Recycling full-period monitoring method and system for waste paper products

Abstract

The invention relates to the field of waste recovery monitoring, in particular to a full-period recovery monitoring method and system for waste paper products. The method comprises the following steps: acquiring a multi-angle real-time monitoring image of a to-be-treated waste paper product pile; local brightness enhancement processing is carried out on the multi-angle real-time monitoring image, panoramic image reconstruction is carried out, and a waste paper pile panoramic image model is constructed; waste paper feature classification is carried out on the waste paper pile panoramic image model, an RFID electronic code is generated, and waste paper pile real-time monitoring parameters are obtained; carrying out real-time positioning and tracking according to the real-time monitoring parameters of the waste paper pile, and carrying out current transportation path analysis so as to generate a current transportation path of the waste paper pile; obtaining a waste paper product operation log; and performing dynamic transportation track mapping according to the waste paper product operation log and the current transportation path of the waste paper pile, and constructing a transportation track site map. According to the invention, the recovery efficiency of waste paper products and real-time recovery state monitoring are improved.

Description

Technical Field

[0001] The present invention relates to the field of waste recycling monitoring, and in particular to a full-cycle monitoring method and system for recycling waste paper products. Background Art

[0002] With the continuous improvement of environmental awareness, the recycling and reuse of waste paper products has gradually become an important issue worldwide. Paper products, as common consumer goods in daily life, have a high recycling value. The recycling of waste paper products can not only reduce resource waste, but also effectively reduce environmental pollution. However, the recycling process of waste paper products involves multiple links, including recycling, classification, treatment and reuse. The effective management and monitoring of each link is crucial to improving recycling efficiency and quality. Traditional monitoring methods for the recycling of waste paper products mainly rely on manual intervention and monitoring of a single link, and often face problems such as inaccurate information, blind spots in monitoring and low efficiency, resulting in poor recycling processes and affecting the recycling rate and maximum utilization of resources.

[0003] With the rapid development of the Internet of Things, big data and artificial intelligence technologies, the intelligence and automation level of modern recycling systems are constantly improving. In the process of recycling waste paper products, the full-cycle monitoring system can collect real-time data and conduct intelligent analysis of the recycling process, so as to fully grasp the various changing factors in the recycling process, such as recycling volume, classification accuracy, processing efficiency, etc. By using sensors, data transmission technology and intelligent analysis algorithms, each stage of the recycling process can be accurately monitored to ensure that the recycling of waste paper products is efficient, accurate and in compliance with environmental protection standards.

[0004] However, although some intelligent monitoring technologies have been applied to the field of waste paper product recycling, the current technology still has problems such as limited data processing capabilities, insufficient monitoring accuracy, and inability to fully capture dynamic changes in the recycling process. Therefore, a new full-cycle monitoring method for waste paper product recycling is urgently needed to address the deficiencies in existing technologies, improve recycling efficiency, and achieve more intelligent and accurate waste paper product recycling management. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a full-cycle monitoring method and system for recycling of waste paper products to solve at least one of the above technical problems.

[0006] To achieve the above object, the present invention provides a method for monitoring the whole cycle of recycling of waste paper products, comprising the following steps: Step S1: acquiring a multi-angle real-time monitoring image of a waste paper product pile to be processed; performing local brightness enhancement processing on the multi-angle real-time monitoring image, and performing panoramic image reconstruction to construct a panoramic image model of the waste paper pile.

[0007] Step S2: classifying waste paper features on the waste paper pile panoramic image model, generating RFID electronic codes, and obtaining real-time monitoring parameters of the waste paper pile; Step S3: performing real-time positioning and tracking according to the real-time monitoring parameters of the waste paper pile, and analyzing the current transportation path, thereby generating the current transportation path of the waste paper pile; Step S4: Obtain the waste paper product operation log; perform dynamic transportation track mapping according to the waste paper product operation log and the current transportation path of the waste paper pile, and construct a transportation track site map; Step S5: Identify the next station based on the transport track station map, perform real-time route scheduling optimization, and build a dynamic route scheduling optimization engine; Step S6: Analyze the abnormal fault status of the real-time monitoring parameters of the waste paper pile, perform global collaborative monitoring based on the dynamic path scheduling optimization engine, and build a full-cycle collaborative monitoring model.

[0008] The present invention can fully understand the shape and condition of the waste paper pile by acquiring real-time monitoring images from multiple angles. Real-time performance ensures that the latest information on the waste paper pile can be quickly obtained to avoid delays. Local brightness enhancement processing helps to improve the clarity and visibility of the image, especially in low light or complex environments, to ensure that the image can provide effective visual data. The reconstruction of the panoramic image can integrate images from multiple angles into a complete waste paper pile model, provide an accurate spatial layout for subsequent processing, and facilitate further analysis of the shape and spatial distribution of the waste paper pile. Through feature classification, different types of paper products in the waste paper pile can be identified to help optimize the recycling process. For example, different types of paper can be processed separately and targeted treatment methods can be adopted. RFID electronic coding can give each type of waste paper product a unique identity, ensuring that each batch of waste paper piles can be uniquely identified, tracked and managed to avoid confusion. Through the construction and classification of the panoramic image model, the system can automatically obtain real-time monitoring parameters about the waste paper pile, covering information such as volume, weight, density, etc., to provide data support for subsequent processing. Through real-time positioning and tracking, the system can accurately obtain the location of the waste paper pile, timely understand its movement during transportation, and avoid resource waste or delays caused by unclear location or scheduling errors. By analyzing the current transportation path, the optimal transportation route can be identified and adjusted according to actual needs to improve transportation efficiency and reduce time and cost. Provide a data basis for subsequent dynamic scheduling optimization to ensure that the transportation path of each station can be flexibly adjusted according to actual conditions. Through the operation log of the waste paper pile, detailed information of each transportation operation (such as transportation starting point, end point, transportation time, etc.) can be systematically recorded to facilitate later review and data analysis. The operation log combined with the current transportation path can generate an accurate transportation trajectory site map, which helps to understand each important node in the transportation process and its corresponding transportation situation, and provide real-time feedback during the transportation process. Through the construction of the transportation trajectory site map, the efficiency of each transportation path can be deeply analyzed, providing a theoretical basis for subsequent transportation scheduling and path optimization. By analyzing the transportation trajectory site map, the next station can be accurately identified, and path scheduling optimization can be performed based on real-time traffic conditions, transportation demand and other factors to ensure the efficiency of the transportation process. By optimizing the route, redundant transportation can be reduced, energy and time can be saved, transportation costs can be reduced, and the economic benefits of waste paper product recycling can be improved. A dynamic route scheduling optimization engine is built to realize intelligent decision-making of transportation routes and improve the response speed and flexibility of the overall recycling and transportation system. By analyzing the real-time monitoring parameters of waste paper piles, abnormal faults that may occur during transportation (such as mechanical failures, environmental problems, etc.) can be discovered in time, so that preventive measures can be taken in advance to avoid systematic errors. Global collaborative monitoring based on the dynamic route scheduling optimization engine enables each link of recycling and transportation to be monitored and optimized in real time, improving the overall efficiency of waste paper product recycling.Through full-cycle collaborative monitoring and dynamic scheduling, we can respond quickly and make adjustments when system abnormalities occur, ensuring the stability and sustainability of the waste paper recycling system.

[0009] Preferably, step S1 comprises the following steps: Step S11: acquiring a multi-angle real-time monitoring image of the waste paper product pile to be processed; performing image region segmentation on the multi-angle real-time monitoring image to obtain a plurality of image regions; Step S12: Calculating the image brightness of each of the multiple image regions one by one to generate a brightness value for each region; Step S13: performing global average flatness calculation based on the brightness value of each area to obtain the average brightness value of the image; Step S14: performing brightness deviation area detection on multiple image areas based on the average brightness value of the image, and extracting the brightness deviation area; Step S15: performing local brightness enhancement processing on the brightness deviation area to obtain a local brightness enhanced image; Step S16: reconstructing the panoramic image of the local brightness enhanced image to construct a panoramic image model of the waste paper pile.

[0010] The present invention can fully capture all aspects of the waste paper pile by acquiring real-time monitoring images from multiple angles, avoiding information loss or inaccuracy caused by a single perspective. Regional segmentation of the image can independently process different parts of the waste paper pile, making subsequent brightness calculation and enhancement processing more refined. For example, the brightness and structure of each region can be processed separately to avoid the impact caused by overall processing. Regional segmentation allows image processing to be performed in parallel, and each region can be analyzed independently, improving processing efficiency and response speed. Calculating brightness values ​​region by region helps to identify brightness differences in the waste paper pile. The brightness values ​​of different regions reflect the characteristics of the region, such as the thickness, density and surface reflectivity of the paper pile. Through local brightness calculation, it is possible to more accurately identify which regions need to further enhance the brightness and which regions do not need to be over-processed. This targeted brightness optimization can avoid image distortion caused by over-processing. By calculating the average brightness value of the image, the brightness characteristics of the entire waste paper pile can be obtained. The average value provides a reference for subsequent brightness deviation area detection, which helps to analyze which regions have brightness deviations from the overall, and ensure the brightness balance of the overall image. Calculating the average brightness value provides a global standard for subsequent image processing, avoiding local errors when processing a single area, thereby improving the processing consistency and accuracy of the entire image. By comparing with the average brightness value, areas with large brightness deviations can be automatically detected. These deviations may indicate uneven distribution in the waste paper pile, or reflect local environmental problems (such as insufficient light or pollution). Identifying these areas can help the system better understand the status of the waste paper pile. The extraction of brightness deviation areas can provide precise positioning for subsequent brightness enhancement or correction, avoid unnecessary processing of the entire image, and improve the intelligence and processing efficiency of the system. Local brightness enhancement of brightness deviation areas can improve the visibility and details of these areas. For the detection of waste paper piles, brightness enhancement helps to highlight the shape, texture and density changes of the paper pile, allowing the system to more clearly identify and analyze different parts of the paper pile. Through local enhancement, areas with insufficient brightness will be corrected to ensure that the clarity and contrast of the entire image are more balanced, thereby improving the accuracy of subsequent processing and reducing misidentification or missed detection problems caused by poor image quality. Only the deviation areas are enhanced, avoiding over-processing of the entire image, retaining the natural features of the original image, and improving the recognizability of the areas that need to be focused on. The local images are stitched into a complete panoramic image model to provide an intuitive and comprehensive view for the overall analysis of the waste paper pile. The panoramic image model can not only show the overall picture of the waste paper pile, but also reveal the internal structure and distribution of the paper pile more clearly, providing detailed image data for subsequent feature classification, tracking and positioning steps. By performing panoramic reconstruction on the enhanced image, the visual differences between different perspectives can be eliminated, making the splicing between images more natural, avoiding distortion at the splicing point, and maintaining high-precision spatial information.Constructing a panoramic image model of waste paper piles helps to comprehensively consider information such as the shape, size, and distribution of the waste paper piles, and provide high-quality image support for subsequent intelligent analysis (such as physical property recognition, recycling path planning, etc.).

[0011] Preferably, the specific steps of step S16 are: Perform image noise recognition on the local brightness enhanced image and mark multiple image noise points; Adaptively filter and optimize image noise to construct a filtered denoised image; Perform multi-angle visual analysis on the filtered and denoised images to identify the visual features of the images at each angle; Extracting an image at each angle based on the multi-angle real-time monitoring image; Recognize feature points of the waste paper pile for each angle of the image and mark multiple feature points of the waste paper pile; Calculate the spatial position coordinates of multiple feature points of the waste paper pile to generate the spatial position coordinates of each feature point; Perform panoramic feature matching on the spatial position coordinates of each feature point according to the image visual features at each angle, so as to generate multi-angle visual matching data of feature points of the waste paper pile; Based on the multi-angle visual matching data of feature points of waste paper pile, the filtered denoised image is reconstructed to construct a panoramic image model of waste paper pile.

