Material acceptance processing method and device, computer equipment and readable storage medium

Abstract

The invention provides a material acceptance processing method and device, computer equipment and a readable storage medium. The method comprises the steps that when a transport vehicle drives into and out of a target position, an identity label of the transport vehicle and materials transported by the transport vehicle are recognized respectively, and an entry record and an exit record are generated; an acceptance record formed by material acceptance of a weight house material acceptance system is obtained, and the acceptance record comprises identity information of a transport vehicle, transport material information and weight information of the transport vehicle; generating a material transportation record according to the entry record, the exit record and the acceptance record; judging whether the material transportation record meets a preset material transportation rule or not; and when the material transportation record does not meet the material transportation rule, outputting alarm information. According to the invention, in combination with an image recognition technology, vehicle identity recognition, material weighing and data linkage, closed-loop management and risk early warning of entry and exit are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of material acceptance, and particularly to a method, device, computer device and readable storage medium for material acceptance processing. Background Art

[0002] Driven by the concept of "Internet +", new generation information technologies such as Internet of Things technology, cloud computing technology, and big data technology are accelerating their integrated applications in the field of construction project management. As an important tool for the management of the entry and exit of construction site materials, the functions of the material acceptance system have gradually evolved from traditional manual records to digital operations such as automatic weighing, electronic document management, and data uploading.

[0003] Current material acceptance systems on the market can automatically complete the weighing of materials through a weighbridge when vehicles enter and exit the construction site, and synchronously upload the weighing data to the cloud for subsequent data statistics and management analysis by enterprises. Such systems have improved the efficiency of material acceptance to a certain extent and reduced the cost of manual intervention.

[0004] However, the existing material acceptance systems still have many management blind spots under the existing technology. Some key links rely on manual operations, lacking systematicness, closed-loop nature, and real-time warning capabilities, resulting in large gaps in aspects such as accuracy, authenticity, and prevention and control in on-site material management. Although the existing material acceptance systems already have a certain degree of automation capabilities, there are still obvious deficiencies in aspects such as data integrity, the correlation between images and data, risk identification capabilities, abnormal behavior control, and information closed-loop management.

[0005] Therefore, when implementing material acceptance, how to better meet the needs of refined management of modern engineering projects has become a technical problem urgently to be solved in this field. Summary of the Invention

[0006] The purpose of the present invention is to provide a method, device, computer device and readable storage medium for material acceptance processing to solve the above technical problems in the existing technology.

[0007] On the one hand, to achieve the above purpose, the present invention provides a method for material acceptance processing.

[0008] The method for material acceptance processing includes: when the transport vehicle enters and exits the target location, respectively identifying the identity identifier of the transport vehicle and the materials transported by the transport vehicle, and generating an inbound record and an outbound record; obtaining the acceptance record formed by the material acceptance system in the weighing house, wherein the acceptance record includes the identity information of the transport vehicle, the transported material information and its weight information; generating a material transport record according to the inbound record, the outbound record and the acceptance record; judging whether the material transport record meets the preset material transport rules; when the material transport record does not meet the material transport rules, outputting an alarm message.

[0009] Further, the step of identifying the materials transported by the transport vehicle includes: extracting a plurality of image frames from the real-time monitoring video stream; inputting each of the image frames into a preset vehicle recognition model for vehicle target detection, and outputting a vehicle recognition result, wherein the vehicle recognition result includes a first category label, a bounding box and a first confidence level, and the first category label includes the vehicle body, the carriage and the tank body of the tanker; performing target tracking processing on the vehicle recognition results of each of the image frames, assigning a unified tracking identifier to the same transport vehicle, and obtaining a target trajectory corresponding to the tracking identifier, wherein the target trajectory includes the bounding box and the frame number; among the vehicle recognition results of the same target trajectory, screening the image frames with the first confidence level greater than a preset threshold and the first category label being the carriage and the tank body as target image frames; cropping the target image frames according to the bounding boxes in the vehicle recognition results of the target image frames to obtain a target image area; inputting the target image area into a preset material recognition model for material recognition, and outputting a material recognition result, wherein the material recognition result includes a second category label and a second confidence level, and the second category label includes the tank body, an empty carriage, steel bars, waste steel bars, sand and gravel, formwork, wooden square timbers and scaffolding; performing aggregation processing on the material recognition results of multiple image frames with the same tracking identifier according to the second confidence level, and outputting the recognition result of the materials transported by the transport vehicle.

[0010] Further, the steps of generating the inbound record and the outbound record include: determining the tracking identifier of the transport vehicle according to the identity identifier; extracting the frame numbers in the target trajectory corresponding to the tracking identifier to obtain an image set; generating the inbound record according to the identity identifier of the transport vehicle, the recognition result of the materials when entering the target location, the image set and the entry time; and generating the outbound record according to the identity identifier of the transport vehicle, the recognition result of the materials when exiting the target location, the image set and the exit time.

[0011] Further, the step of determining whether the material transportation record meets the preset material transportation rules includes: when the inbound record and the outbound record in the material transportation record match, and there is no acceptance record in the material transportation record, it is determined that the transport vehicle has a risk of not being weighed.

[0012] Further, the step of determining whether the material transportation record meets the preset material transportation rules includes: when the material transportation record has an acceptance record and does not have an inbound record and / or an outbound record, it is determined that the transport vehicle has a risk of false material receipt.

[0013] Further, the step of determining whether the material transportation record meets the preset material transportation rules includes: when the inbound record and the acceptance record in the material transportation record are consistent, and the material identification result in the outbound record indicates that the transport vehicle does not leave the site empty, it is determined that the transport vehicle has a risk of stealing materials.

