. The information of unqualified materials includes the numbers, types, and sizes of the materials. Use the summarized information of unqualified materials to construct a dataset of unqualified materials. The dataset of unqualified materials contains the unique numbers, types, and size information of each unqualified material.

[0028] Specifically, according to the production standard or quality standard, set the upper threshold of the size for each qualified material type. For example, during the production process, if the volume of gypsum is too large, it will block the chute. However, since the belt is still continuously transporting and producing, it will eventually lead to material overflow, which not only affects production but also requires organizing manpower and material resources to clean up the overflow. Therefore, set the length threshold of gypsum to 300 mm for limitation. Compare the recognized material type with the preset qualified material types to determine whether it falls within the qualified range. If the material type is metal, according to the pre-set rules, regardless of whether its size exceeds the threshold, directly mark this material as unqualified. For example, if the recognized material type is "iron", and iron belongs to metal materials, then this "iron" material will be marked as unqualified; If the material type is a non-metal material, then compare its size with the preset size threshold to determine whether it exceeds the threshold. According to the comparison result, if the material size exceeds the threshold range, mark it as unqualified; Summarize the information of all materials marked as unqualified. This summarization process needs to collect relevant information from each detection link or recognition process. For each material marked as unqualified, obtain its number, type, and size.

[0029] Use the summarized unqualified material information to construct an unqualified material data set. In this data set, each unqualified material has its unique number, type, and size information. This unqualified material data set is stored in the database, which is convenient for subsequent query and statistical analysis. For example, by querying the data set, the number of unqualified materials of different types can be counted, and it can be analyzed which material or which size is more likely to have problems, so as to provide data support for the improvement of production processes, the management of raw material suppliers, etc.

[0030] Furthermore, according to the moving speed of the materials on the conveyor belt and the position information of the recognized unqualified materials, calculate the real-time position of the unqualified materials transported on the conveyor belt, including: Obtain the moving speed of the materials through a belt speed sensor; When a material is recognized as unqualified, the intelligent control system records the specific position of the material on the belt and the moment when the material is recognized. The specific position is the distance relative to the starting point of the belt; The calculation formula for the real-time position of unqualified materials on the belt is as follows: D d = D i + V×(T c -T i ) ; Wherein, D d is the real-time position of the unqualified material on the belt, D i is the initial position of the material, T c is the current moment, T i is the moment when the material is recognized, V is the moving speed of the material, T c -T i is the time difference from the moment when the material is recognized to the current moment, V×(T c -Ti ) Calculate the moving distance of the material on the conveyor belt within the time difference by multiplying the material moving speed and the time difference, D i + V×(T c -T i ) Calculate the real-time position of the unqualified material on the conveyor belt by adding its initial position and the moving distance.

[0031] Specifically, the running speed of the conveyor belt is measured and obtained in real time by a speed sensor installed on the conveyor belt, and this speed is the moving speed of the material on the conveyor belt; When the material is identified as unqualified, the intelligent control system records the specific position of the material on the conveyor belt and the moment when the material is identified. The specific position is represented by the distance relative to the starting point of the conveyor belt, with the unit of meter (m); the time record should be accurate to seconds (s) to ensure the accuracy of the time difference.

[0032] Obtain the unqualified material data set. When it is judged that the type of the unqualified material is non-metal, compare the size of the non-metal material. When the size is greater than the preset threshold, read the real-time position of the non-metal material, and drive the propulsion device to push the non-metal material into the crushing device according to the real-time position. The crushing device is configured on the side of the conveyor belt to crush the material, and the crushed material is transported back to the starting position of the conveyor belt. Specifically, obtain the data set of unqualified materials, which should include the number, type, and size information of the materials. Traverse the unqualified material data set for judgment. If the type of the unqualified material is non-metal, further judge the size of the non-metal material. Compare the size of the non-metal material with the preset threshold. If the size of the non-metal material is greater than the preset threshold, read the real-time position of the non-metal material, and drive the propulsion device to push the non-metal material into the crushing device. The crushing device has sufficient crushing capacity and adaptability to handle non-metal materials of different types and sizes, and is configured on the side of the conveyor belt to ensure that the non-metal materials can be crushed without affecting the transportation of other materials. During the pushing process, the position and state of the material should be monitored in real time to ensure that the material can reach the crushing device smoothly. After crushing, collect the crushed material from the crushing device and transport it back to the starting position of the conveyor belt.