[0012] By identifying and marking image noise points, the present invention can provide a clear target area for subsequent filtering and noise reduction, ensure the accuracy and efficiency of image processing, and reduce interference from irrelevant areas. Effective noise point identification can ensure the quality of the image, thereby improving the accuracy of subsequent processing processes. In particular, in the feature extraction and recognition of waste paper piles, noise points often affect the extraction of key features. Marking noise points helps to accurately process. Noise reduction processing can effectively eliminate stray signals or interference noise in the image, ensuring that the image is visually clearer and smoother, which is convenient for subsequent feature recognition and analysis. The denoised image helps to improve the recognition accuracy of the texture, spectrum and spatial features of the waste paper pile, ensuring that the subsequent feature point extraction and spatial matching are more accurate. Multi-angle analysis can provide a reference for subsequent spatial features. Feature matching provides rich visual information, which helps the system better understand the three-dimensional structure of the paper pile. Multi-angle visual analysis helps to extract more comprehensive and accurate feature information, enhance the detail performance of the image, and make the subsequent feature point recognition more accurate. Extracting images from multiple angles can provide more perspective information, which helps to build a three-dimensional model of the waste paper pile. Images from each angle help to reveal different parts of the paper pile, ensuring that sufficient information can be obtained no matter which direction you observe from. Multi-angle images can provide a comprehensive observation of the waste paper pile, reduce the information omission that may be caused by a single perspective, and improve the comprehensiveness of image analysis. Feature points are usually the most recognizable parts of the image. Identifying these feature points helps with subsequent image alignment, spatial matching and Three-dimensional modeling, by marking the feature points of the waste paper pile, the system can accurately identify the geometric shape and position of the paper pile. Marking feature points can enhance image contrast, especially in complex backgrounds. The clear marking of feature points makes the core features of the waste paper pile more prominent, which is convenient for subsequent processing. Calculating the spatial coordinates of the feature points of the waste paper pile helps to provide actual spatial reference for subsequent three-dimensional modeling, panoramic image reconstruction, etc. This step can help the system establish the three-dimensional spatial relationship of the paper pile and avoid the accumulation of errors. Through the calculation of spatial coordinates, the system can accurately match feature points at different angles to ensure the consistency and accuracy of feature points at different viewing angles, thereby optimizing the subsequent feature matching process. By matching the visual features of each angle with the spatial coordinates of the feature points The system can generate more accurate panoramic images through matching. Multi-angle visual matching of feature points helps solve image distortion and alignment problems, ensuring the correct position of each feature point of the paper pile at different viewing angles. Multi-angle visual matching can effectively reduce errors caused by factors such as viewing angle differences and lighting changes, ensuring accurate fusion of images at different viewing angles and improving image quality. Panoramic image reconstruction based on multi-angle feature matching data can achieve seamless splicing between multiple viewing angles and generate a complete and highly accurate panoramic image model of waste paper piles. Through the reconstruction of panoramic images, a comprehensive and realistic visual experience can be provided for subsequent monitoring, analysis and decision-making, helping operators to fully understand the shape, structure and potential problems of waste paper piles.The panoramic image model can provide high-quality image data for further classification, positioning, feature extraction and recycling path planning of waste paper piles, supporting full-cycle monitoring and decision-making.

[0013] Preferably, the specific steps of step S2 are: Step S21: performing a paper pile texture analysis on the waste paper pile panoramic image model to generate waste paper pile texture features; Step S22: performing multi-angle paper pile spectral feature mining on the waste paper pile panoramic image model to extract the waste paper pile spectral features; Step S23: quantifying the degree of waste paper stains according to the spectral characteristics of the waste paper pile, thereby generating a waste paper stain quantification value; Step S24: performing paper component identification on the spectral characteristics of the waste paper pile according to the texture characteristics of the waste paper pile to generate waste paper component characteristics; Step S25: performing waste paper feature classification according to waste paper component features and waste paper stain quantification values, thereby generating feature information of a waste paper pile to be processed; Step S26: Generate an RFID electronic code based on the characteristic information of the waste paper pile to be processed, and associate the RFID electronic code with the waste paper product pile to be processed for monitoring, so as to obtain real-time monitoring parameters of the waste paper pile.

[0014] The present invention generally reflects the surface structure, material distribution and density change of paper through the texture of the paper pile. Through texture analysis, the unique features of the paper pile, such as surface roughness, paper type (such as single-layer paper or multi-layer paper) and the degree of aging of the paper, can be effectively extracted. The extraction of texture features not only helps to identify the components of the paper pile, but also provides valuable information for subsequent paper component identification and stain analysis. For example, a rougher texture may indicate the aging or contamination of the paper pile, which needs to be processed in a targeted manner. Texture feature analysis enhances the deep understanding of the waste paper pile image, making subsequent processing decisions more accurate. The spectral features of the waste paper pile can reveal the chemical composition, degree of contamination and physical properties of the paper. Through multi-angle spectral mining, spectral data reflected at different angles can be obtained to avoid errors or insufficient data caused by a single perspective. The spectral information of the paper pile can reflect the pollutants (such as oil, ink or other chemical residues) that may exist on the surface of the paper, providing a scientific basis for the subsequent quantification of the degree of stains. Through multi-angle analysis of spectral features, the system can fully understand the physical and chemical characteristics of the waste paper pile, which helps to classify and recycle more accurately. Waste paper piles often contain stains of varying degrees (such as oil stains, ink stains, etc.). Spectral features can effectively identify and quantify the severity of these stains. By quantifying the degree of stains, the system can provide clear guidance for subsequent recycling and treatment. The quantification of stains allows the recycling system to adjust the treatment strategy according to the degree of stains. For waste paper with high stains, more sophisticated cleaning and treatment processes may be required to ensure the quality of the recycled paper. By quantifying the degree of stains, the system can identify which waste paper requires special treatment and which can be directly recycled, avoiding unnecessary cleaning and treatment steps, thereby improving recycling efficiency and saving time and cost. Combining the texture characteristics and spectral characteristics of the paper pile, the composition of the paper, such as pulp type, filler, coating layer, etc., can be more accurately identified. This helps to identify different types of paper and determine whether it is suitable for recycling or requires special treatment. Composition feature identification helps to classify waste paper according to the material, type and difficulty of handling. For example, some special composition papers (such as paper with chemical coatings) may require additional recycling treatment, while ordinary paper can be directly recycled. By accurately identifying the composition of paper, the system can ensure that each type of paper is handled most appropriately, thereby maximizing resource utilization and reducing unnecessary waste. Combining the composition characteristics and the quantification value of stains, waste paper piles can be classified more finely. This classification can be based on multiple dimensions such as the composition of the paper and the degree of stains, ensuring that subsequent processing steps can be optimized according to the characteristics of different waste paper piles. Through the comprehensive feature classification of waste paper piles, the system can accurately identify which paper piles can be recycled normally and which require special treatment or cleaning. This will help to improve the intelligence of the recycling system, reduce manual intervention, and improve processing efficiency.The classification of characteristic information can help recycling companies or processing manufacturers to allocate resources reasonably, optimize each link of the recycling chain, and improve the efficiency of the entire cycle. Generating RFID electronic codes and embedding them into the waste paper pile can give each pile of waste paper an independent digital identity, thereby realizing the accurate identification, tracking and management of the waste paper pile. This electronic code not only contains the basic information of the waste paper pile, but also can dynamically track its transportation and processing status. Through RFID electronic codes, the system can collect monitoring parameters of the waste paper pile (such as weight, volume, degree of stains, etc.) in real time. This real-time data feedback helps to respond to problems in the recycling process in a timely manner and provide data support for subsequent optimization. RFID technology can realize the automated management of waste paper piles without manual intervention, thereby improving the efficiency, accuracy and operability of waste paper pile monitoring.

[0015] Preferably, step S3 specifically comprises the following steps: Step S31: performing real-time positioning and tracking according to the real-time monitoring parameters of the waste paper pile, and extracting the real-time positioning and tracking data of the paper pile; Step S32: performing time-series transportation trajectory analysis on the real-time positioning and tracking data of the paper pile to generate a time-series transportation trajectory of the paper pile; Step S33: performing stop node identification on the time-series transportation trajectory of the paper pile to obtain a plurality of stop nodes; Step S34: Analyze the current transportation path based on multiple stop nodes to generate the current transportation path of the waste paper pile.

[0016] The present invention can grasp the exact position and status of the waste paper pile at any time through real-time positioning and tracking of the waste paper pile. This real-time tracking capability ensures that the waste paper pile is always under monitoring during the entire recycling and transportation process, reducing the occurrence of problems such as loss, misdelivery or transportation delays. Through precise positioning technology (such as GPS, RFID or sensor networks, etc.), the real-time position data of the paper pile can be collected. These data provide a basis for subsequent trajectory analysis, path optimization and transportation monitoring, ensuring that all operations are based on the latest positioning information. The time-series transportation trajectory analysis generates a transportation trajectory by performing time series analysis on the real-time positioning data of the waste paper pile, which can provide detailed records for each position and time node of the paper pile during the transportation process. Through time series analysis, the complete transportation path of the waste paper pile from the starting point to the end point can be drawn, which helps to review the path selection during the transportation process and identify potential bottlenecks or optimization space. Through the analysis of the transportation trajectory, the system can identify the stop nodes of the waste paper pile during the transportation process. The stop node usually refers to the time when the paper pile stays at a certain location for more than a certain period of time. The identification of stop nodes helps to understand whether there are unnecessary stops or delays in the transportation process. These stops may be caused by traffic problems, equipment failures or other reasons. Identifying stop nodes can help managers take timely measures to reduce invalid stops and improve overall transportation efficiency. The monitoring of stop nodes also helps to analyze potential safety issues. For example, areas with frequent stop nodes may need to be monitored to avoid transportation delays due to bad weather or traffic congestion. Through multiple stop nodes and time-series trajectory data, the system can analyze the current transportation path of the waste paper pile. This analysis can reflect the route taken during the actual transportation process, which is different from the predetermined ideal route, thereby revealing any unnecessary detours, repeated paths or inefficient route choices. Based on the analysis results of the stop nodes, the transportation path can be optimized. For example, if the stop nodes are concentrated in certain specific areas, the system can suggest bypassing these areas and choosing a smoother path, thereby reducing stagnation time and improving transportation efficiency.

[0017] Preferably, the specific steps of step S4 are: Step S41: Obtaining a waste paper product operation log; identifying all recycling sites according to the waste paper product operation log; Step S42: performing spatial distribution analysis on the recycling sites to generate spatial distribution data of recycling sites; Step S43: performing site distribution fitting on the recycling site spatial distribution data to construct a site distribution map; Step S44: dynamically mapping the transportation track of the site distribution map based on the current transportation path of the waste paper pile to construct a transportation track site map.