[0014] Further, the step of determining whether the material transportation record meets the preset material transportation rules includes: when the outbound record in the material transportation record is inconsistent with the acceptance record, and the material identification result in the outbound record indicates that the transport vehicle does not leave the site empty, it is determined that the transport vehicle has a risk of scrap steel not being weighed.

[0015] On the other hand, to achieve the above object, the present invention provides a material acceptance processing device.

[0016] The material acceptance processing device includes: a processing module for respectively identifying the identity identifier of the transport vehicle and the material transported by the transport vehicle when the transport vehicle enters and exits the target position, and generating an inbound record and an outbound record; an acquisition module for acquiring an acceptance record formed by the material acceptance system in the weighing house, where the acceptance record includes the identity information of the transport vehicle, the transported material information and its weight information; a generation module for generating a material transportation record according to the inbound record, the outbound record and the acceptance record; a judgment module for judging whether the material transportation record meets the preset material transportation rules; and an early warning module for outputting an alarm message when the material transportation record does not meet the material transportation rules.

[0017] On the other hand, to achieve the above object, the present invention further provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor implements the steps of the above method when executing the computer program.

[0018] On the other hand, to achieve the above object, the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above method are implemented.

[0019] The material acceptance processing method, device, computer equipment and readable storage medium provided by the present invention respectively identify the vehicle identity and the information of the transported materials when the transport vehicle enters and exits the target location, and generate an inbound record and an outbound record. Obtain the acceptance record formed by the weighbridge material acceptance system, which includes vehicle identity, transported material type and its weight. Structurally integrate the inbound record, outbound record and acceptance record to form a complete material transport record. Based on the transport rules preset by the system, perform consistency verification, process integrity judgment and risk assessment on the material transport record. If a transport record that does not meet the rules is identified, it is considered that there is an abnormal behavior, and the system automatically generates an alarm message and pushes it. Through the present invention, with the vehicle identity as the core identifier, three types of records, namely inbound, acceptance and outbound, are associated to construct a complete material transport chain with clear time clues, identity information and image evidence, realizing the whole process traceability and retrievability of material transfer, and realizing the full-link structured closed loop of the inbound and outbound processes. By setting rules to automatically identify risk scenarios, without relying on manual backtracking and verification, early warning, proactive discovery and timely intervention are achieved, realizing intelligent identification of abnormal behaviors and improving the on-site risk control ability. At the same time, the system automatically completes data collection, matching, judgment and warning, greatly reducing the workload of manual verification, statistics and comparison, and improving the automation level and transparent management ability of engineering project material management. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings: Figure 1 It is a flowchart of the material acceptance processing method provided in Embodiment 1 of the present invention; Figure 2 It is an application flowchart of the material acceptance processing method provided in Embodiment 1 of the present invention; Figure 3 It is a block diagram of the material acceptance processing device provided in Embodiment 2 of the present invention; Figure 4 It is a hardware structure diagram of the computer equipment provided in Embodiment 3 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] In order to make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. 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. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0022] Regarding the material acceptance system in the related art, the inventor has found the following problems.

[0023] First, it is difficult to prevent and control false material receipts. In some projects, there is a problem of creating false acceptance documents by weighing the empty truck and then weighing the truck with the material. The current system relies on paper signatures or document image verification and cannot effectively verify whether the materials have actually arrived.

[0024] Second, the act of stealing materials is highly concealed. After materials are lost or missing, the project side can only rely on checking access control surveillance videos for post-event investigation, which is a lagging process and cannot effectively identify the materials in the truck bed.

[0025] Third, the departure of waste steel bars is not standardized. As recyclable resources in engineering projects, the transfer and departure of waste steel bars lack a standardized management process, and irregularities such as not being weighed, not being approved, or leaving the site with a non-empty truck are likely to occur.

[0026] Fourth, the evidence chain is incomplete and the risk is difficult to close. In the current system, records such as vehicle entry, weighing, and departure do not form a structured closed-loop chain, and images and data are not effectively associated. As a result, when disputes or problems occur, it is difficult to trace and define the responsible party.

[0027] Fifth, the discovery of risk events is lagging and the handling is inefficient. Problems such as the driver leaving the site without weighing back, the materials being put into use without being accepted as required, and the failure to timely warn of risk vehicles may exist normally on-site. There is a lack of a timely discovery and handling mechanism, which is likely to cause management loopholes.

[0028] Therefore, this application proposes a material acceptance processing method, device, computer device, and readable storage medium. The specific embodiments are described in detail as follows.

[0029] Embodiment 1 Embodiment 1 of the present invention provides a material acceptance processing method, which is applicable to the material in-and-out management system at the construction site. Through this method, combined with image recognition technology, vehicle identity recognition, material weighing, and data linkage, closed-loop management of entry and exit and risk warning are realized, so as to better meet the needs of refined management of modern engineering projects. Specifically, Figure 1 is a flowchart of the material acceptance processing method provided in Embodiment 1 of the present invention, as Figure 1As shown, the material acceptance processing method provided by this embodiment includes steps S101 to S105 as follows.

[0030] Step S101: When the transport vehicle enters and exits the target location, respectively identify the identity identifier of the transport vehicle and the materials transported by the transport vehicle, and generate an entry record and an exit record.

[0031] Specifically, the target location can be the gate of the construction site. When the transport vehicle enters the gate of the construction site, the system obtains the license plate information of the vehicle through the license plate recognition camera deployed at the entrance as the vehicle identity identifier. At the same time, using the AI image recognition model, the material in the load area such as the vehicle hopper or tank is identified to recognize the type of materials transported (such as steel bars, sand and gravel, formwork, etc.) as the material information for this entry. The system binds the vehicle identity identifier with the recognized material information and records the entry time to generate an entry record.