[0033] When it is judged that the type of the unqualified material is metal, obtain the type and size of the metal material, obtain the real-time position of the metal material, and call the iron removal mechanism to grab the metal material off the conveyor belt according to the preset grabbing scheme; Specifically, when the type of unqualified material is metal, it is necessary to obtain the type and size of the metal material. The size information includes the length, width, and height of the metal material. In addition, it is also necessary to obtain the real-time position of the metal material on the conveyor belt. After obtaining the position, call the iron removal mechanism to grab the metal material off the conveyor belt according to the preset grabbing scheme.

[0034] Further, when it is determined that the type of unqualified material is metal, obtaining the type and size of the metal material, obtaining the real-time position of the metal material, and calling the iron removal mechanism to grab the metal material off the conveyor belt includes: According to the type and size of the metal material, preset the grabbing scheme of the iron removal mechanism; According to the real-time obtained material position, the control system issues an instruction to start the iron removal mechanism for grabbing operation.

[0035] Specifically, according to the type and size of the metal material, preset corresponding iron removal mechanism grabbing schemes for different metal materials and sizes in advance. When it is detected that the material is a metal material, the control system will automatically select the most suitable grabbing scheme from the grabbing scheme library according to the type and size information of the material; Obtain the position information of the metal material on the conveyor belt in real time. The control system will automatically generate a grabbing instruction according to the real-time position information and the preset grabbing scheme. And according to the type and hardness of the metal material, select a suitable grabbing method. Transmit the grabbing instruction to the iron removal mechanism for grabbing operation. The iron removal mechanism moves above the material according to the grabbing instruction and precisely grabs the material through magnetism or mechanical jaws. After successful grabbing, move the material smoothly to a safe area outside the conveyor belt to avoid interfering with the production line.

[0036] Further, the presetting of the grabbing scheme of the iron removal mechanism according to the type and size of the metal material includes: Configure the grabbing scheme according to the size and shape of the metal material and establish a grabbing scheme library; Judge that when the type of the metal to be grabbed is a magnetic metal, the iron removal mechanism sucks the magnetic metal through the magnetic field generated by the permanent magnet and grabs the magnetic metal off the conveyor belt; When the type of the metal to be grabbed is a non-magnetic metal, the iron removal mechanism uses the clamping mechanism to clamp and grab the non-magnetic metal, and the clamping mechanism realizes the grabbing of the material through the opening and closing of the mechanical jaws.

[0037] Specifically, first, collect the size and shape data of various metal materials that may appear on the production line. Classify and file the collected data according to the type of metal material (such as iron, copper, aluminum, etc.) and size range to form a detailed material database. Design specific grabbing schemes for each combination of metal material and size; Install a magnetic field sensor on the iron removal mechanism to detect whether the metal to be grabbed is a magnetic metal. When it is detected that the metal to be grabbed is a magnetic metal, the iron removal mechanism uses the magnetic field generated by the permanent magnet to absorb the metal. The magnetic force and position of the permanent magnet should be adjusted according to the type and size of the metal to ensure that the metal can be stably absorbed and avoid interfering with other materials. After successful absorption, move the metal smoothly to a safe area outside the conveyor belt and release the metal; When it is detected that the metal to be grabbed is a non-magnetic metal, the iron removal mechanism uses the clamping mechanism to grab it. The clamping mechanism clamps and grabs the non-magnetic metal away from the conveyor belt by opening and closing the mechanical jaws. The clamping force should be adjusted according to the type and size of the metal to avoid damaging the metal or unstable clamping. After successful clamping, also move the metal smoothly to a safe area and release the jaws to let the metal fall.

[0038] Real-time monitor the operating states of the image acquisition device, the propulsion device and the iron removal mechanism. When the intelligent control system detects equipment anomalies or reaches the preset alarm threshold, immediately send out an alarm signal.