[0018] The present invention can automatically identify the locations of all sites involved in recycling through operation log analysis, which helps to ensure the integrity and rationality of the recycling network and provides a basis for subsequent spatial analysis and path optimization. The waste paper product operation log contains detailed recycling process data, such as the operation time, transportation route, processing status and other information of each recycling site. By obtaining and analyzing these logs, the key nodes and sites of the waste paper pile in the recycling process can be accurately identified. By performing spatial distribution analysis on the recycling sites, the layout and distribution characteristics of these sites in geographical space can be clearly understood, for example, whether the sites are evenly distributed, whether there are blank areas or concentrated areas, which can effectively help optimize the site settings. The spatial distribution data provides a visual basis for subsequent optimization decisions. For example, it can identify the excessive concentration or lack of recycling sites in certain areas, help plan a more reasonable layout of recycling sites, and fit the spatial distribution data to reveal the regularity or potential pattern of site distribution. For example, some areas may have a large number of sites due to urban areas. Due to reasons such as urban layout and traffic convenience, the distribution density of recycling stations is relatively high; while other areas may have sparse distribution of stations due to low population density or inconvenient transportation. By constructing a station distribution map, the spatial layout of recycling stations can be intuitively displayed, which provides an important basis for the planning of the recycling network, the rational allocation of resources, and the optimization of transportation routes. Combining the current transportation route of the waste paper pile with the recycling station distribution map can realize dynamic mapping of the transportation route and the layout of the recycling station, which can help optimize the association between the transportation route and the station, making the transportation of the waste paper pile more efficient. Through dynamic mapping, the positional relationship between the transportation route and the recycling station can be adjusted in real time. For example, if a transportation route is congested or delayed, the system can automatically adjust the route to ensure that the waste paper pile can reach the next recycling station quickly and smoothly. By combining the station distribution map with the dynamic mapping of the transportation route, the operating status and transportation needs of each station can be accurately understood, thereby optimizing the transportation route, avoiding empty loads or detours, and improving the overall recycling efficiency.

[0019] Preferably, the specific steps of step S5 are: Step S51: Calculating the current position coordinates of the discarded paper pile based on the real-time positioning and tracking data of the paper pile; Step S52: identifying the next station on the transport track station map according to the current waste paper pile location coordinates, and marking the next operation station; Step S53: Calculate the operating distance of the next operating station to obtain the operating distance parameter of the next station; Step S54: Perform dynamic transportation route planning according to the next station operation distance parameter, thereby constructing a dynamic transportation planning path; Step S55: Perform real-time path scheduling optimization on the dynamic transportation planning path and build a dynamic path scheduling optimization engine.

[0020] The present invention can accurately calculate the position coordinates of the current waste paper pile by obtaining the real-time positioning data of the waste paper pile (such as GPS, RFID or sensor network). This provides an important basis for subsequent transportation path analysis and path optimization. Real-time position calculation can monitor the position of the waste paper pile at any time to help managers understand the transportation status of the waste paper pile. Through the real-time update of the position coordinates, lags or misalignments in the transportation process can be avoided, ensuring that the waste paper pile is smoothly transferred to the next station. According to the current positioning coordinates of the waste paper pile, the system can automatically identify the next operating station. This step can ensure the continuity of the transportation process and avoid interruption or duplication of the path. By dynamically identifying the next station, the transportation scheduling can be optimized to guide the waste paper pile to the most suitable recycling station. This process is based on real-time transportation needs and the current position of the waste paper pile, which can improve the accuracy of transportation. By calculating the operating distance between the current position of the waste paper pile and the next station, the transportation time and cost can be more accurately estimated. This helps to further optimize the configuration of transportation resources, such as reasonably arranging transportation tools and personnel scheduling. The calculation of the operating distance provides a basis for subsequent path optimization and transportation scheduling. Knowing the transportation distance between each station can help predict possible problems during transportation, such as whether a break is needed or whether the route needs to be replanned. Dynamic path planning can adjust the transportation path in real time based on factors such as the current transportation distance, traffic conditions, and station location. This not only improves the efficiency of transportation, but also reduces unnecessary detours and repeated routes. By optimizing the transportation path, unnecessary detours are reduced, and fuel consumption and time costs are reduced. This helps to reduce operating costs in the recycling of waste paper products. The dynamic path scheduling optimization engine can automatically optimize the transportation path based on real-time transportation status, traffic information, and the location information of the waste paper pile. The engine continuously monitors every link in the transportation process to ensure that the waste paper pile is always on the optimal path. The path scheduling engine not only optimizes a single transportation path, but also considers global scheduling, coordinates the relationship between multiple transportation vehicles and stations, and conducts resource sharing and collaborative optimization. For example, the transportation paths of multiple waste paper piles can achieve optimal resource allocation through the engine to avoid path conflicts or repeated transportation.

[0021] Preferably, the specific steps of step S6 are: Step S61: Analyze the abnormal fault state of the real-time monitoring parameters of the waste paper pile and extract the abnormal fault state data; Step S62: Calculate the fault location according to the abnormal fault state data to generate the abnormal fault occurrence location; Step S63: Visually mark the transport track site map according to the location where the abnormal fault occurs, thereby generating a fault site map; Step S64: Make emergency fault decisions based on the fault occurrence site map and build an emergency fault handling strategy; Step S65: Perform global collaborative monitoring on the emergency fault handling strategy and the dynamic path scheduling optimization engine to build a full-cycle coordinated monitoring model.

[0022] The present invention can timely discover potential faults in the transportation process by real-time monitoring of the status of the waste paper pile (such as temperature, humidity, position, transportation conditions, etc.). For example, abnormal temperature rise or humidity fluctuation may mean equipment failure or transportation problem. Abnormal fault state analysis helps to identify these potential problems in advance and prevent small faults from becoming big problems. Through data analysis algorithms (such as pattern recognition based on machine learning), fault states can be extracted from real-time monitoring data. For example, if a certain monitoring parameter (such as vibration, speed, position change, etc.) exceeds a preset range, the system can automatically identify possible fault states. Through abnormal fault state data, the location where the fault occurred can be accurately calculated, which is very important for subsequent Emergency handling is very critical. It can help staff quickly find the source of the fault and reduce the spread time of the fault impact. By locating the fault location, maintenance personnel and necessary resources can be quickly deployed to ensure that fault handling can start in the shortest time possible, reducing downtime and transportation delays. By visually marking the fault location and generating a fault site map, the fault point and its surrounding environment can be clearly presented. This map helps managers understand the spatial distribution of faults, facilitating subsequent emergency decision-making and resource allocation. Managers can understand the latest developments of the fault point and monitor the repair progress through the real-time updated fault site map. This visual management method improves the transparency of fault handling. and collaborative efficiency. Once a fault occurs and its location is calibrated, the system can quickly formulate an emergency response strategy based on the type and severity of the fault. For example, if it is a failure of the means of transport, a maintenance team can be dispatched immediately; if it is a road blockage or traffic problem, the transportation route can be replanned. The system helps managers determine the priority of emergency response based on the location and time of the fault and the importance of transportation. For example, faults near important recycling sites are given priority to ensure that the recycling of waste paper piles is not affected. By integrating the emergency fault handling strategy with the dynamic path scheduling optimization engine, global collaborative monitoring can track all problems in the transportation and recycling processes in real time. The system can span multiple transportation routes and multiple sites. Collaborative monitoring is carried out with multiple vehicles to ensure that all links are under monitoring and to respond to various abnormal situations in a timely manner. Based on the full-cycle collaborative monitoring model, the system can intelligently adjust the transportation route, maintenance plan and other resource allocations after a fault occurs. For example, if a transportation route fails, the system can automatically guide the waste paper pile to an alternative route and dispatch maintenance personnel for on-site processing. Through the full-cycle coordinated monitoring model, the recycling process of the waste paper pile can be monitored and optimized at each link to ensure the efficient operation of multiple links such as recycling, transportation, and fault handling. This model can be continuously optimized during long-term use, adjusted according to historical data and feedback, and gradually improve the operating efficiency of the entire system.

[0023] In this specification, a system for monitoring the full cycle of recycling of waste paper products is provided, which is used to execute the above-mentioned method for monitoring the full cycle of recycling of waste paper products, including: The image reconstruction module is used to obtain multi-angle real-time monitoring images of the waste paper product pile to be processed; perform local brightness enhancement processing on the multi-angle real-time monitoring images, and perform panoramic image reconstruction to build a panoramic image model of the waste paper pile.

[0024] The RFID encoding module is used to classify waste paper features in the waste paper pile panoramic image model, generate RFID electronic codes, and obtain real-time monitoring parameters of the waste paper pile; A real-time positioning module is used to perform real-time positioning and tracking based on real-time monitoring parameters of the waste paper pile, and to analyze the current transportation path, thereby generating the current transportation path of the waste paper pile; The site map module is used to obtain the waste paper product operation log; dynamically map the transportation track according to the waste paper product operation log and the current transportation path of the waste paper pile, and build a transportation track site map; The route scheduling optimization module is used to identify the next station based on the transport trajectory station map, perform real-time route scheduling optimization, and build a dynamic route scheduling optimization engine; The collaborative monitoring module is used to analyze the abnormal fault status of the real-time monitoring parameters of the waste paper pile, and to perform global collaborative monitoring based on the dynamic path scheduling optimization engine to build a full-cycle collaborative monitoring model.

[0025] The present invention can obtain comprehensive information of waste paper piles through multi-angle image capture and avoid blind spots caused by single-angle monitoring, which helps to more comprehensively analyze the shape, state and distribution of paper piles. Brightness enhancement processing ensures that the details of darker areas in the image are clearly displayed, avoiding the loss of image information due to insufficient light, especially in complex or low-light environments. The monitoring effect is more significant. Through the synthesis and reconstruction of multi-angle images, the system can generate a high-precision panoramic image model of waste paper piles. This image model provides basic data support for subsequent paper pile feature analysis, component identification and other work. By classifying the waste paper features of the panoramic image model, the composition, type and quality of the waste paper pile can be deeply analyzed. For example, it can Identify key features of paper such as material and degree of contamination to provide reference for subsequent recycling or processing decisions. The generated RFID electronic code can not only create an independent identification for each batch of waste paper piles, but also automatically track the flow of waste paper to avoid errors or omissions in manual identification. The implantation of RFID tags allows the real-time data of waste paper piles to be read at any time and supports seamless docking with other modules. RFID encoding technology enables the monitoring parameters of each waste paper pile to be updated in real time (such as location, temperature and humidity, transportation status, etc.), ensuring that the system can grasp the various states of the paper pile in real time and respond quickly to abnormal situations. Through the real-time monitoring system, the location of the waste paper pile can be accurately tracked, providing a basis for path optimization and resource scheduling during transportation. Data, positioning information can accurately reach each transportation node to ensure that the waste paper pile is always on the predetermined path. Real-time positioning information combined with transportation path analysis, the system can dynamically evaluate whether the transportation path is optimal and avoid redundant or inefficient route selection. For example, it can promptly identify problems such as traffic jams and route deviations, and adjust the transportation path. Once an abnormality is detected (such as an unexpected stop or transportation delay), the system can adjust the transportation path of the waste paper pile according to the real-time positioning data, thereby ensuring the efficiency of the overall process. The acquisition and storage of operation logs ensures that each transportation process has a detailed record. These data can be used for subsequent fault diagnosis, path analysis and optimization work, and can also be used as a reference for optimization plans. The transportation trajectory site map will be The specific situation of each transportation is visualized, showing each station that the waste paper pile passes through and the key nodes on the path, helping the staff to clearly understand the overall transportation process of the waste paper pile. According to the real-time location of the waste paper pile, the system can dynamically adjust the transportation trajectory to avoid unnecessary redundant paths and improve transportation efficiency. By real-time analysis of the transportation trajectory site map, the system can automatically identify the next transportation site of the waste paper pile and intelligently optimize the path to ensure the shortest, fastest and safest transportation route. The engine automatically optimizes the transportation route through real-time data analysis to cope with various emergencies (such as traffic jams, temporary changes in sites, etc.). Dynamic scheduling can flexibly respond to changes in the external environment and improve the adaptability and responsiveness of the transportation system.The path scheduling optimization module not only optimizes the path of a single waste paper pile, but also coordinates scheduling between multiple waste paper piles, reasonably arranges transportation tools and resources, and ensures the efficient operation of the entire transportation system. Real-time monitoring and analysis of various data of waste paper piles (such as location, humidity, temperature, vibration, etc.) can timely detect system abnormalities or faults. For example, too high temperature may mean that the transportation tool has failed, or some paper piles are too wet, which may affect the recycling quality. Fault warning can be responded in advance to avoid greater losses. The collaborative monitoring module integrates the data of various links (such as positioning, path scheduling, transportation logs, etc.) and realizes real-time monitoring of the transportation process of all waste paper piles through a full-cycle monitoring system. It can not only improve the fault response speed, but also coordinate the optimization of paths and resources, intelligently coordinate the relationship between various system modules, and optimize the use of resources. For example, when a certain transportation tool fails, the system can reallocate resources according to real-time monitoring data and path optimization solutions to ensure the continuous and efficient operation of the entire transportation system. , BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A schematic flow chart of the steps of a method for monitoring the entire recycling cycle of waste paper products according to the present invention; Figure 2 Detailed implementation flow chart of step S1; Figure 3 Detailed implementation flow chart of step S2. DETAILED DESCRIPTION

[0027] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0028] The present application example provides a method and system for monitoring the whole cycle of recycling of waste paper products. The execution subjects of the method and system for monitoring the whole cycle of recycling of waste paper products include but are not limited to: mechanical equipment, data processing platform, cloud server node, network upload device, etc. equipped with the system can be regarded as the general computing node of the present application, and the data processing platform includes but is not limited to: at least one of an audio and image management system, an information management system, and a cloud data management system.