[0032] When the transport vehicle exits the project site after unloading, weighing or other processes, the system identifies its license plate and hopper status again, records the exit time and the material recognition result, and generates an exit record. The entry and exit images, recognition results and timestamps are stored together for subsequent comparison and auditing.

[0033] Step S102: Obtain the acceptance record formed by the material acceptance system in the weighing house.

[0034] Among them, the acceptance record includes the identity information of the transport vehicle, the transport material information and its weight information. During the vehicle weighing process, the weighing house system automatically recognizes the license plate number, binds the vehicle information with the weighing data, and generates an acceptance record. This record contains the vehicle identity information (license plate number), the transport material information (such as concrete, scrap steel, etc.) and its weight information, and may also include fields such as the acceptance time and the acceptance method (automatic weighing or manual entry). The acceptance record is uploaded to the cloud database in real time through the system and participates in subsequent data integration.

[0035] Step S103: Generate a material transport record according to the entry record, the exit record and the acceptance record.

[0036] Specifically, the system uses the license plate number as the associated primary key to match the entry record, the exit record and the acceptance record of the same vehicle, and summarizes these three records to form a structured material transport record as the complete process data of this transport task.

[0037] Step S104: Determine whether the material transport record meets the preset material transport rules.

[0038] The system presets a set of material transportation rules to identify abnormal behavior, such as: whether the acceptance record is missing; whether the entry and exit recognition images are the same license plate; whether the acceptance materials are consistent with the AI recognition results; whether the exit truck bed is empty; whether the transportation time is too short or too long.

[0039] The system matches each material transportation record with the above rules one by one to determine whether there are any exceptions in the record.

[0040] Step S105: When the material transportation record does not meet the material transportation rules, an alarm message is output.

[0041] When the judgment result is "not satisfied", the system will automatically trigger the early warning mechanism and generate an alarm message, including vehicle identity, risk type (such as failure to weigh, false material collection, stolen material, etc.), entry and exit time, corresponding image evidence, etc. The alarm information can be pushed to the site administrator, material clerk or company management platform in real time through the Web or mobile APP for timely intervention and processing.

[0042] In the material acceptance processing method provided in this embodiment, data structuring and abnormality identification based on identity recognition, image recognition and weighing information are realized for the whole process of material flow of transport vehicles entering and leaving the project. When the transport vehicle enters and leaves the target location, the vehicle identity and the material information transported are respectively identified to generate entry records and exit records. The acceptance record formed by the material acceptance system of the weighing room is obtained, and the acceptance record includes the vehicle identity, the type of transported material and its weight. The entry record, exit record and acceptance record are structured and integrated to form a complete material transportation record. Based on the transportation rules preset by the system, the material transportation record is subjected to consistency verification, process integrity judgment and risk assessment. If a transportation record that does not meet the rules is identified, it is considered that there is abnormal behavior, and the system automatically generates and pushes an alarm message. The material acceptance processing method provided in this embodiment is adopted, with the vehicle identity as the core identifier, and the three types of records of entry, acceptance and exit are associated to construct a complete material transportation chain with clear time clues, identity information and image evidence, so that the whole process of material flow can be checked and traced, and the full-link structured closed loop of the entry and exit process is realized. By setting rules to automatically identify risk scenarios, without relying on manual backtracking and verification, early warning, active discovery, and timely intervention can be achieved, realizing intelligent identification of abnormal behaviors and improving on-site risk control capabilities. At the same time, the system automatically completes data collection, matching, judgment, and early warning, greatly reducing the workload of manual verification, statistics, and comparison, and improving the automation level and transparent management capabilities of project material management.

[0043] Optionally, in one embodiment, the step of identifying the material transported by the transport vehicle includes: extracting a plurality of image frames from the real-time monitoring video stream; inputting each image frame into a preset vehicle recognition model for vehicle target detection, and outputting a vehicle recognition result, where the vehicle recognition result includes a first category label, a bounding box, and a first confidence level, and the first category label includes the vehicle body, the hopper, and the tank body of the tanker; performing target tracking processing on the vehicle recognition results of each image frame, assigning a unified tracking identifier to the same transport vehicle, and obtaining a target trajectory corresponding to the tracking identifier, where the target trajectory includes a bounding box and a frame number; in the vehicle recognition results of the same target trajectory, screening out the image frames with the first confidence level greater than a preset threshold and the first category label being the hopper and the tank body as target image frames; cropping the target image frames according to the bounding boxes in the vehicle recognition results of the target image frames to obtain a target image area; inputting the target image area into a preset material recognition model for material recognition, and outputting a material recognition result, where the material recognition result includes a second category label and a second confidence level, and the second category label includes the tank body, an empty hopper, steel bars, waste steel bars, sand and gravel, formwork, wooden squares, and scaffolding; aggregating the material recognition results of multiple image frames with the same tracking identifier according to the second confidence level, and outputting the recognition result of the material transported by the transport vehicle.

[0044] In this embodiment, the accuracy and stability of the material recognition of the transport vehicle are further optimized. In view of the large number of image frames, perspective changes, and interference of some images during the process of the transport vehicle entering and leaving the site, a preferred material acceptance processing method is proposed. Specifically, when the transport vehicle passes through the monitoring area at the entrance and exit of the construction project, continuous video stream images are extracted from the high-definition cameras deployed on the passage, and multiple image frames are automatically extracted from the video stream. Optionally, 3 to 5 frames are collected per second to ensure that the moving vehicle is completely captured. The above-mentioned multiple image frames are input into a preset vehicle recognition model for vehicle detection. Optionally, the vehicle recognition model is a multi-category target detection network trained based on YOLOv5, and the output vehicle recognition result includes a category label, a bounding box, and a confidence level. Among them, the category label is marked as "vehicle body", "hopper", or "tank body"; the bounding box is used to locate the coordinate position of the target area in the image; the confidence level represents the credibility score of the target recognition.

