Crowdsourcing mode-based natural resource investigation and monitoring task management method, device and equipment and medium

Abstract

The application discloses a natural resource investigation and monitoring task management method and device based on a crowdsourcing mode, equipment and medium, relates to the technical field of data management, and comprises the following steps: constructing a structured data set by standardizing and packaging a natural resource investigation polygon task; calculating a matching degree and generating an allocation scheme by using a multi-constraint algorithm in combination with the state of crowdsourcing personnel; issuing an instruction to guide a terminal to collect encrypted data packets on site; obtaining a review result through automatic preliminary screening and multi-source cross verification; and dynamically updating a personnel credit file and adjusting a matching weight to realize hierarchical self-adaptive task allocation. The method can expand investigation manpower, intelligently schedule resources, control evidence quality in the whole process, form a credit self-optimization closed loop, and improve investigation efficiency and data standardization.

Description

Technical Field

[0001] This invention relates to the field of data management technology, and in particular to a method, apparatus, equipment and medium for managing natural resource survey and monitoring tasks based on a crowdsourcing model. Background Technology

[0002] Currently, natural resource surveys and monitoring mainly adopt three operational modes: First, relying on professional surveying equipment and desktop geographic information software, dedicated professionals conduct stand-alone, closed-loop field surveys and evidence collection; second, based on mobile professional applications, fixed contract technicians complete integrated field and office surveys; and third, drawing on the crowdsourcing model of public geographic information, simple geographic information collection tasks are opened to the public. Meanwhile, the logistics industry has formed a real-time task matching and route scheduling mechanism based on location services.

[0003] The professional team operation mode is limited by staffing and scale, resulting in a significant human resource bottleneck. Tasks are mostly assigned manually and in a rough manner by area, which cannot be dynamically optimized by combining task attributes, personnel location and workload. This leads to low resource utilization and slow response. The crowdsourcing mode of public geographic information lacks unified collection standards, accuracy constraints and process guidance and verification methods. The output data is not accurate enough and the structure is not standardized, which cannot meet the professional requirements of statutory natural resource surveys and decision-making. The real-time logistics scheduling system only takes timeliness and cost as the single optimization goal. It does not take into account multiple constraints such as the complexity of natural resource task land types, confidentiality level, personnel qualification level, etc. It also lacks quality control of field collection process and post-event multi-source verification mechanism. Furthermore, it lacks personnel credit assessment and adaptive task matching system linked to data quality, making it difficult to adapt to high-precision, highly standardized and auditable natural resource survey and monitoring scenarios.

[0004] Therefore, how to absorb social crowdsourcing manpower and achieve intelligent distribution of survey tasks, standardized on-site evidence collection, automated quality review, and adaptive matching of credit closed loop has become an urgent problem to be solved. Summary of the Invention

[0005] The main purpose of this application is to provide a method, device, equipment and medium for managing natural resource survey and monitoring tasks based on a crowdsourcing model, aiming to solve the technical problem of how to absorb socialized non-professional manpower and standardize the completion of the entire process of high-precision natural resource survey and monitoring.

[0006] To achieve the above objectives, this application proposes a crowdsourcing-based method for managing natural resource survey and monitoring tasks, comprising: Receive a list of natural resource survey and monitoring tasks, and standardize and encapsulate each survey patch in the list to obtain a structured task dataset, wherein the structured task dataset includes spatial attribute information, technical attribute information and management attribute information; The comprehensive matching score is calculated based on the structured task dataset and the status information of online crowdsourcing personnel using a multi-constraint matching algorithm. A task allocation scheme is generated based on the comprehensive matching score; According to the task allocation scheme, a task assignment instruction is sent to the crowdsourcing personnel terminal so that the crowdsourcing personnel terminal can perform on-site evidence collection according to the task assignment instruction and obtain an encrypted evidence data packet. Receive the encrypted evidence data packet uploaded by the crowdsourcing personnel's terminal, perform automated initial screening and multi-source cross-validation on the encrypted evidence data packet, and obtain the quality review result; The crowdsourcing personnel credit profiles are updated based on the quality review results to obtain the updated crowdsourcing personnel credit profiles. Based on the updated crowdsourcing personnel credit profiles, the task matching weight parameters are adjusted to obtain adaptive matching configuration information, thereby achieving hierarchical adaptive matching of tasks and personnel.