[0029] See also Figures 1 to 3 The present invention provides a full-cycle monitoring method for recycling waste paper products, and the full-cycle monitoring method for recycling waste paper products comprises the following steps: Step S1: acquiring a multi-angle real-time monitoring image of a waste paper product pile to be processed; performing local brightness enhancement processing on the multi-angle real-time monitoring image, and performing panoramic image reconstruction to construct a panoramic image model of the waste paper pile.

[0030] Step S2: classifying waste paper features on the waste paper pile panoramic image model, generating RFID electronic codes, and obtaining real-time monitoring parameters of the waste paper pile; Step S3: performing real-time positioning and tracking according to the real-time monitoring parameters of the waste paper pile, and analyzing the current transportation path, thereby generating the current transportation path of the waste paper pile; Step S4: Obtain the waste paper product operation log; perform dynamic transportation track mapping according to the waste paper product operation log and the current transportation path of the waste paper pile, and construct a transportation track site map; Step S5: Identify the next station based on the transport track station map, perform real-time route scheduling optimization, and build a dynamic route scheduling optimization engine; Step S6: Analyze the abnormal fault status of the real-time monitoring parameters of the waste paper pile, perform global collaborative monitoring based on the dynamic path scheduling optimization engine, and build a full-cycle collaborative monitoring model.

[0031] In the embodiment of the present invention, refer to Figure 1 , is a schematic flow chart of the steps of a method for monitoring the full cycle of recycling of waste paper products according to the present invention. In this example, the steps of the method for monitoring the full cycle of recycling of waste paper products include: Step S1: obtaining a multi-angle real-time monitoring image of a waste paper product pile to be processed; performing local brightness enhancement processing on the multi-angle real-time monitoring image, and performing panoramic image reconstruction to construct a panoramic image model of the waste paper pile; In this embodiment, a high-resolution industrial camera (e.g., 16 MP or higher) is used to ensure that images with rich details are captured. When selecting the sensor type of the camera, CMOS sensors are preferred to improve performance in low-light environments. 4 to 6 cameras are set to cover different angles of the paper pile. The field of view of each camera should overlap by about 30% to ensure image matching during subsequent stitching. The image acquisition frequency is set to 1 frame per second to obtain real-time monitoring data. If conditions permit, it can be increased to 5 frames per second to capture rapidly changing situations. The exposure time is adjusted according to the ambient light conditions. For example, it is set to 1 / 1000 second in a bright environment, and can be extended to 1 / 30 second in a low-light environment to prevent motion blur. The automatic white balance setting is used to ensure that the images captured by different cameras maintain consistent color performance. A network attached storage (NAS) device is used, and a RAID 1 configuration is set to improve data security and reading speed. All image files should contain a timestamp and camera number (e.g., 20231224_153000_cam1.jpg) for easy subsequent processing. A 5x5 Gaussian filter is applied for denoising to reduce random noise in the image. σ is set The value is 1.5 to ensure smoothing effect while retaining important details. Convert each image to grayscale to simplify subsequent processing. Use adaptive histogram equalization (CLAHE) to set the size of each small block to 8x8 pixels and the maximum contrast limit to 2.0. This can effectively enhance local contrast and avoid over-enhancement. Visually compare the image histograms before and after enhancement to ensure that the image contrast has been significantly improved. Use the ORB feature extraction algorithm to extract the key points of each image and set the number of key points to 500. Perform feature matching through RANSAC and set the matching threshold to 0.75 to ensure the accuracy of matching. Use affine transformation to align the matched images. Optimize the stitching effect through the least squares method to generate a seamless panoramic image. Use the stitcher function in the OpenCV library for image stitching to improve stitching speed and quality. Make sure to set the matching threshold of the overlapping area to 0.8 to improve the robustness of stitching. Monitor the stitching progress in real time to ensure that the output of each step meets expectations. Use the bilateral filtering algorithm and set the diameter to 9, σColor and σSpace. The values ​​are all set to 75 to reduce artifacts at the stitching boundary. A color balancing algorithm is applied to ensure the color consistency of the stitching area and improve the visual effect of the overall image. The final panoramic image is saved in TIFF format to preserve the image quality and avoid compression loss. The relevant metadata of the image, including the shooting time, camera settings, and the list of images involved in the stitching, is saved to facilitate subsequent analysis and retrieval.

[0032] Step S2: classifying waste paper features on the waste paper pile panoramic image model, generating RFID electronic codes, and obtaining real-time monitoring parameters of the waste paper pile; In this embodiment, the panoramic image model is preprocessed to ensure the accuracy of subsequent feature extraction, including denoising, grayscale and normalization operations. A Gaussian filter (5x5 kernel, σ = 1.5) is used to remove noise in the image, and then the image is converted into a grayscale image to reduce computational complexity. Finally, the image is normalized to ensure that the features of different images are in the same scale. Gabor filter and local binary pattern (LBP) are used to extract the texture features of the image. The Gabor filter sets multiple directions (such as 0°, 45°, 90°, 135°) and frequencies to capture texture information of different directions and scales; LBP By binarizing the neighborhood around each pixel, a texture histogram is generated. The spectral features of different areas in the image are extracted using spectral analysis methods. The reflectance data of different wavelengths are obtained through a multispectral sensor to form a spectral feature vector. According to the nature of the feature data, a suitable classification algorithm is selected. Commonly used algorithms include support vector machine (SVM), random forest and k-nearest neighbor (KNN). In this step, the random forest algorithm is selected because it performs well in processing high-dimensional data and has less overfitting. The labeled waste paper feature samples are used to construct a training data set. Each sample should contain a corresponding feature vector and label (such as paper type, quality grade, etc.). The RandomForestClassifier in the Scikit-learn library is used, and the number of trees is set to 100 and the maximum depth is 10 to improve the classification effect. After the training is completed, the accuracy of the model is evaluated to ensure that its accuracy on the validation set is above 85%. A unique RFID is generated based on the classification results. Electronic coding, using the UUID (Universally Unique Identifier) ​​format, ensures the uniqueness of each code. The code contains the paper type, processing date and serial number, for example: UUID-PaperType-Date-Seq (such as 123e4567-e89b-12d3-a456-426614174000-recyclable-20231224-001). Write the generated RFID electronic code into the RFID tag and operate it with the RFID encoder to ensure that the material of the tag is suitable for the environmental conditions of the waste paper pile. Install the RFID reader in the storage area of ​​the waste paper pile to ensure that it can cover the entire area and read the RFID tag information in real time. Set the reader's operating frequency to 13.56 MHz to ensure a good reading distance and stability. The reader transmits real-time monitoring data to the central database, including each RFID The tag's coding, reading time, location and other information are used to set the database structure to support fast query and analysis, and a user interface is developed to display the monitoring parameters of the waste paper pile in real time, including the current paper type, quantity, degree of stains and other information, to facilitate management personnel to make decisions.

[0033] Step S3: performing real-time positioning and tracking according to the real-time monitoring parameters of the waste paper pile, and analyzing the current transportation path, thereby generating the current transportation path of the waste paper pile; In this embodiment, an RFID reader is installed between the transport vehicle and the storage area to ensure that the RFID tag information of the waste paper pile can be captured in real time. When the vehicle passes the reader, the system automatically reads the tag information, including the RFID code, reading time and location, and uses a wireless network (such as Wi-Fi or cellular network) to transmit the real-time collected data to the central database. This process needs to ensure the stability and timeliness of data transmission to avoid data loss caused by network delays. Each time the RFID tag is read, the system records the current timestamp, location coordinates (using GPS positioning technology) and corresponding paper information, such as type, quantity and degree of stain. This information will be used for subsequent path analysis. A GPS positioning device is installed on the transport vehicle to obtain the vehicle's location information in real time. By setting the GPS update frequency to once per second, real-time performance can be guaranteed. The GPS data will include longitude, latitude and altitude information to ensure that the current status of the vehicle can be accurately located. The RFID data and GPS data are integrated to form a comprehensive data set. Each record should contain the RFID code, reading time, location coordinates, paper information, etc., and the data processing software (such as Python's Pandas) can be used to process the data. Database), merge data from different sources for subsequent analysis, use geographic information system (GIS) software (such as QGIS or ArcGIS) to visualize the integrated data, plot GPS positioning data on the map, show the transportation path of the waste paper pile, mark the location of each RFID tag reading point on the map, and connect these points with lines to form a complete transportation path. To improve readability, different colors and marks can be used to represent different types of paper or transportation status. By analyzing the transportation path, the length of the path, transportation time, and possible traffic bottlenecks can be identified. The Dijkstra algorithm or A* algorithm can be used to optimize the path to find the shortest path or optimal transportation strategy. Combined with real-time traffic data (such as using Google Maps API), the transportation path can be dynamically adjusted to ensure efficient transportation.

[0034] Step S4: Obtain the waste paper product operation log; perform dynamic transportation track mapping according to the waste paper product operation log and the current transportation path of the waste paper pile, and construct a transportation track site map; In this embodiment, during each transportation process, an operation log is automatically generated using the vehicle's onboard recorder or logistics management system. The log should include key information, such as departure time, arrival time, stay time, transportation speed, transportation status (such as normal, delayed, etc.) and the type and quantity of paper transported. The operation log is ensured to be in a standard format (such as CSV or JSON) to facilitate subsequent data processing and analysis. A sample log record may include the following fields: vehicle_id, timestamp, location, status, load_type, load_quantity. The integrated data is cleaned to remove duplicate and invalid records to ensure data accuracy and completeness. Python's Pandas library can be used for data cleaning to filter out valid time periods and statuses. A transportation trajectory is generated based on the integrated data. The dynamic transportation trajectory of the vehicle is drawn by analyzing the GPS coordinates and operation status at each time point. An interpolation algorithm (such as linear interpolation) is used to fill in missing data points to ensure the continuity of the trajectory. The generated transportation trajectory data is visualized using GIS software (such as QGIS). or ArcGIS) to draw the path on the map, use different colors or line types to represent different transportation status, such as normal transportation, delay or stop, mark key transportation sites (such as starting point, end point and intermediate stop) on the map, and display relevant information (such as residence time, cargo type, etc.), identify common transportation sites based on the location information in the operation log, these sites may include distribution centers, transfer stations, customer locations, etc., count the residence time and pass frequency of each site, identify key transportation nodes, and record relevant attributes for each identified site, such as coordinates, name, transportation frequency and historical residence time, etc., to facilitate subsequent analysis and decision-making, use GIS software to build a transportation trajectory site map, combine the identified site information with the transportation trajectory data to form a complete transportation trajectory map, the map should contain different layers, respectively showing the transportation path, key sites, and transportation status, so that users can intuitively understand the transportation situation, set up a system automatic update mechanism, and continuously adjust and improve the site map according to real-time data, use Web GIS technology (such as Leaflet or Mapbox) to create interactive online maps, where users can view transportation status and trajectory in real time, analyze transportation trajectory maps, identify bottlenecks and potential problems in the transportation process, such as sections with frequent delays or stations with too long stays, and use statistical analysis tools (such as R or Python) to generate analysis reports to support management decisions. Based on the data analysis results, it can also make suggestions for optimizing the transportation path, such as adjusting the transportation route, adding stops, or optimizing transportation scheduling.