[0045] After obtaining the vehicle recognition results corresponding to each image frame, perform object tracking processing on the vehicle recognition results in all image frames. Optionally, use the SORT (Simple Online Realtime Tracking) or DeepSORT tracking algorithm to associate the detection results belonging to the same transport vehicle according to the bounding box motion trajectory and visual features, and assign a unified tracking identifier Track ID to form an object trajectory. This object trajectory contains the bounding box positions and frame numbers of this transport vehicle in all image frames, representing the movement process of the transport vehicle in the image sequence.

[0046] Among all the detection results corresponding to this object trajectory, filter out the image frames whose first category label is "dump truck body" or "tank body" and the first confidence level is greater than a set threshold (e.g., 0.85). Such image frames are considered to have a good visual angle of the material and are selected as target image frames. For the above target image frames, according to the bounding boxes in the vehicle recognition results, crop the areas of the dump truck body or tank body in the image to obtain the corresponding target image areas for subsequent material recognition.

[0047] Then, input the target image area into a preset material recognition model. This material recognition model can use a convolutional neural network classifier trained based on the ResNet architecture, and the supported recognition categories include but are not limited to: tank body, empty dump truck body, steel bars, waste steel bars, sand and gravel, formwork, wooden square timbers, and scaffolding, etc. The material recognition model outputs the material recognition results, specifically including the category label and the confidence level. Among them, the category label here represents the specific material type identified, and the confidence level represents the credibility score of this recognition result.

[0048] Finally, perform aggregation processing on the material recognition results of multiple target image frames under the same Track ID. Optionally, during the aggregation processing, the category with the highest frequency of occurrence can be taken as the final recognition result, or the average confidence level of each category can be statistically calculated, and the category with the highest confidence level is selected. Finally, output the recognition result of the material transported by this transport vehicle and record it in the entry record and the exit record accordingly.

[0049] Using the material recognition method provided in this embodiment, by performing unified tracking and fusion processing on multiple image frames, it avoids the problem of misjudgment caused by interference such as occlusion and blurring in a single image frame, can enhance the stability of recognition, and at the same time avoids the problem of repeated recognition and repeated processing of the same vehicle in consecutive frames, improving the utilization efficiency of system resources. Finally, by fusing and judging the material category based on the recognition results, it effectively improves the recognition ability of materials, improves the recognition accuracy and reliability, and is applicable to typical engineering site environments such as multiple vehicles entering simultaneously, complex backgrounds, and overlapping materials.

[0050] Optionally, in one embodiment, the steps of generating an entry record and an exit record include: determining a tracking identifier of the transport vehicle according to the identity identifier; extracting the frame numbers in the target trajectory corresponding to the tracking identifier to obtain an image set; generating an entry record according to the identity identifier of the transport vehicle, the recognition result of the material when entering the target position, the image set, and the entry time; and generating an exit record according to the identity identifier of the transport vehicle, the recognition result of the material when exiting the target position, the image set, and the exit time.

[0051] In this embodiment, the generation methods of the entry record and the exit record during the entry and exit process of the transport vehicle are further refined, ensuring the consistent binding of images and recognition information, and improving the integrity and traceability of the records. When generating the entry record and the exit record, determine the tracking identifier of the transport vehicle according to the identity identifier. When the transport vehicle enters or exits the target position of the project site, first extract the identity identifier of the vehicle, that is, the license plate number, through the license plate recognition algorithm. Combining with the target tracking system, retrieve whether there is a vehicle trajectory matching the identity identifier in the current video frame sequence. If it exists, directly bind the tracking identifier (Track ID) corresponding to the license plate to it; if it does not exist, generate a new tracking identifier for the license plate and start the tracking process to ensure that the subsequent image frame data can be aggregated and processed. Then extract the frame numbers in the target trajectory corresponding to the tracking identifier to obtain an image set. From the video stream, based on the tracking identifier corresponding to the transport vehicle, extract its complete trajectory information. The trajectory contains multiple image frames and corresponding frame numbers. According to the set rules (such as 1 frame per second, select 1 frame every 3 frames, etc.), select several key frame images to form an image set for binding to the entry record or the exit record. The image set can truly reflect the visual information during the vehicle's entry or exit process, providing an image basis for subsequent verification, traceability, and anomaly comparison operations. When the vehicle enters the target position, execute the material recognition model to obtain the recognition result of the material in the vehicle's hopper or tank (such as steel bars, formwork, etc.), and record the entry timestamp recognized by the system. Subsequently, combine the following information to generate an entry record: identity identifier (license plate number), tracking identifier (Track ID), entry time, material recognition result when entering, and image set during the entry process. Among them, image confidence, camera number, recognition device number, etc. can be used as optional fields. The entry record will be stored in the database in a structured manner, and a unique record ID will be generated for subsequent data matching.

[0052] Similar to the inbound record, when the vehicle exits the target location, license plate recognition and material recognition are performed again. The material situation and the exit timestamp of the vehicle are recorded, and the image frames during the exit period are extracted based on the tracking identifier to form an exit image set. Subsequently, the following information is used to generate an exit record: identification (license plate number), tracking identifier (Track ID), exit time, material recognition result at the time of exit, and the image set during the exit process. This exit record is also stored in a structured manner, corresponding one-to-one with the inbound record, providing basic data for subsequent construction of a complete transportation chain (i.e., material transportation record).