[0039] Specifically, the system real-time monitors the operating states of the image acquisition device, the propulsion device and the iron removal mechanism. Set various alarm thresholds (such as vibration amplitude, temperature range, image recognition accuracy, etc.) according to the equipment type, operating environment and historical data. When the system detects equipment anomalies (such as abnormal vibration, too high temperature, abnormal image recognition, etc.) or reaches the preset alarm threshold, immediately trigger the alarm mechanism. The alarm signal can be notified to relevant personnel in the form of sound, warning box and mobile terminal notification, etc., so as to repair, adjust or clean the image acquisition device, the propulsion device and the iron removal mechanism in time. After receiving the alarm signal, relevant personnel immediately view the alarm details and the on-site monitoring screen, understand the specific situation and location of the equipment anomaly, and quickly check, repair or replace the faulty parts of the equipment according to the alarm information to ensure that the equipment resumes normal operation.

[0040] Embodiment 2, based on the same inventive concept as the image recognition-based intelligent material processing method in the foregoing embodiment, as Figure 2 shown, the present application provides an image recognition-based intelligent material processing system. The system in the embodiment of the present application and the method embodiment are based on the same inventive concept. Among them, the system includes: A digital connection establishment module 10, where the digital connection establishment module 10 is used to establish a digital connection between the intelligent control system and the image acquisition device, the propulsion device and the iron removal mechanism; A qualified material data set configuration module 20, where the qualified material data set configuration module 20 is used to configure a qualified material data set, and the qualified material data set includes a material picture library, material types, material sizes and thresholds, and material numbers; An identification module 30, which is used to identify the material type and size by capturing the material image on the conveyor belt in real time and comparing and analyzing it with a preset material picture library. The material size includes the length, width, and height of the material; A data set establishment module 40, which is used to compare the identified material type and size with the type and size thresholds of preset qualified materials, identify materials with sizes exceeding the thresholds, and establish a data set of unqualified materials. The unqualified materials include those with non-metallic material sizes exceeding the thresholds and materials with a metal material type. The data set of unqualified materials includes the numbers, types, and sizes of unqualified materials; A real-time position calculation module 50, which is used to calculate the real-time position of the unqualified material during transportation on the conveyor belt according to the moving speed of the material on the conveyor belt and the position information of the identified unqualified material; A material propulsion module 60, which is used to obtain the data set of unqualified materials, determine when the type of unqualified material is non-metallic, compare the size of the non-metallic material, read the real-time position of the non-metallic material when the size is greater than a preset threshold, and drive the propulsion device to push the non-metallic material into the crushing device according to the real-time position. The crushing device is configured on the side of the conveyor belt to crush the material and transport the crushed material back to the starting position of the conveyor belt; A metal grasping module 70, which is used to determine when the type of unqualified material is metal, obtain the type and size of the metal material, obtain the real-time position of the metal material, and call the iron removal mechanism to grasp the metal material away from the conveyor belt according to a preset grasping scheme; An intelligent monitoring module 80, which is used to monitor the operating states of the image acquisition device, the propulsion device, and the iron removal mechanism in real time. When the intelligent control system detects equipment abnormalities or reaches a preset alarm threshold, it immediately issues an alarm signal.

[0041] Furthermore, the identification module 30 is used to execute the following method: Use a high-resolution industrial camera or camera as the image acquisition device and install it above or on the side of the conveyor belt; Perform denoising, contrast enhancement, and grayscale preprocessing on the collected original image; Use an edge detection algorithm to identify the position and shape of the material in the image; Extract the features of the material, where the features include the type and size of the material. The type includes gypsum and metal, and the size includes length, width, and height; Match the extracted material type with a preset material picture library and use a convolutional neural network to identify the material type; Calculate the material volume based on the extracted dimensions.

[0042] Furthermore, the dataset building module 40 is used to execute the following method: Based on each qualified material type, set the upper threshold of the dimension; Compare the identified material type with the preset qualified material types. If the material type is metal, directly mark the current material as unqualified; If the material type is non-metal, compare the dimensions of the non-metal material with the preset dimension threshold. If the material dimensions exceed the threshold range, mark it as unqualified; Summarize the information of all materials marked as unqualified. The unqualified material information includes the material number, type, and dimensions; Use the summarized unqualified material information to construct an unqualified material dataset, which contains the unique number, type, and dimension information of each unqualified material.