[0035] Step S5: Identify the next station based on the transport track station map, perform real-time route scheduling optimization, and build a dynamic route scheduling optimization engine; In this embodiment, the transportation trajectory site map and real-time monitoring data established in the early stage are used to obtain the current vehicle's position, status and estimated arrival time. These data are updated in real time through GPS and RFID systems to ensure the accuracy of the information. Combined with the order management system, the information of the waste paper products to be transported, including the destination, priority and time limit requirements, is obtained. By analyzing the transportation demand, the target site to be transported is determined. According to the time limit and priority of the transportation demand, the sites to be transported are sorted, and time-sensitive orders are given priority to ensure that high-priority transportation needs can be met in a timely manner. The Dijkstra algorithm or the A* algorithm is used to calculate the shortest path to each candidate site based on the current vehicle position and the geographical location of the site to be transported. The site with the shortest path and meeting the time requirements is selected as the next site. A dynamic path scheduling optimization engine is constructed, which can automatically adjust the transportation path according to real-time data. The engine should integrate GPS data, transportation demand and current traffic conditions to realize dynamic path calculation. The system should be connected to real-time traffic data (such as using Google Maps API or local traffic monitoring system), ensure that the dispatch engine can obtain the latest traffic conditions, including congestion, accidents and road closures, implement dispatch algorithms in the dispatch engine, such as genetic algorithms or particle swarm optimization (PSO), which can seek a balance between multiple objectives, optimize transportation paths and time costs, consider multiple optimization objectives, such as the shortest transportation time, the lowest transportation cost and maximize resource utilization, and achieve comprehensive optimization by setting weights for different objectives. The dynamic path dispatch engine is designed into multiple functional modules, including data acquisition module, path calculation module, dispatch decision module and user interface module. The modules interact with each other through API interfaces to ensure the flexibility and scalability of the system. Develop a user-friendly interface so that managers can view transportation status, dispatch suggestions and current path information in real time. The interface should provide a dynamic update function to ensure that users can obtain the latest dispatch information. During the transportation process, feedback data is collected in real time, including actual arrival time, transportation status and user satisfaction, and these data are fed back to the dispatch engine to continuously optimize the algorithm and dispatch strategy. Based on the collected feedback data, the dispatch engine can use machine learning technology to self-learn and adjust the dispatch algorithm to improve the intelligence level of path dispatch.

[0036] Step S6: Analyze the abnormal fault status of the real-time monitoring parameters of the waste paper pile, perform global collaborative monitoring based on the dynamic path scheduling optimization engine, and build a full-cycle collaborative monitoring model.

[0037] In this embodiment, real-time monitoring data of waste paper piles are collected, including RFID tag information, transportation status, temperature and humidity, location coordinates, etc., to ensure the accuracy and completeness of the data. Duplicate and invalid records are removed through data cleaning, historical transportation data and real-time data are integrated, a benchmark model is established, and historical data are used for statistical analysis to identify parameter ranges and change trends under normal conditions. Statistical methods (such as Z-score or IQR method) are used to detect anomalies in real-time monitoring parameters. According to the set threshold (such as Z-score ≥ 3 or IQR exceeds 1.5), anomalies are detected. times), identify outliers, apply machine learning algorithms (such as isolation forests or support vector machines) for more complex anomaly detection, train models using historical data, identify potential abnormal patterns, and monitor in real time whether new data deviates from normal patterns, classify detected anomalies, determine the type of fault state, such as transportation delays, environmental anomalies (such as excessive temperature and humidity), etc., establish a fault classification model to ensure that different types of faults can be identified and responded to in a timely manner, build a global collaborative monitoring system, integrate real-time monitoring data, anomaly detection results and dynamic path scheduling information, the system should have data integration, real-time processing and visualization functions, fuse data from different sources (such as RFID, GPS, environmental sensors) to form a unified monitoring view, use data fusion technology (such as Kalman filtering) to improve the accuracy and reliability of monitoring data, design an interface for the dynamic path scheduling optimization engine, and ensure that the monitoring system can access the scheduling engine in real time The system should automatically adjust the transportation route or transportation plan through the scheduling engine in time when an abnormality occurs. The real-time feedback mechanism ensures that the scheduling engine can respond to abnormal conditions and make optimization adjustments. The periodicity of the monitoring model is set according to the transportation cycle and monitoring needs of the waste paper pile. The model should support real-time monitoring and regular analysis to ensure that it can cover all stages of the entire life cycle. Key performance indicators (KPIs) for full-cycle collaborative monitoring are set, including transportation efficiency, fault response time, environmental impact, etc., to ensure that the monitoring model can fully reflect the status of the transportation process. The model is verified through historical data, and the performance of the monitoring model under different conditions is checked to ensure that the model can accurately identify abnormalities and adjust the transportation route in time. According to the monitoring results and user feedback, the model parameters and monitoring strategies are continuously optimized, and machine learning algorithms are introduced to self-learn the model so that it can adapt to the changing environment and needs.

[0038] In this embodiment, refer to Figure 2 , is a flowchart of detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Step S11: acquiring a multi-angle real-time monitoring image of the waste paper product pile to be processed; performing image region segmentation on the multi-angle real-time monitoring image to obtain a plurality of image regions; Step S12: Calculating the image brightness of each of the multiple image regions one by one to generate a brightness value for each region; Step S13: performing global average flatness calculation based on the brightness value of each area to obtain the average brightness value of the image; Step S14: performing brightness deviation area detection on multiple image areas based on the average brightness value of the image, and extracting the brightness deviation area; Step S15: performing local brightness enhancement processing on the brightness deviation area to obtain a local brightness enhanced image; Step S16: reconstructing the panoramic image of the local brightness enhanced image to construct a panoramic image model of the waste paper pile.

[0039] In this embodiment, multiple high-resolution cameras are installed around the waste paper product pile to ensure that a complete view of the paper pile can be captured from different angles. It is recommended to configure at least four cameras to cover different viewing angles. These cameras should have a real-time transmission function and be able to transmit the captured image data to the central processing unit. Monitoring software is written to set a timed shooting or trigger shooting mechanism to obtain image data from each camera in real time, ensuring that the time interval for image capture is short enough to timely reflect changes in the paper product pile (such as stacking height, shape changes, etc.). After image acquisition, each image is preliminarily processed, such as denoising, distortion correction, and exposure adjustment, to improve image quality. Processing algorithms (such as Gaussian blur, histogram equalization, etc.) are used to ensure that the acquired images have good analyzability. The acquired multi-angle images are stored in a cloud server or local storage system and classified and managed to ensure that they can be quickly accessed during subsequent processing. Image segmentation algorithms (such as threshold-based segmentation, region growing method, or semantic segmentation based on deep learning) are used to segment each image into multiple regions. The goal is to identify different parts of the paper pile and the background area, ensuring that the segmentation of each region can accurately reflect the characteristics of the paper product, which is helpful for subsequent brightness calculations. For each segmented image area, its brightness value is calculated. The brightness calculation formula can be used: L=0.299R+0.587G+0.114B, R, G, B The pixel values ​​of the red, green, and blue channels are recorded in the data structure for subsequent analysis. The calculated brightness values ​​of each region are summarized and stored in the database to form a data set that is easy to query and analyze. The brightness values ​​of all regions are extracted from the previous brightness calculation step to form a brightness value list to ensure the integrity and accuracy of the data for subsequent average calculation. The average brightness value of all regions is calculated using a simple arithmetic average method. The calculated average brightness value is recorded as the standard for global brightness. The rationality of the average brightness value is checked to ensure that it is within the expected range. If an abnormality is found, it is necessary to return to the previous step to check the accuracy of the data. The threshold of brightness deviation is set, which can usually be set to ±10% of the average brightness (adjustable according to specific circumstances). This standard will be used to determine whether there is a brightness deviation in each region. The brightness values ​​of each region are traversed and compared with the calculated average brightness value to identify the brightness deviation area. The ID or coordinates of the brightness deviation area are recorded for subsequent processing. The identified brightness deviation area information is stored in the database, and a visual image is generated to intuitively display the location and range of the deviation area. Select a suitable local brightness enhancement algorithm, such as histogram equalization, gamma correction or Adaptive Histogram Equalization (CLAHE), these methods can effectively improve the brightness of dark areas in the image, making it easier to observe and analyze. For each detected brightness deviation area, apply the selected enhancement algorithm to improve the local brightness. Note that the processed image should remain natural and avoid over-enhancement or artifacts. Save the processed local brightness enhanced image and compare it with the original image to ensure that the enhancement effect reaches the expected result. Generate a processing report to record the enhancement process and effect of each area for subsequent analysis. Collect all the regional images that have been locally enhanced and prepare for splicing to ensure that the overlapping parts between images are The parts can be accurately aligned for subsequent seamless stitching. Select appropriate image stitching algorithms, such as feature point matching methods (SIFT, SURF) or stitching methods based on optical flow. These algorithms can effectively identify overlapping areas and smoothly synthesize different images into panoramic images. Apply the selected stitching algorithm to seamlessly stitch the local enhanced images to generate a panoramic image of the waste paper pile. During the processing, attention should be paid to edge fusion to ensure the naturalness and coherence of the final image. Check the quality of the panoramic image to ensure that it can clearly show the overall situation of the waste paper pile. Save the final panoramic image model and provide it for subsequent analysis and utilization. .

[0040] In this embodiment, the specific steps of step S16 are: Perform image noise recognition on the local brightness enhanced image and mark multiple image noise points; Adaptively filter and optimize image noise to construct a filtered denoised image; Perform multi-angle visual analysis on the filtered and denoised images to identify the visual features of the images at each angle; Extracting an image at each angle based on the multi-angle real-time monitoring image; Recognize feature points of the waste paper pile for each angle of the image and mark multiple feature points of the waste paper pile; Calculate the spatial position coordinates of multiple feature points of the waste paper pile to generate the spatial position coordinates of each feature point; Perform panoramic feature matching on the spatial position coordinates of each feature point according to the image visual features at each angle, so as to generate multi-angle visual matching data of feature points of the waste paper pile; Based on the multi-angle visual matching data of feature points of waste paper pile, the filtered denoised image is reconstructed to construct a panoramic image model of waste paper pile.