[0007] In one embodiment, the step of receiving a list of natural resource survey and monitoring tasks, and standardizing and encapsulating each survey patch in the list to obtain a structured task dataset, includes: Receive the list of survey patches issued by the superior system, and extract the boundary coordinate information and center point coordinate information of each survey patch from the list of survey patches to obtain spatial attribute information; Based on the survey patch list, determine the task type information, land category code information, preset collection element list information, and preset positioning accuracy threshold information for each survey patch to obtain technical attribute information; Based on the list of survey patches, determine the task level information, promised completion time limit information, and basic reward information for each survey patch to obtain management attribute information; A structured task dataset is generated based on the spatial attribute information, the technical attribute information, and the management attribute information.

[0008] In one embodiment, the step of calculating the comprehensive matching score based on the structured task dataset and the online crowdsourcing personnel status information using a multi-constraint matching algorithm includes: Obtain real-time location information, qualification level information, historical credit score, current task load information, and land type proficiency information of the online crowdsourcing personnel to obtain the status information of the online crowdsourcing personnel; The spatiotemporal matching degree value is calculated based on the spatial attribute information of the structured task dataset and the real-time location information in the online crowdsourcing personnel status information; Based on the task level information in the management attribute information of the structured task dataset and the qualification level information, historical credit score, current task load information and land type proficiency information in the status information of the online crowdsourcing personnel, calculate the qualification compliance value, credit weighted score, load balancing factor value and skill preference matching value. The weighted summation result is obtained by performing a weighted summation operation on the spatiotemporal matching degree value, the qualification conformity degree value, the credit weighted score, the load balancing factor value, and the skill preference matching value. The weighted summation result is globally optimized using a preset optimization algorithm to obtain a comprehensive matching score.

[0009] In one embodiment, the step of generating a task allocation scheme based on the comprehensive matching score includes: Candidate crowdsourcing personnel are ranked according to the comprehensive matching score to obtain a candidate priority sequence; The task assignment method is determined based on the candidate priority sequence and the preset assignment mode rules, wherein the task assignment method includes system forced assignment, market order grabbing, or a combination of guaranteed assignment and overflow order grabbing. A task allocation scheme is generated based on the task assignment method, and the task assignment instruction corresponding to the task allocation scheme is sent to the target crowdsourcing personnel's terminal.

[0010] In one embodiment, the step of sending a task assignment instruction to the crowdsourcing personnel terminal according to the task allocation scheme, so that the crowdsourcing personnel terminal performs on-site evidence collection according to the task assignment instruction and obtains an encrypted evidence data packet, includes: A task assignment instruction is generated according to the task allocation scheme, and the task assignment instruction is sent to the crowdsourcing personnel terminal so that the crowdsourcing personnel terminal receives the task assignment instruction and displays an augmented reality guidance interface; The system receives real-time sensor data sent by the crowdsourcing personnel's terminal, and performs location compliance verification, posture compliance verification, and element compliance verification based on the real-time sensor data and the structured task dataset to obtain real-time compliance verification results. The element compliance verification is performed by identifying ground features in the viewfinder using a preset lightweight visual recognition model. Based on the real-time compliance verification result, a shooting permission instruction is sent to the crowdsourcing personnel terminal, so that the crowdsourcing personnel terminal can perform image acquisition according to the shooting permission instruction to obtain evidence images and associated metadata information; The system receives the evidence image and associated metadata information sent by the crowdsourcing personnel's terminal, and encrypts and packages the evidence image and associated metadata information to obtain an encrypted evidence data packet.

[0011] In one embodiment, the step of receiving the encrypted evidence data packet uploaded by the crowdsourcing personnel's terminal, performing automated initial screening and multi-source cross-validation on the encrypted evidence data packet, and obtaining a quality review result includes: The system receives the encrypted evidence data packet uploaded by the crowdsourcing personnel's terminal and performs integrity and logical consistency checks on the encrypted evidence data packet to obtain the initial screening results. Acquire time-series remote sensing image data, historical survey database data, and crowdsourced data of adjacent map features; Based on the time-series remote sensing image data, the historical survey database data, and the crowdsourced data of adjacent image patches, the evidence image information in the encrypted evidence data package is cross-referenced from multiple sources to obtain a cross-validation credibility score. The encrypted evidence data packet is graded and marked according to the initial screening and verification results and the cross-validation credibility score to obtain the quality review results, wherein the quality review results include automatic pass mark, doubtful review mark or invalid return mark.

[0012] In one embodiment, the step of updating the crowdsourcing personnel credit profile based on the quality review result to obtain the updated crowdsourcing personnel credit profile includes: The quality score and completion time score for this task are calculated based on the automated verification results, cross-validation results, and manual verification results in the quality review results. The overall score for this task is obtained by weighted summation of the task quality score and completion time score. The crowdsourcing personnel's long-term credit score is updated based on the overall score of this task, and their proficiency rating for specific task types is also updated based on the overall score of this task, resulting in an updated credit profile for the crowdsourcing personnel.