[0043] Furthermore, the real-time position calculation module 50 is used to execute the following method: Obtain the moving speed of the material through the belt speed sensor; When the material is identified as unqualified, the intelligent control system records the specific position of the material on the belt and the moment when the material is identified. The specific position is the distance relative to the starting point of the belt; The formula for calculating the real-time position of the unqualified material on the belt is as follows: D d = D i + V×(T c -T i ) ; Where, D d is the real-time position of the unqualified material on the belt, D i is the initial position of the material, T c is the current moment, T i is the moment when the material is identified, V is the moving speed of the material, T c -T i is the time difference from the moment when the material is identified to the current moment, and V×(T c -T i ) is to calculate the moving distance of the material on the belt during the time difference by multiplying the material moving speed and the time difference. D i + V×(T c -T i ) is to calculate the real-time position of the unqualified material on the belt by adding its initial position and the moving distance.

[0044] Furthermore, the metal grasping module 70 is used to execute the following method: According to the type and size of the metal materials, preset the grasping scheme of the iron removal mechanism; Based on the real-time obtained material position, the control system issues an instruction to start the iron removal mechanism for grasping operation.

[0045] It should be noted that the above-mentioned sequence of embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above description of specific embodiments of this specification has been made. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be executed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require the specific order or sequential order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0046] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

[0047] This specification and the drawings are only exemplary descriptions of the present application and are considered to have covered any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. The intelligent material processing method based on image recognition is characterized in that: The method comprises: Establish digital connection between intelligent control system and image acquisition equipment, propulsion device and iron removal mechanism; Configure a qualified material data set, wherein the qualified material data set includes a material image library, material type, material size and threshold, and material number; By capturing the image of the material on the conveyor belt in real time and comparing and analyzing it with the preset material image library, the material type and material size can be identified, and the material size includes the length, width and height of the material; Compare the identified material type and material size with the preset qualified material type and size threshold, identify the material whose size exceeds the threshold, and establish an unqualified material data set, wherein the unqualified material includes non-metallic materials whose size exceeds the threshold and materials whose material type is metal, and the unqualified material data set includes the number, type, and size of the unqualified material; Calculate the real-time position of the unqualified material on the conveyor belt according to the material moving speed on the conveyor belt and the identified position information of the unqualified material; Obtaining a data set of unqualified materials, determining that when the type of unqualified materials is non-metallic, comparing the size of the non-metallic materials, reading the real-time position of the non-metallic materials when the size is greater than a preset threshold, driving the propulsion device to push the non-metallic materials into the crushing device according to the real-time position, the crushing device is arranged on the side of the conveyor belt, crushing the materials, and re-transporting the crushed materials to the starting position of the conveyor belt; When it is determined that the type of unqualified material is metal, the type and size of the metal material are obtained, the real-time position of the metal material is obtained, and the iron removal mechanism is called to grab the metal material from the conveying belt according to a preset grabbing scheme; Monitor the operating status of the image acquisition equipment, propulsion device and iron removal mechanism in real time. When the intelligent control system detects equipment abnormality or reaches the preset alarm threshold, it will immediately issue an alarm signal.

2. The material intelligent processing method based on image recognition according to claim 1 is characterized in that: The method captures the material image on the conveyor belt in real time and compares and analyzes it with the preset material image library to identify the material type and material size, including: Use a high-resolution industrial camera or webcam as an image acquisition device, installed above or on the side of the conveyor belt; Perform denoising, contrast enhancement, and grayscale preprocessing on the collected original images; Use edge detection algorithms to identify the location and shape of materials in images; Extracting characteristics of the material, the characteristics including the type and size of the material, the type including gypsum and metal, and the size including length, width, and height; Match the extracted material types with the preset material image library, and use convolutional neural networks to identify the material types; The volume of material is calculated based on the extracted size.

3. The material intelligent processing method based on image recognition according to claim 1, characterized in that: The identified material type and material size are compared with the preset qualified material type and size threshold, materials with sizes exceeding the threshold are identified, and an unqualified material data set is established, wherein the unqualified materials include non-metallic materials with sizes exceeding the threshold and materials of metal type, and the unqualified material data set includes the number, type, and size of the unqualified materials, including: Set upper size thresholds based on each qualified material type; Compare the identified material type with the preset qualified material type. If the material type is metal, the material will be directly marked as unqualified. If the material type is non-metal, the size of the non-metal material is compared with a preset size threshold, and if the material size exceeds the threshold range, it is marked as unqualified; Summarize the information of all materials marked as unqualified, including the material number, type and size; The aggregated unqualified material information is used to construct an unqualified material data set, wherein the unqualified material data set includes a unique number, type, and size information of each unqualified material.