[0041] In this embodiment, a threshold-based method, such as the Otsu method, is used to identify noise points. This method can effectively distinguish noise from real image information by automatically selecting a threshold. Other algorithms, such as a Gaussian mixture model (GMM), can also be used to model the image and identify high-probability noise points. Before noise point identification, the image is preprocessed, including conversion to a grayscale image, smoothing to reduce noise interference, and preliminary denoising using methods such as Gaussian filtering or median filtering to improve the accuracy of noise point identification. According to the set threshold, the processed image is analyzed pixel by pixel to identify pixels with brightness values ​​below the threshold and mark them as noise points. A connected domain marking algorithm (such as connected component analysis) is used to aggregate adjacent noise points to form a noise area. The identified noise area is marked to generate an image with noise marks. The noise area can be marked with different colors or shapes for subsequent viewing and processing, and the marking results are saved. , and generate a noise identification report, record the number and distribution of noise points, and provide a basis for subsequent filtering processing. Select algorithms such as adaptive median filtering or adaptive convolution filtering. These methods can dynamically adjust the filtering parameters according to the characteristics of the local area to effectively remove noise while retaining edge information. Set the filter window size to adapt to different types of noise. For each pixel, calculate the statistical characteristics of its surrounding neighborhood (such as mean value, standard deviation), and dynamically adjust the filtering parameters based on these characteristics. Traverse the entire image through the adaptive filter and apply filtering processing pixel by pixel. After the processing is completed, compare the image quality before and after denoising. Use indicators such as peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) to evaluate the filtering effect. According to the evaluation results, adjust the filtering parameters if necessary, and repeat the filtering process to ensure the best effect. Use feature extraction algorithms such as edge detection (Canny Edge detection), corner detection (Harris corner detection) or feature descriptors (SIFT, ORB, etc.). These algorithms can effectively extract key features in images and provide a basis for subsequent analysis. For the denoised images at each angle, apply the selected feature extraction algorithm to identify and extract important visual features in the image, classify and mark the extracted features for subsequent analysis, use clustering algorithms (such as K-means or DBSCAN) to cluster the extracted features, identify similar feature points, and classify them into different types (such as edge features, texture features, etc.) according to their distribution for subsequent analysis. Store the extracted visual feature information in a database, generate an analysis report, record the visual features of each angle in detail, and provide a visual display. Extract images of each angle from the stored multi-angle monitoring images. Classify and extract according to camera number or shooting time. Use image processing libraries (such as OpenCV) to read image files under the specified path, and preprocess each extracted image, such as adjusting the image size, normalizing the brightness, etc., to ensure the consistency and comparability of the image.Ensure that all images are analyzed at the same scale and format to facilitate subsequent feature point identification. Save the preprocessed images as a new image set to facilitate processing and analysis in subsequent steps. Record the transformation parameters and processing methods in the extraction process and form a document for subsequent reference. Select a suitable feature point detection algorithm, such as SIFT, SURF or ORB. These methods can effectively identify key points in the image and calculate their descriptors. These algorithms have good robustness to image rotation and scale changes and are suitable for multi-angle image analysis. Apply the selected feature point detection algorithm to the image at each angle to identify the key feature points. Record the coordinates and descriptor information of each feature point to form a feature point dataset. Mark the identified feature points on the original image using circles or crosses of different colors for subsequent viewing and analysis. Save the marked images to ensure that the feature point information at each angle is clearly visible. Store the identified feature point coordinates and descriptors in the database and generate an analysis report to record the feature points of the image at each angle. Information, use 3D reconstruction algorithms, such as triangulation or stereo vision methods, to calculate the 3D coordinates of feature points in space. It is necessary to pre-establish the internal and external parameter models of the camera in order to perform accurate spatial calculations, collect the coordinate information of each feature point in images at different angles, and the corresponding camera parameters, use the 3D reconstruction algorithm to convert the 2D coordinates of the feature points into 3D spatial coordinates, record the calculated 3D coordinates in the database, provide basic data for subsequent matching, verify the calculation results, check the accuracy and consistency of the spatial coordinates, generate an analysis report on the spatial coordinates of the feature points, record the spatial position of each feature point and its calculation method, and select a suitable feature matching algorithm, such as KNN Matching, FLANN matcher or descriptor-based matching algorithms, which can efficiently match similar feature points between different images, match feature points at each angle, find the same feature points at different viewing angles, record the matching results, including the successfully matched feature point pairs and their spatial coordinates, filter the matching results with the RANSAC algorithm, remove false matches to improve the accuracy and robustness of the matching, and ensure that the final retained matching results have a high degree of credibility, select a suitable panoramic image reconstruction algorithm, such as feature point-based image stitching or 3D reconstruction technology, ensure that the selected algorithm can handle multi-angle images and generate seamless panoramic images, stitch the images at various angles according to the matching results, use the matched feature points as references for geometric transformation, calculate the transformation matrix of each image, ensure that the reconstructed panoramic image can be naturally integrated, apply image processing technology for post-processing after the panoramic image is reconstructed, such as color correction, smoothing and edge fusion, to improve the overall quality of the image, generate the final panoramic image, and ensure that it shows the true situation of the waste paper pile, check the quality of the generated panoramic image, ensure its clarity and coherence, and meet the analysis requirements, save the final generated panoramic image model,And provide effective visual data support for subsequent analysis and application.

[0042] In this embodiment, refer to Figure 3 , is a flowchart of detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Step S21: performing a paper pile texture analysis on the waste paper pile panoramic image model to generate waste paper pile texture features; Step S22: performing multi-angle paper pile spectral feature mining on the waste paper pile panoramic image model to extract the waste paper pile spectral features; Step S23: quantifying the degree of waste paper stains according to the spectral characteristics of the waste paper pile, thereby generating a waste paper stain quantification value; Step S24: performing paper component identification on the spectral characteristics of the waste paper pile according to the texture characteristics of the waste paper pile to generate waste paper component characteristics; Step S25: performing waste paper feature classification according to waste paper component features and waste paper stain quantification values, thereby generating feature information of a waste paper pile to be processed; Step S26: Generate an RFID electronic code based on the characteristic information of the waste paper pile to be processed, and associate the RFID electronic code with the waste paper product pile to be processed for monitoring, so as to obtain real-time monitoring parameters of the waste paper pile.

[0043] In this embodiment, the gray level co-occurrence matrix (GLCM) method or the local binary pattern (LBP) method is used to extract texture features. These methods can effectively describe the texture information of the image and reflect the surface characteristics of the paper pile. GLCM can be used to extract texture features such as contrast, energy, homogeneity and correlation, and LBP can effectively capture the local texture structure. The panoramic image model of the waste paper pile is preprocessed, such as converted into a grayscale image and smoothed to reduce noise interference, and the GLCM is calculated. Or LBP, extract the texture features of the image. For GLCM, it is necessary to select appropriate offsets and directions to ensure that the texture features are fully captured. The extracted texture features are stored in the database to form structured data of the texture features of the paper pile. The distribution of texture features is displayed through visualization tools (such as Matplotlib), which is convenient for analysts to intuitively understand the texture features of the paper pile. Select appropriate spectral feature extraction algorithms, such as principal component analysis (PCA) or independent component analysis (ICA), for dimensionality reduction and extraction of effective spectral features. These methods can help identify reflectivity changes in different bands, thereby extracting representative spectral features. Perform spectral analysis on the panoramic image of the waste paper pile to extract reflectivity data in different bands. Multispectral imaging equipment can be used to obtain image data of the paper pile in different spectral regions, calculate the average reflectivity of each band, and generate spectral feature curves for subsequent analysis. The extracted spectral feature data is stored in the database to form structured information of spectral features. Spectral feature curves are drawn using visualization tools to facilitate intuitive display of spectral features in different bands. The stain recognition standard is set according to the spectral feature curve. Usually, the difference between the spectral reflectivity and the reflectivity of normal paper can be compared to determine the stain. Determine the degree of stains and set a stain threshold. Usually, when the reflectivity drops by more than a certain percentage, it is marked as a stain. Analyze each spectral band, calculate the reflectivity difference with normal paper, and obtain a quantitative value of the degree of stains. The weighted average method can be used to weight the reflectivity differences of different bands to generate a comprehensive stain quantification value. The calculated stain quantification value is stored in the database for subsequent analysis and query. Generate a stain degree quantification report, record the calculation method and results, and provide a basis for subsequent component identification. Select an appropriate machine learning model, such as support vector machine (SVM) or random forest (R F), to identify paper components. These models can classify based on texture features and spectral features. The models need to be trained. Sample data with known components are used as training sets. The extracted texture features and spectral features are combined into feature vectors for model training and prediction. The feature data are normalized to improve the training effect of the model. The model is trained using the prepared sample data. The model parameters are adjusted to improve the classification accuracy. Unlabeled waste paper pile data is input into the trained model to predict its paper components and generate component feature data. The classification criteria are set according to the paper component features and stain quantification values.The paper can be classified according to the type (such as book paper, newspaper, wrapping paper, etc.) and the degree of stain. For example, a threshold of the stain quantification value is set, and the paper exceeding the value is marked as a heavy stain category. Each waste paper pile is traversed and divided into different categories according to the set classification criteria. A decision tree or other classification algorithm can be used for automatic classification. The quantity and characteristic information of each category are recorded to form the result of characteristic classification. The classification result is stored in a database for subsequent query and management. A waste paper characteristic classification report is generated, and the characteristic information and classification criteria of each category are recorded in detail. The format and content of the RFID electronic code are set, which usually includes the characteristic information of the waste paper pile, the classification result and the unique identification code. The uniqueness of the code is ensured to avoid duplication. According to the characteristic information of the waste paper pile to be processed, the corresponding RFID electronic code is generated. The barcode generation library or RFID encoding tool can be used for generation. The generated code is associated with the characteristic information of the waste paper pile to form complete monitoring data. The generated RFID code is printed on an RFID tag, and the tag is attached to the waste paper pile for real-time monitoring. The readability and durability of the tag are ensured to adapt to the storage and transportation environment of the waste paper pile. RFID is deployed Reading equipment monitors the status and characteristic information of waste paper piles in real time, obtains various monitoring parameters associated with the code through the RFID reader, stores the monitoring data in the database, and analyzes it for subsequent management and decision support.

[0044] In this embodiment, the detailed implementation steps of step S3 include: Step S31: performing real-time positioning and tracking according to the real-time monitoring parameters of the waste paper pile, and extracting the real-time positioning and tracking data of the paper pile; Step S32: performing time-series transportation trajectory analysis on the real-time positioning and tracking data of the paper pile to generate a time-series transportation trajectory of the paper pile; Step S33: performing stop node identification on the time-series transportation trajectory of the paper pile to obtain a plurality of stop nodes; Step S34: Analyze the current transportation path based on multiple stop nodes to generate the current transportation path of the waste paper pile.