[0053] In this embodiment, by introducing tracking identifier management and structured binding of the image set during the generation of inbound and outbound records, the target tracking technology is used to ensure that the extracted image frames truly reflect the motion state of the same transportation vehicle, preventing image mismatch or missed capture, and enabling accurate binding of the recorded images and events. Both inbound and outbound records are bound to the corresponding image set and timestamp, providing a verifiable image basis for subsequent auditing, risk control, or problem backtracking operations, enhancing the integrity and traceability of the evidence chain. By logically binding image recognition, trajectory tracking, and identity extraction, it is ensured that each record has accurate, complete, and structured identifiers and content, guaranteeing the consistency of material recognition and license plate identity. The material category, image information, and entry and exit times included in the inbound and outbound records provide key inputs for judging risks such as unauthorized pulling, false receipt, or non-weighing, providing data support for subsequent rule judgment and alarm output.

[0054] Optionally, in one embodiment, the step of judging whether the material transportation record meets the preset material transportation rules includes: when the inbound record and the outbound record in the material transportation record match, and there is no acceptance record in the material transportation record, it is determined that there is a risk of non-weighing for the transportation vehicle.

[0055] Specifically, first, field verification is performed on the generated material transportation record. If it is found that the material transportation record contains both valid inbound and outbound records, but no acceptance record corresponding to the license plate number can be matched in the database, it is determined that the material weighing process has not been completed for this transportation process. This risk type is attached to the transportation record and pushed to the administrator terminal, including the license plate number, entry and exit times, photos, material categories, etc. When judging the validity of the inbound record and the outbound record, it includes judging whether the license plate recognition is successful, whether the image acquisition time exists, whether the image recognition result is legal, whether the image source matches the position of the door camera, whether the entry time is earlier than the exit time, whether consecutive trajectory Track IDs are assigned to the same license plate for entry and exit, and so on.

[0056] Optionally, in one embodiment, the step of determining whether a material transportation record meets a preset material transportation rule includes: when there is an acceptance record in the material transportation record and there is no inbound record and / or outbound record, it is determined that there is a risk of false receipt of materials by the transportation vehicle.

[0057] Specifically, if it is detected that there is an acceptance record in a certain material transportation record but the inbound record and / or outbound record is missing, it means that the inbound and outbound images or trajectories of the corresponding license plate number cannot be found, and this acceptance record is not associated with the actual transportation process. There may be behaviors such as manual entry, supplementary entry, and fabrication. Attach this risk type to the transportation record and push it to the administrator terminal.

[0058] Optionally, in one embodiment, the step of determining whether a material transportation record meets a preset material transportation rule includes: when the inbound record and the acceptance record in the material transportation record are consistent, and the material identification result in the outbound record indicates that the transportation vehicle does not leave the site empty, it is determined that there is a risk of stealing materials by the transportation vehicle.

[0059] Specifically, compare the material identification result in the inbound record with the content of the acceptance record. If the two are consistent, it means that the vehicle was loaded with materials when it entered the site and the weighing operation has been completed. However, if in the outbound record image, the AI image recognition model determines that there are still obvious materials (such as steel bars, formworks, etc.) in the hopper or tank body, it means that the vehicle left the site in a non-empty state and there may be a behavior of privately pulling materials. For example, if the outbound image recognition category is not an empty hopper or tank body, and the recognition confidence is greater than the threshold, and the material category does not belong to the category of materials that can remain, then output a risk warning for stealing materials.

[0060] Optionally, in one embodiment, the step of determining whether a material transportation record meets a preset material transportation rule includes: when the outbound record in the material transportation record is inconsistent with the acceptance record, and the material identification result in the outbound record indicates that the transportation vehicle does not leave the site empty, it is determined that there is a risk that the waste steel has not been weighed.

[0061] Specifically, if the image recognition result in the outbound record of the transportation vehicle is waste steel, but no weighing data on waste steel can be found in the acceptance record, or the material category in the acceptance record is significantly inconsistent with the outbound image, it is determined that the waste steel has not been weighed. For example, if the outbound image recognition category is waste steel and the confidence is greater than the threshold, and the material type does not appear in the acceptance record, or the recorded weight is 0, then output the risk that the waste steel has not been weighed. This mechanism can effectively control the loss of waste steel and illegal handling behaviors.

[0062] Through the material identification method provided by the above embodiments, when judging whether the material transportation record meets the preset material transportation rules, automatic identification of high-frequency abnormal scenarios is realized, including non-weighing, false acceptance, stealing materials, and non-weighing of waste steel bars, etc. The risk types are refined to improve the accuracy and interpretability of early warnings, facilitate project managers to quickly locate problem vehicles and problem documents, strengthen the integration ability of image recognition and data structure, ensure that each risk event has a verifiable image evidence chain, and form a digital platform driven by data, rules, and closed-loop management.

[0063] Among them, when a suspected behavior of false material receipt occurs in a project, the system will automatically analyze and identify it. Managers can always keep track of suspicious behaviors and trace back to the photos and responsible persons on the site for follow-up management to ensure that every truckload of materials actually arrives at each project.

[0064] The records of vehicles entering and leaving the main gate are automatically associated and analyzed with the acceptance records to timely master the concrete inspection rate and avoid replacing full inspection with sampling inspection; for vehicles leaving the site without weighing back, according to the business rules, an early warning can be broadcast on the spot to guide the vehicle to weigh back, prevent the vehicle from leaving without weighing back, and avoid cost losses caused by getting out of control.