4. The material intelligent processing method based on image recognition according to claim 1, characterized in that: The method of calculating the real-time position of the unqualified material on the conveyor belt according to the material moving speed on the conveyor belt and the identified position information of the unqualified material comprises: The moving speed of the material is obtained through the belt speed sensor; When a material is identified as unqualified, the intelligent control system records the specific position of the material on the belt and the time when the material is identified. The specific position is the distance relative to the starting point of the belt. The formula for calculating the real-time position of unqualified materials on the belt is as follows: D d = D i + V×(T c -T i ) ; Among them, D d is the real-time position of the unqualified material on the belt, D i is the initial position of the material, T c is the current moment, T i is the moment when the material is identified, V is the moving speed of the material, T c -T i is the time difference from the time when the material is identified to the current time, V×(T c -T i ) is to calculate the moving distance of the material on the belt during the time difference by multiplying the material moving speed and the time difference, D i + V×(T c -T i ) is to calculate the real-time position of the rejected material on the belt by adding its initial position to the travel distance.

5. The material intelligent processing method based on image recognition according to claim 1, characterized in that: When the unqualified material is determined to be metal, the type and size of the metal material are obtained, the real-time position of the metal material is obtained, and the iron removal mechanism is called to grab the metal material from the conveying belt according to a preset grabbing scheme, including: Pre-set the grabbing scheme of the iron removal mechanism according to the type and size of the metal material; Based on the material position acquired in real time, the control system issues instructions to start the iron removal mechanism to perform grabbing operations.

6. The material intelligent processing method based on image recognition according to claim 5 is characterized in that: The grabbing scheme of the iron removal mechanism is pre-set according to the type and size of the metal material, including: Configure grabbing solutions according to the size and shape of metal materials and establish a grabbing solution library; When it is determined that the type of metal to be grabbed is magnetic metal, the iron removal mechanism absorbs the magnetic metal through the magnetic field generated by the permanent magnet, and grabs the magnetic metal away from the conveyor belt; When the type of metal to be grasped is non-magnetic metal, the iron removal mechanism uses a clamping mechanism to clamp the non-magnetic metal and grab it off the conveyor belt. The clamping mechanism grasps the material by opening and closing the mechanical clamp.

7. The material intelligent processing system based on image recognition is characterized by: The system is used to implement the material intelligent processing method based on image recognition according to any one of claims 1 to 6, and the system comprises: A digital connection establishment module, which is used to establish a digital connection between the intelligent control system and the image acquisition device, the propulsion device and the iron removal mechanism; A qualified material data set configuration module, wherein the qualified material data set configuration module is used to configure a qualified material data set, wherein the qualified material data set includes a material image library, material type, material size and threshold, and material number; An identification module, which is used to capture the image of the material on the conveyor belt in real time and compare and analyze it with a preset material image library to identify the type and size of the material, wherein the material size includes the length, width and height of the material; A data set establishment module, the data set establishment module is used to compare the identified material type and material size with the preset qualified material type and size threshold, identify the material whose size exceeds the threshold, and establish an unqualified material data set, the unqualified material includes non-metallic material whose size exceeds the threshold and material whose material type is metal, and the unqualified material data set includes the number, type and size of the unqualified material; A real-time position calculation module, which is used to calculate the real-time position of the unqualified material on the conveyor belt according to the material moving speed on the conveyor belt and the identified position information of the unqualified material; A material pushing module, the material pushing module is used to obtain a data set of unqualified materials, and when the type of unqualified materials is non-metallic, compare the size of the non-metallic materials, and read the real-time position of the non-metallic materials when the size is greater than a preset threshold, and drive the pushing device to push the non-metallic materials into the crushing device according to the real-time position. The crushing device is arranged on the side of the conveyor belt to crush the materials and re-transport the crushed materials to the starting position of the conveyor belt; A metal grabbing module, which is used to determine when the unqualified material is metal, obtain the type and size of the metal material, obtain the real-time position of the metal material, and call the iron removal mechanism to grab the metal material from the conveying belt according to a preset grabbing scheme; The intelligent monitoring module is used to monitor the operating status of the image acquisition device, the propulsion device and the iron removal mechanism in real time. When the intelligent control system detects that the equipment is abnormal or reaches a preset alarm threshold, an alarm signal is immediately issued.

Images