[0045] In this embodiment, RFID readers and GPS positioning devices are deployed on the transport vehicles of the waste paper piles to ensure that the location data can be obtained in real time. Configure the monitoring system and set the data collection frequency (such as once per second) to ensure real-time location updates. The RFID tag information of the waste paper pile is collected by the RFID reader, and the real-time positioning data is formed in combination with the geographic location data obtained by the GPS device. The system needs to have a data cache function to ensure that the data can be saved even when the network is unstable, and automatically uploaded to the cloud server or local database after the network is restored. Format the acquired real-time monitoring data to ensure that the data structure is consistent. Usually includes timestamps, GPS coordinates, RFID tag ID and other information. Perform preliminary data cleaning to remove outliers (such as invalid coordinates or duplicate data) to ensure the accuracy and reliability of the data. Store the processed real-time positioning tracking data in the database for subsequent analysis and query. Use visualization tools (such as map API) to display the real-time location of the waste paper pile to ensure that managers can intuitively understand the dynamic situation of the paper pile. Extract real-time positioning tracking data from the database and ensure that the data is sorted by timestamp for time series analysis. The data is processed in segments and the analysis time window (e.g., every minute or every hour) is set to facilitate subsequent trajectory analysis. The transportation trajectory of the waste paper pile is constructed based on the timestamp and GPS coordinate data. The coordinates of each time point are connected into a line to form an overall trajectory map. The trajectory is visualized using geographic information system (GIS) tools to display the transportation path. The continuity and changes of the trajectory are analyzed to identify the main movement stages and stop stages during the transportation process. The total distance and time of the transportation path are calculated to evaluate the transportation efficiency. A suitable stop node identification algorithm is selected, such as a threshold-based method or a clustering algorithm (e.g., DBSCAN). These methods can effectively identify points with small position changes within a specific time period and mark them as stop nodes. A stop time threshold (e.g., 5 minutes) is set to determine whether a location is a stop node. The time series transportation trajectory data is traversed to calculate the displacement change at each time point. When the displacement is less than the set threshold and the stop time exceeds the set threshold, it is marked as a stop node. For continuous stop nodes, merge them to ensure the accuracy and effectiveness of the stop nodes. The time, GPS coordinates, and stop duration of each stop node are recorded to form a stop node dataset. Analyze the distribution of stop nodes and identify common stop areas (such as transfer stations, warehouses, etc.). Store the identified stop node data in the database for subsequent query and analysis. Use visualization tools to display the location of the stop nodes in the transportation trajectory, so that managers can intuitively understand the stop situation during transportation. Define the current transportation path based on the identified stop nodes and movement trajectories. The current path should include the nearest stop node and movement direction to reflect the latest dynamics of the paper pile. Consider time factors to ensure the real-time and effectiveness of the path.Analyze the efficiency of the current transportation path, including the length of the path, the time required, and the impact of the stop nodes. Identify possible path optimization points (such as detours, excessive stops, etc.). Use path optimization algorithms (such as Dijkstra algorithm or A* algorithm) to optimize the path to ensure the efficiency of the transportation process. Record the information of the current transportation path in the database, including path points, stop nodes and their related data. Generate a transportation path analysis report, which records the characteristics of the transportation path, optimization suggestions and implementation effects in detail for subsequent management and decision-making. Use map visualization tools to display the current transportation path, including movement trajectories and stop nodes, so that managers can monitor and adjust in real time.

[0046] In this embodiment, step S4 includes the following steps: Step S41: Obtaining a waste paper product operation log; identifying all recycling sites according to the waste paper product operation log; Step S42: performing spatial distribution analysis on the recycling sites to generate spatial distribution data of recycling sites; Step S43: performing site distribution fitting on the recycling site spatial distribution data to construct a site distribution map; Step S44: dynamically mapping the transportation track of the site distribution map based on the current transportation path of the waste paper pile to construct a transportation track site map.

[0047] In this embodiment, an operation log is extracted from the waste paper product transportation management system. The log should include information such as timestamp, transportation vehicle ID, transportation route, stay time, recycling site ID, etc., to ensure the integrity and accuracy of the data, remove any abnormal or duplicate records, and format them. The collected operation logs are cleaned to ensure that all fields are complete, especially the recycling site ID and timestamp. If there is missing data, interpolation or other methods can be used to complete it to ensure the continuity of subsequent analysis. According to the recycling site information in the operation log, all independent recycling site IDs and related information (such as location, type, etc.) are extracted. The set deduplication method is used to ensure that each recycling site is recorded only once to form a list of recycling sites. The identified recycling site information is stored in a database, including site ID, name, location and related attributes. Visualization tools (such as GIS software) are used to display the distribution of recycling sites on the map to ensure that managers can intuitively view the layout of the recycling sites. The location information (such as longitude and latitude) of all recycling sites is extracted from the database, and spatial analysis is prepared to ensure that the coordinate system of the data is consistent. The WGS84 coordinate system is usually used. A spatial analysis tool (such as ArcGIS, QGIS or other GIS) is selected. 1. Use GIS software to analyze the spatial distribution of recycling sites, use spatial aggregation analysis (such as Getis-Ord G* statistic) and heat map analysis to identify dense areas of sites, input the recycling site data into the selected GIS software, perform spatial distribution analysis, calculate the distribution density of recycling sites, and identify high-density areas, generate spatial distribution data, including the number of sites, distribution density and aggregation degree, to provide a basis for subsequent site distribution fitting, store the analysis results in a database to form structured spatial distribution data, generate a recycling site spatial distribution analysis report, record the analysis methods, parameters and results, and provide a basis for subsequent steps, select a suitable spatial fitting model (such as Kriging interpolation, Voronoi diagram or rasterization method) to describe the spatial distribution characteristics of recycling sites, determine the parameters and range of the fitting model in order to perform accurate spatial analysis, input the spatial distribution data of recycling sites into the selected fitting model, perform spatial interpolation processing, calculate the distribution of recycling sites in each area, generate raster data or contour map of site distribution, and use GIS The software visualizes the fitting results and generates a site distribution map. The map should include the density of recycling sites in different areas and the distribution characteristics of the sites, ensuring the readability and intuitiveness of the map to facilitate subsequent analysis and decision-making. The generated site distribution map is stored in the database and the corresponding file format (such as PDF or image file) is provided for subsequent use. A site distribution fitting analysis report is generated to record the fitting model, method and results to provide a basis for subsequent steps. The current transportation path data of the waste paper pile is extracted from the real-time positioning and tracking system to ensure that the data includes timestamps, GPS coordinates and other information.Clean and format the transport path data for subsequent analysis, select appropriate mapping tools or GIS software, overlay the transport path data with the site distribution map, determine how to display the transport path (such as different colors or line types) to distinguish different states, overlay the current transport path data on the recycling site distribution map, generate a dynamic transport trajectory site map, ensure that the relationship between the transport path and the recycling site is clearly visible, mark the recycling sites that the transport path passes through on the map, and record the relevant stay time and transport status.

[0048] In this embodiment, step S5 includes the following steps: Step S51: Calculating the current position coordinates of the discarded paper pile based on the real-time positioning and tracking data of the paper pile; Step S52: identifying the next station on the transport track station map according to the current waste paper pile location coordinates, and marking the next operation station; Step S53: Calculate the operating distance of the next operating station to obtain the operating distance parameter of the next station; Step S54: Perform dynamic transportation route planning according to the next station operation distance parameter, thereby constructing a dynamic transportation planning path; Step S55: Perform real-time path scheduling optimization on the dynamic transportation planning path and build a dynamic path scheduling optimization engine.

[0049] In this embodiment, the location tracking data of the waste paper pile is extracted from the real-time positioning system, including GPS coordinates, timestamps and related RFID tag information. Ensure the integrity and accuracy of the data, remove any outliers, and ensure the reliability of subsequent calculations. If the acquired coordinate data uses different coordinate systems (such as geographic coordinate systems and plane coordinate systems), necessary coordinate conversions are required to ensure that all coordinate data are in the same coordinate system. Use appropriate geographic information system (GIS) tools or libraries (such as Proj4) for coordinate conversion. Calculate the current location of the waste paper pile based on the latest GPS coordinates (usually longitude and latitude). Ensure real-time updates to reflect the dynamic changes of the paper pile. Record timestamps and coordinate values ​​as basic data for subsequent analysis. Store the calculated coordinates of the current waste paper pile location in a database for subsequent query and analysis. Extract the identified recycling site information from the database, including the coordinates, name and related attributes of each site. Ensure the accuracy of all site data for effective matching and identification. Use the nearest neighbor algorithm or distance-based matching method to identify the nearest recycling site based on the current waste paper pile location coordinates. A distance threshold can be set to filter out stations that are too far away. Use a geographic distance formula (such as the Hafsin formula) to calculate the distance between the current paper pile location and each recycling station, and find the station corresponding to the minimum distance. Mark the next identified operation station and record its coordinates and related information. Store the information of the next identified operation station in the database and update the transportation trajectory map for subsequent analysis and monitoring. Highlight the next operation station on the visual interface to ensure that managers can clearly identify it. Choose an appropriate distance calculation method, usually using Euclidean distance or geographic distance (such as the Hafsin formula) to calculate the distance of the transportation path. Considering the actual road conditions during transportation, network analysis tools (such as GIS path analysis function) can be used for more accurate distance calculation. Calculate the distance between the current waste paper pile location coordinates and the identified next operation station coordinates. Record the calculated operation distance parameters and ensure that their units are consistent (such as meters or kilometers). Store the calculated operation distance parameters in the database for subsequent query and analysis. Generate an operation distance calculation report to record the calculation process, results and related information. Select an appropriate path planning algorithm, such as Dijkstra, A*, or genetic algorithm, to ensure that the optimal or near-optimal transport path can be found. Consider factors such as traffic conditions, road conditions, and transportation time during transportation to improve the accuracy of path planning. Use the current waste paper pile location and the next operation station location as the starting and end points, and call the path planning algorithm for calculation. Consider real-time traffic data and dynamically adjust the path planning to ensure that the optimal route is selected. Store the calculated dynamic transportation path in the database, including waypoints, total distance, and estimated arrival time.Use map visualization tools to display the planned transportation routes to ensure that managers can intuitively understand the transportation routes. Select a suitable scheduling optimization model, such as linear programming, integer programming, or heuristic algorithms, to optimize for real-time traffic conditions and transportation needs. Set optimization goals, such as minimizing transportation time, minimizing transportation costs, or maximizing transportation efficiency. Based on the current transportation route and real-time traffic data, call the scheduling optimization algorithm for calculation and adjust the transportation route to deal with emergencies (such as traffic congestion, accidents, etc.). Generate the optimized path and calculate the corresponding adjustment parameters (such as the total distance and estimated arrival time of the new path). Store the optimized dynamic transportation path in the database to ensure data update and accuracy. Use map tools to display the optimized path to ensure that managers can monitor the transportation situation in real time. Based on the above steps, build a dynamic path scheduling optimization engine that can receive data, analyze and adjust in real time. The engine should have real-time monitoring, path update and alarm functions to respond to changes in the transportation process in a timely manner.

[0050] In this embodiment, step S6 includes the following steps: Step S61: Analyze the abnormal fault state of the real-time monitoring parameters of the waste paper pile and extract the abnormal fault state data; Step S62: Calculate the fault location according to the abnormal fault state data to generate the abnormal fault occurrence location; Step S63: Visually mark the transport track site map according to the location where the abnormal fault occurs, thereby generating a fault site map; Step S64: Make emergency fault decisions based on the fault occurrence site map and build an emergency fault handling strategy; Step S65: Perform global collaborative monitoring on the emergency fault handling strategy and the dynamic path scheduling optimization engine to build a full-cycle coordinated monitoring model.