[0065] Through the AI algorithm, 24-hour all-weather defense and monitoring are realized to identify whether the vehicle carries steel bars when leaving the site and whether the truck hopper is blocked to prevent theft, automatically identify the number of times waste steel bars leave the site, analyze whether they are weighed, master the actual situation of waste steel bar treatment on the site, and standardize the departure.

[0066] An acceptance evidence chain that links business data and real image photos in all links of entering the site, acceptance, and leaving the site is constructed to form a comprehensive monitoring, and there is evidence to rely on in case of problems.

[0067] Through the built-in rules, once risks such as vehicles not weighing back, not weighing the gross weight, concrete mixer trucks not being weighed, and stealing steel bar materials are identified on the site, an early warning will be broadcast immediately, the risk vehicles will not be let go, and at the same time, an early warning notice will be sent to relevant personnel for handling, realizing a complete closed-loop of "instant identification and early warning - rapid response and handling - post-event control with a handle" for on-site risks.

[0068] Optionally, Figure 2 is the application flowchart of the material acceptance processing method provided in the first embodiment of the present invention, as Figure 2As shown, in one embodiment, when the transport vehicle enters the project gate, the vehicle identification camera captures and identifies the license plate number, that is, the license plate recognition is performed on the image of the front of the transport vehicle to obtain the license plate information of the corresponding vehicle. Then, the AI camera captures and identifies the material. Based on the hardware installed and deployed on the project site, when the transport vehicle enters the project gate, the image of the vehicle body and the transport material is AI algorithm material recognition, and the corresponding identification is added to it in the system background. The detection and tracking algorithm is applied to detect and track the vehicle in the video screen, and the detection categories are divided into: vehicle as a whole, tanker tank body, and truck bed. Then the image classification algorithm is applied to realize the identification of the material type in the truck bed, and the tanker tank body is secondary verified. Specifically, the truck bed and tank body detected above are cut out. The categories preset by the material recognition algorithm are: tanker tank body, empty truck bed, steel bar, scrap steel bar, sand and gravel, template, wood, scaffolding, and others. The input of this material recognition algorithm is the cutout area, and the output is the corresponding category label and confidence. Finally, the results output from the first two stages are post-processed to improve the accuracy. From the moment the vehicle enters the video screen, the results are saved every second. The categories with confidence levels greater than a preset threshold are selected from the multiple results. Then, the category results are voted on and the category with the most votes is taken as the final recognition result.

[0069] When the vehicle enters the gate, the license plate of each weighed vehicle is recognized, the license plate information of the corresponding weighed vehicle is obtained, and the material acceptance is carried out based on the license plate information of each weighed vehicle, and the acceptance document is generated. Alternatively, the license plate information of the corresponding vehicle is entered on the mobile phone, and the material acceptance is carried out for each material vehicle, and the acceptance document is generated to complete the material delivery and receipt on the project site.

[0070] When the vehicle leaves the project gate, the vehicle recognition camera captures and identifies the license plate number, and the AI camera captures and identifies the material, and also adds a corresponding mark to it in the system background.

[0071] The system automatically matches entry and exit records with material delivery and receipt records. That is, based on the vehicle entry and exit records, and acceptance document records, the system connects the data of the weighing room material acceptance system and the import and export linkage acceptance system at the gate in real time through the Lora device. The system automatically associates this part of the data into a complete vehicle entry, acceptance, and exit document according to the chronological order, and permanently retains the on-site photos, building an acceptance evidence chain that links business data with real image photos in the entire process of entry, acceptance, and exit.

[0072] Through built-in rules, once risks such as material vehicles not being weighed, weighing records being incomplete, and vehicles leaving the project site are identified, the warning display screen at the site entrance will immediately broadcast text and voice warnings to achieve real-time warnings for risky vehicles, and at the same time send warning message notifications to the mobile phones of relevant personnel for processing.

[0073] All vehicle entry and exit records of the project can be viewed in the cloud. The information includes entry time, exit time, entry photo, and exit photo. The system automatically associates vehicle import and export records with weighing records, and it is possible to view whether all vehicles have been weighed and the weighing rates of various materials. If a vehicle has import and export records but lacks receiving document information, it can be determined that the vehicle has not been weighed. The risk warning generated by non - weighing is permanently stored in the cloud system and can be pushed to the mobile APP of project management personnel in real - time. The warning message includes license plate number, non - weighed material category, vehicle entry and exit time, and photos, etc.

[0074] In the cloud, it can be checked whether the license plate number in the receiving document has import and export records. If there is a receiving document in the system but lacks photos of the vehicle entering and leaving the project entrance, it can be determined that there is a risk of false receiving in the receiving document. The types of receiving documents include weighing acceptance, mobile acceptance, and supplementary recording. The risk warning generated by false receiving is permanently stored in the cloud system and can be pushed to the mobile APP of project management personnel in real - time. The warning message includes license plate number, vehicle entry and exit time, photos, and receiving documents, etc.

[0075] In the cloud, it can be viewed the vehicle exit photos. The AI algorithm automatically identifies the material types carried by the vehicle in the picture and associates the import and export record with the acceptance document. If the receiving vehicle is not empty when exiting, for example, carrying materials such as steel bars and sand and gravel, it can be determined that the vehicle has a risk of stealing and pulling materials. The risk warning generated by stealing and pulling materials is permanently stored in the cloud system and can be pushed to the mobile APP of project management personnel in real - time. The warning message includes license plate number, stolen and pulled material category, vehicle entry and exit time, and photos, etc.

[0076] Based on the AI algorithm to automatically identify waste steel bars, in the cloud, it can be viewed the import and export records of vehicles carrying waste steel bars. According to whether it is associated with the material issuing document, it is judged whether the waste steel bars have been weighed when leaving the factory. The risk warning generated by non - weighing of waste steel bars is permanently stored in the cloud system and can be pushed to the mobile APP of project management personnel in real - time. The warning message includes license plate number, non - weighed material category, vehicle entry and exit time, and photos, etc.