[0051] In this embodiment, various monitoring parameters of the waste paper pile are extracted from the real-time monitoring system, including location data, temperature, humidity, weight, vibration and other information, to ensure data integrity, and the monitoring parameters are integrated into a database for subsequent analysis. Statistical methods (such as Z-score detection), machine learning methods (such as isolation forest algorithm) or deep learning methods (such as autoencoders) are used for anomaly detection. These methods can effectively identify data points that deviate significantly from the normal state, set reasonable thresholds for different monitoring parameters, so as to quickly screen out abnormal values, analyze the collected real-time monitoring parameters, identify abnormal fault states that exceed the set thresholds, record the time when the abnormal state occurs, parameter values ​​and their types, generate a detailed abnormal state report for the detected abnormal situation, including possible fault causes and related data, extract abnormal fault state data from the database, ensure that the data includes timestamps, monitoring parameters and their corresponding waste paper pile location coordinates, record the specific characteristics of each abnormal state for subsequent analysis, and select a suitable fault location calculation method based on the abnormal state data. Usually, the addition method can be used. The weighted average method or the interpolation method based on neighboring points is combined with historical location data to calculate the location of the fault. If there are multiple abnormal records, you can consider using cluster analysis methods (such as K-means) to identify the hot spots where the fault occurs. Analyze the extracted abnormal state data, calculate the fault location coordinates corresponding to each abnormal state, ensure the accuracy of the position calculation, record the location of the fault, including longitude and latitude and the corresponding abnormal state characteristics, store the calculated fault location in the database, and generate a fault location analysis report, record the characteristics and related data of each fault location in detail, and provide a basis for subsequent emergency decision-making. Extract the transportation trajectory site map data from the database, including the location coordinates of all recycling sites and the corresponding transportation path information, ensure the accuracy of the map data, so as to carry out effective visual marking, and use GIS Use software or map visualization tools (such as Leaflet, Mapbox) for visual marking, select appropriate marking styles (such as color, icon) to clearly display the fault location, add marks on the transportation trajectory site map, identify the locations where all abnormal faults occur, and ensure that each mark contains specific information about the fault (such as abnormal type, occurrence time, etc.). Different colors or graphics can be used to distinguish different types of faults to enhance visualization effects. Establish an emergency fault handling strategy framework, determine the emergency response process and responsible persons, ensure that emergency measures can be quickly initiated after a fault occurs, and clarify the goals of emergency decision-making, such as resuming normal operations as soon as possible and reducing losses. Evaluate the location and nature of the fault and its impact on the transportation process, identify the transportation routes and recycling sites that may be affected, combine real-time monitoring data, evaluate the impact of the fault on the overall transportation efficiency, formulate corresponding handling measures, and formulate specific emergency handling strategies based on the fault impact assessment results.Including adjusting the transportation route, temporarily stopping operations, adding vehicles, etc., formulating clear operating steps to ensure that responsibilities are assigned to people, quickly responding to faults, storing the formulated emergency response strategies in the database, and generating emergency decision reports to record the decision-making process, basis and expected results, designing and building a collaborative monitoring platform, integrating functional modules such as real-time monitoring, fault analysis, and path planning, realizing instant data sharing and processing, ensuring that the platform has high availability and scalability to adapt to dynamic scheduling and emergency response needs, integrating data from various modules (such as real-time monitoring system, fault analysis module, path scheduling engine) to form a global monitoring view, ensuring that data is updated in real time so that managers can grasp the current status at any time, linking emergency response strategies with dynamic path scheduling optimization engines to ensure that transportation routes can be quickly adjusted in the event of a fault, and corresponding resource scheduling, setting monitoring indicators, real-time tracking of emergency response effects and transportation route adjustments, ensuring that corrections can be made in a timely manner, and storing the operating data of the full-cycle coordinated monitoring model in the database for subsequent analysis and optimization, generating collaborative monitoring reports, recording the model's operating conditions, emergency response effects and improvement suggestions, and providing decision support for subsequent management. ,

[0052] In this embodiment, a system for monitoring the full cycle of recycling of waste paper products is provided, which is used to execute the above-mentioned method for monitoring the full cycle of recycling of waste paper products, including: The image reconstruction module is used to obtain multi-angle real-time monitoring images of the waste paper product pile to be processed; perform local brightness enhancement processing on the multi-angle real-time monitoring images, and perform panoramic image reconstruction to build a panoramic image model of the waste paper pile.

[0053] The RFID encoding module is used to classify waste paper features in the waste paper pile panoramic image model, generate RFID electronic codes, and obtain real-time monitoring parameters of the waste paper pile; A real-time positioning module is used to perform real-time positioning and tracking based on real-time monitoring parameters of the waste paper pile, and to analyze the current transportation path, thereby generating the current transportation path of the waste paper pile; The site map module is used to obtain the waste paper product operation log; dynamically map the transportation track according to the waste paper product operation log and the current transportation path of the waste paper pile, and build a transportation track site map; The route scheduling optimization module is used to identify the next station based on the transport trajectory station map, perform real-time route scheduling optimization, and build a dynamic route scheduling optimization engine; The collaborative monitoring module is used to analyze the abnormal fault status of the real-time monitoring parameters of the waste paper pile, and to perform global collaborative monitoring based on the dynamic path scheduling optimization engine to build a full-cycle collaborative monitoring model.

[0054] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0055] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for monitoring the whole cycle of recycling of waste paper products, characterized in that: The following steps are involved: Step S1: acquiring a multi-angle real-time monitoring image of a waste paper product pile to be processed; performing local brightness enhancement processing on the multi-angle real-time monitoring image, and performing panoramic image reconstruction to construct a panoramic image model of the waste paper pile; Step S2: classifying waste paper features on the waste paper pile panoramic image model, generating RFID electronic codes, and obtaining real-time monitoring parameters of the waste paper pile; Step S3: performing real-time positioning and tracking according to the real-time monitoring parameters of the waste paper pile, and analyzing the current transportation path, thereby generating the current transportation path of the waste paper pile; Step S4: Obtain the waste paper product operation log; perform dynamic transportation track mapping according to the waste paper product operation log and the current transportation path of the waste paper pile, and construct a transportation track site map; Step S5: Identify the next station based on the transport track station map, perform real-time route scheduling optimization, and build a dynamic route scheduling optimization engine; Step S6: Analyze the abnormal fault status of the real-time monitoring parameters of the waste paper pile, perform global collaborative monitoring based on the dynamic path scheduling optimization engine, and build a full-cycle collaborative monitoring model.

2. The method for monitoring the whole cycle of recycling of waste paper products according to claim 1, characterized in that: The specific steps of step S1 are: Step S11: acquiring a multi-angle real-time monitoring image of the waste paper product pile to be processed; performing image region segmentation on the multi-angle real-time monitoring image to obtain a plurality of image regions; Step S12: Calculating the image brightness of each of the multiple image regions one by one to generate a brightness value for each region; Step S13: performing global average flatness calculation based on the brightness value of each area to obtain the average brightness value of the image; Step S14: performing brightness deviation area detection on multiple image areas based on the average brightness value of the image, and extracting the brightness deviation area; Step S15: performing local brightness enhancement processing on the brightness deviation area to obtain a local brightness enhanced image; Step S16: reconstructing a panoramic image of the local brightness enhanced image to construct a panoramic image model of the waste paper pile.

3. The method for monitoring the whole cycle of recycling of waste paper products according to claim 2, characterized in that: The specific steps of step S16 are: Perform image noise recognition on the local brightness enhanced image and mark multiple image noise points; Adaptively filter and optimize image noise to construct a filtered denoised image; Perform multi-angle visual analysis on the filtered and denoised images to identify the visual features of the images at each angle; Extracting an image at each angle based on the multi-angle real-time monitoring image; Recognize feature points of the waste paper pile for each angle of the image and mark multiple feature points of the waste paper pile; Calculate the spatial position coordinates of multiple feature points of the waste paper pile to generate the spatial position coordinates of each feature point; Perform panoramic feature matching on the spatial position coordinates of each feature point according to the image visual features at each angle, so as to generate multi-angle visual matching data of feature points of the waste paper pile; Based on the multi-angle visual matching data of feature points of waste paper pile, the filtered denoised image is reconstructed to construct a panoramic image model of waste paper pile.

4. The method for monitoring the whole cycle of recycling of waste paper products according to claim 1, characterized in that: The specific steps of step S2 are: Step S21: performing a paper pile texture analysis on the waste paper pile panoramic image model to generate waste paper pile texture features; Step S22: performing multi-angle paper pile spectral feature mining on the waste paper pile panoramic image model to extract the waste paper pile spectral features; Step S23: quantifying the degree of waste paper stains according to the spectral characteristics of the waste paper pile, thereby generating a waste paper stain quantification value; Step S24: performing paper component identification on the spectral characteristics of the waste paper pile according to the texture characteristics of the waste paper pile to generate waste paper component characteristics; Step S25: performing waste paper feature classification according to waste paper component features and waste paper stain quantification values, thereby generating feature information of a waste paper pile to be processed; Step S26: Generate an RFID electronic code based on the characteristic information of the waste paper pile to be processed, and associate the RFID electronic code with the waste paper product pile to be processed for monitoring, so as to obtain real-time monitoring parameters of the waste paper pile.

5. The method for monitoring the whole cycle of recycling of waste paper products according to claim 1, characterized in that: The specific steps of step S3 are: Step S31: performing real-time positioning and tracking according to the real-time monitoring parameters of the waste paper pile, and extracting the real-time positioning and tracking data of the paper pile; Step S32: performing time-series transportation trajectory analysis on the real-time positioning and tracking data of the paper pile to generate a time-series transportation trajectory of the paper pile; Step S33: performing stop node identification on the time-series transportation trajectory of the paper pile to obtain a plurality of stop nodes; Step S34: Analyze the current transportation path based on multiple stop nodes to generate the current transportation path of the waste paper pile.

6. The method for monitoring the whole cycle of recycling of waste paper products according to claim 1, characterized in that: The specific steps of step S4 are: Step S41: obtaining a waste paper product operation log; identifying all recycling sites according to the waste paper product operation log; Step S42: performing spatial distribution analysis on the recycling sites to generate spatial distribution data of recycling sites; Step S43: performing site distribution fitting on the recycling site spatial distribution data to construct a site distribution map; Step S44: dynamically mapping the transportation track of the site distribution map based on the current transportation path of the waste paper pile to construct a transportation track site map.

7. The method for monitoring the whole cycle of recycling of waste paper products according to claim 1, characterized in that: The specific steps of step S5 are: Step S51: Calculating the current position coordinates of the discarded paper pile based on the real-time positioning and tracking data of the paper pile; Step S52: identifying the next station on the transport track station map according to the current waste paper pile location coordinates, and marking the next operation station; Step S53: Calculate the operating distance of the next operating station to obtain the operating distance parameter of the next station; Step S54: Perform dynamic transportation route planning according to the next station operation distance parameter, thereby constructing a dynamic transportation planning path; Step S55: Perform real-time path scheduling optimization on the dynamic transportation planning path and build a dynamic path scheduling optimization engine.

8. The method for monitoring the whole cycle of recycling of waste paper products according to claim 1, characterized in that: The specific steps of step S6 are: Step S61: Analyze the abnormal fault state of the real-time monitoring parameters of the waste paper pile and extract the abnormal fault state data; Step S62: Calculate the fault location according to the abnormal fault state data to generate the abnormal fault occurrence location; Step S63: Visually mark the transport track site map according to the location where the abnormal fault occurs, thereby generating a fault site map; Step S64: Make emergency fault decisions based on the fault occurrence site map and build an emergency fault handling strategy; Step S65: Perform global collaborative monitoring on the emergency fault handling strategy and the dynamic path scheduling optimization engine to build a full-cycle coordinated monitoring model.

9. A full-cycle monitoring system for recycling waste paper products, characterized in that: The method for monitoring the whole cycle of recycling of waste paper products according to claim 1 comprises: An image reconstruction module is used to obtain multi-angle real-time monitoring images of the waste paper product pile to be processed; perform local brightness enhancement processing on the multi-angle real-time monitoring images, and perform panoramic image reconstruction to build a panoramic image model of the waste paper pile; The RFID encoding module is used to classify waste paper features in the waste paper pile panoramic image model, generate RFID electronic codes, and obtain real-time monitoring parameters of the waste paper pile; A real-time positioning module is used to perform real-time positioning and tracking based on real-time monitoring parameters of the waste paper pile, and to analyze the current transportation path, thereby generating the current transportation path of the waste paper pile; The site map module is used to obtain the waste paper product operation log; dynamically map the transportation track according to the waste paper product operation log and the current transportation path of the waste paper pile, and build a transportation track site map; The route scheduling optimization module is used to identify the next station based on the transport trajectory station map, perform real-time route scheduling optimization, and build a dynamic route scheduling optimization engine; The collaborative monitoring module is used to analyze the abnormal fault status of the real-time monitoring parameters of the waste paper pile, and to perform global collaborative monitoring based on the dynamic path scheduling optimization engine to build a full-cycle collaborative monitoring model.

Images