[0077] By using the material acceptance processing method provided in this embodiment, through the acceptance document to automatically associate the vehicle's import and export records and photos, high - risk acceptance documents can be quickly located, and false receiving behaviors can be efficiently investigated. The weighing rates of materials such as concrete can be statistically calculated in real - time to ensure that the company truly grasps the concrete inspection situation and reduces cost losses. The hardware combined with the AI algorithm monitors the material leaving the factory in real - time to reduce the risk of stealing and pulling materials. The waste steel bar exit record is automatically associated with the material issuing record to ensure the standardized management of waste steel bar exits and provide evidence to follow. The risks of material entry and exit are discovered in time and warned in real - time, supporting the closed - loop of risk handling.

[0078] Embodiment 2 Corresponding to the first embodiment above, the second embodiment of the present invention provides a material acceptance processing device. For the corresponding technical feature details and corresponding technical effects, reference can be made to the first embodiment above, and they will not be elaborated herein again. Figure 3 It is a block diagram of the material acceptance processing device provided by the second embodiment of the present invention. As Figure 3 shown, the device includes: a processing module 201, an acquisition module 202, a generation module 203, a judgment module 204, and an early warning module 205.

[0079] The processing module 201 is configured to identify the identity identifier of the transport vehicle and the materials transported by the transport vehicle when the transport vehicle enters and exits the target position, and generate an inbound record and an outbound record; the acquisition module 202 is configured to acquire the acceptance record formed by the material acceptance system in the weighing house for accepting materials, where the acceptance record includes the identity information of the transport vehicle, the transport material information, and its weight information; the generation module 203 is configured to generate a material transport record according to the inbound record, the outbound record, and the acceptance record; the judgment module 204 is configured to judge whether the material transport record meets the preset material transport rules; the early warning module 205 is configured to output an alarm message when the material transport record does not meet the material transport rules.

[0080] Optionally, the processing module includes: a first extraction unit configured to extract a plurality of image frames from the real-time monitoring video stream; a first processing unit configured to input each of the image frames into a preset vehicle recognition model for vehicle target detection and output a vehicle recognition result, where the vehicle recognition result includes a first category label, a bounding box, and a first confidence level, and the first category label includes the vehicle body, the carriage, and the tank body of the tanker; a tracking unit configured to perform target tracking processing on the vehicle recognition results of each of the image frames, assign a unified tracking identifier to the same transport vehicle, and obtain a target trajectory corresponding to the tracking identifier, where the target trajectory includes the bounding box and the frame number; a screening unit configured to screen, from the vehicle recognition results of the same target trajectory, the image frames with the first confidence level greater than a preset threshold and the first category label being the carriage and the tank body as target image frames; a cropping unit configured to crop the target image frames according to the bounding boxes in the vehicle recognition results of the target image frames to obtain a target image area; a second processing unit configured to input the target image area into a preset material recognition model for material recognition and output a material recognition result, where the material recognition result includes a second category label and a second confidence level, and the second category label includes the tank body, the empty carriage, steel bars, waste steel bars, sand and gravel, formwork, wooden squares, and scaffolding; a third processing unit configured to perform aggregation processing on the material recognition results of multiple image frames with the same tracking identifier according to the second confidence level and output the recognition result of the materials transported by the transport vehicle.

[0081] The processing module includes: a determination unit configured to determine a tracking identifier of the transport vehicle according to the identity identifier; a second extraction unit configured to extract frame numbers in a target trajectory corresponding to the tracking identifier to obtain an image set; a generation unit configured to generate the inbound record according to the identity identifier of the transport vehicle, the recognition result of the material when entering the target position, the image set, and the inbound time, and generate the outbound record according to the identity identifier of the transport vehicle, the recognition result of the material when leaving the target position, the image set, and the outbound time.

[0082] The judgment module includes: a first judgment unit configured to determine that the transport vehicle has a risk of not being weighed when the inbound record and the outbound record in the material transport record match and there is no acceptance record in the material transport record.

[0083] The judgment module includes: a first judgment unit configured to determine that the transport vehicle has a risk of false material receipt when there is an acceptance record in the material transport record and there is no inbound record and / or outbound record.

[0084] The judgment module includes: a first judgment unit configured to determine that the transport vehicle has a risk of stealing materials when the inbound record and the acceptance record in the material transport record are consistent and the material recognition result in the outbound record indicates that the transport vehicle is not empty when leaving the site.

[0085] The judgment module includes: a first judgment unit configured to determine that the transport vehicle has a risk of scrap steel not being weighed when the outbound record in the material transport record is inconsistent with the acceptance record and the material recognition result in the outbound record indicates that the transport vehicle is not empty when leaving the site.

[0086] Embodiment III This embodiment further provides a computer device, such as a smart phone, a tablet computer, a notebook computer, a desktop computer, a rack server, a blade server, a tower server, or a cabinet server (including an independent server, or a server cluster composed of multiple servers) that can execute programs. As Figure 4 shown, the computer device 01 of this embodiment at least includes, but is not limited to, a memory 012 and a processor 011 that can communicate with each other through a system bus. As Figure 4 shown. It should be noted that Figure 4 only the computer device 01 having components of the memory 012 and the processor 011 is shown, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented.

[0087] In this embodiment, the memory 012 (i.e., the readable storage medium) includes flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 012 may be an internal storage unit of the computer device 01, such as the hard disk or memory of the computer device 01. In other embodiments, the memory 012 may also be an external storage device of the computer device 01, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc. equipped on the computer device 01. Of course, the memory 012 may also include both the internal storage unit and the external storage device of the computer device 01. In this embodiment, the memory 012 is generally used to store the operating system installed on the computer device 01 and various application software, such as the program code of the material acceptance processing device in Embodiment 2. In addition, the memory 012 can also be used to temporarily store various data that have been output or will be output.

[0088] In some embodiments, the processor 011 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips. The processor 011 is generally used to control the overall operation of the computer device 01. In this embodiment, the processor 011 is used to run the program code stored in the memory 012 or process data, such as the material acceptance processing method, etc.

[0089] Embodiment 4 This embodiment also provides a computer-readable storage medium, such as flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, server, App reference mall, etc., on which a computer program is stored, and when the program is executed by a processor, corresponding functions are implemented. The computer-readable storage medium of this embodiment is used to store the material acceptance processing device, and when executed by a processor, it implements the material acceptance processing method of Embodiment 1.

[0090] It should be noted that in this text, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising that element.

[0091] The serial numbers of the above embodiments of the present invention are only for description and do not represent the superiority or inferiority of the embodiments.

[0092] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-described embodiment methods can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation.

[0093] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, is equally included in the patent protection scope of the present invention.

Claims

1. A method for material acceptance and processing, characterized in that, include: When a transport vehicle enters and leaves a target location, the identity of the transport vehicle and the materials transported by the transport vehicle are respectively identified to generate an entry record and an exit record; Obtaining an acceptance record formed by the material acceptance system of the weighing room, wherein the acceptance record includes the identity information of the transport vehicle, the transport material information and its weight information; Generate a material transportation record according to the entry record, the exit record and the acceptance record; Determining whether the material transportation record meets the preset material transportation rules; When the material transportation record does not meet the material transportation rules, an alarm message is output.

2. The material acceptance and processing method according to claim 1, characterized in that, The step of identifying the material transported by the transport vehicle comprises: Extracting multiple image frames from a real-time surveillance video stream; Input each of the image frames into a preset vehicle recognition model to perform vehicle target detection, and output a vehicle recognition result, wherein the vehicle recognition result includes a first category label, a bounding box, and a first confidence level, and the first category label includes the entire vehicle, the truck bed, and the tank body of the tank truck; Performing target tracking processing on the vehicle recognition results of each of the image frames, assigning a unified tracking identifier to the same transport vehicle, and obtaining a target track corresponding to the tracking identifier, wherein the target track includes the bounding box and the frame number; In the vehicle recognition results of the same target trajectory, image frames whose first confidence is greater than a preset threshold and whose first category labels are truck bed and tank body are selected as target image frames; According to the bounding box in the vehicle recognition result of the target image frame, the target image frame is cropped to obtain a target image area; Inputting the target image area into a preset material recognition model for material recognition, and outputting a material recognition result, wherein the material recognition result includes a second category label and a second confidence level, and the second category label includes the tank body, empty truck bed, steel bar, scrap steel bar, sand and gravel, formwork, wood and scaffolding; Aggregate the material recognition results of multiple image frames of the same tracking mark according to the second confidence level, and output a recognition result of identifying the material transported by the transport vehicle.

3. The material acceptance and processing method according to claim 2, wherein, The steps to generate entry and exit records include: Determining a tracking identifier of the transport vehicle according to the identity identifier; Extracting the frame number in the target trajectory corresponding to the tracking mark to obtain an image set; Generate the entry record according to the identity of the transport vehicle, the identification result of the material when entering the target location, the image set and the entry time; and The exit record is generated according to the identity mark of the transport vehicle, the recognition result of the material when leaving the target location, the image set and the exit time.

4. The material acceptance processing method according to claim 3, characterized in that, The step of determining whether the material transportation record meets the preset material transportation rules includes: When the entry record and the exit record in the material transportation record match, and the acceptance record does not exist in the material transportation record, it is determined that the transport vehicle has a risk of not being weighed.

5. The material acceptance processing method according to claim 3, characterized in that The step of determining whether the material transportation record meets the preset material transportation rules includes: When there is the acceptance record in the material transportation record and there is no inbound record and / or outbound record, it is determined that there is a risk of false material receipt for the transport vehicle.

6. The material acceptance and processing method according to claim 3, characterized in that The steps of determining whether the material transportation record meets the preset material transportation rules include: When the inbound record and the acceptance record in the material transportation record are consistent, and the material identification result in the outbound record indicates that the transport vehicle does not leave the site with an empty load, it is determined that there is a risk of stealing materials for the transport vehicle.

7. The material acceptance and handling method according to claim 3, wherein The steps of determining whether the material transportation record meets the preset material transportation rules include: When the outbound record in the material transportation record is inconsistent with the acceptance record, and the material identification result in the outbound record indicates that the transport vehicle does not leave the site with an empty load, it is determined that there is a risk that the waste steel bars are not weighed for the transport vehicle.

8. A material acceptance and processing device, characterized in that, including: a processing module, configured to identify the identity identifier of the transport vehicle and the materials transported by the transport vehicle when the transport vehicle enters and exits the target location, and generate an inbound record and an outbound record; an acquisition module, configured to acquire an acceptance record formed by the weighbridge material acceptance system for accepting materials, where the acceptance record includes the identity information of the transport vehicle, the transported material information, and its weight information; a generation module, configured to generate a material transportation record according to the inbound record, the outbound record, and the acceptance record; a judgment module, configured to judge whether the material transportation record meets the preset material transportation rules; a warning module, configured to output an alarm message when the material transportation record does not meet the material transportation rules.

9. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images