[0013] Furthermore, to achieve the above objectives, this application also proposes a crowdsourcing-based natural resource survey and monitoring task management device, which includes: The task encapsulation module is used to receive a list of natural resource survey and monitoring tasks, and to encapsulate each survey patch in the list of natural resource survey and monitoring tasks in a standardized manner to obtain a structured task dataset, wherein the structured task dataset includes spatial attribute information, technical attribute information and management attribute information. The calculation module is used to calculate the comprehensive matching score based on the structured task dataset and the status information of online crowdsourcing personnel using a multi-constraint matching algorithm. The scheme generation module is used to generate a task allocation scheme based on the comprehensive matching score; The task dispatch module is used to send a task dispatch instruction to the crowdsourcing personnel terminal according to the task allocation scheme, so that the crowdsourcing personnel terminal can perform on-site evidence collection according to the task dispatch instruction and obtain an encrypted evidence data packet. The quality review module is used to receive the encrypted evidence data packet uploaded by the crowdsourcing personnel's terminal, perform automated initial screening and multi-source cross-validation on the encrypted evidence data packet, and obtain the quality review result. The credit update module is used to update the credit profiles of crowdsourcing personnel based on the quality review results, and obtain the updated credit profiles of crowdsourcing personnel. The weight adjustment module is used to adjust the task matching weight parameters according to the updated crowdsourcing personnel credit profile, so as to obtain adaptive matching configuration information and realize hierarchical adaptive matching of tasks and personnel.

[0014] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the crowdsourcing-based natural resource survey and monitoring task management method described above.

[0015] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the crowdsourcing-based natural resource survey and monitoring task management method described above.

[0016] This application constructs a structured dataset by standardizing and encapsulating natural resource survey map tasks. It then calculates the matching degree using a multi-constraint algorithm, combined with the status of crowdsourced personnel, and generates an allocation scheme. Instructions are issued to guide terminals to collect encrypted data packets for on-site evidence collection. The results are verified through automated initial screening and multi-source cross-validation. Personnel credit files are dynamically updated, and matching weights are adjusted to achieve tiered adaptive task assignment. This approach expands survey manpower, intelligently schedules resources, controls the quality of evidence collection throughout the entire process, forms a self-optimizing credit loop, and improves survey efficiency and data standardization. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the first embodiment of the natural resource survey and monitoring task management method based on the crowdsourcing model of this application. Figure 2This is a flowchart illustrating the second embodiment of the natural resource survey and monitoring task management method based on the crowdsourcing model of this application. Figure 3 This is a schematic diagram of the module structure of the natural resource survey and monitoring task management device based on the crowdsourcing model of this application; Figure 4 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the crowdsourcing-based natural resource survey and monitoring task management method in this application embodiment.

[0019] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0021] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0022] Currently, natural resource surveys and monitoring mainly adopt three operational modes: First, relying on professional surveying equipment and desktop geographic information software, dedicated professionals conduct stand-alone, closed-loop field surveys and evidence collection; second, based on mobile professional applications, fixed contract technicians complete integrated field and office surveys; and third, drawing on the crowdsourcing model of public geographic information, simple geographic information collection tasks are opened to the public. Meanwhile, the logistics industry has formed a real-time task matching and route scheduling mechanism based on location services.

[0023] The professional team operation mode is limited by staffing and scale, resulting in a significant human resource bottleneck. Tasks are mostly assigned manually and in a rough manner by area, which cannot be dynamically optimized by combining task attributes, personnel location and workload. This leads to low resource utilization and slow response. The crowdsourcing mode of public geographic information lacks unified collection standards, accuracy constraints and process guidance and verification methods. The output data is not accurate enough and the structure is not standardized, which cannot meet the professional requirements of statutory natural resource surveys and decision-making. The real-time logistics scheduling system only takes timeliness and cost as the single optimization goal. It does not take into account multiple constraints such as the complexity of natural resource task land types, confidentiality level, personnel qualification level, etc. It also lacks quality control of field collection process and post-event multi-source verification mechanism. Furthermore, it lacks personnel credit assessment and adaptive task matching system linked to data quality, making it difficult to adapt to high-precision, highly standardized and auditable natural resource survey and monitoring scenarios.

[0024] Therefore, how to absorb social crowdsourcing manpower and achieve intelligent distribution of survey tasks, standardized on-site evidence collection, automated quality review, and adaptive matching of credit closed loop has become an urgent problem to be solved.

[0025] Based on the above, this application also provides a method for managing natural resource survey and monitoring tasks based on a crowdsourcing model, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the natural resource survey and monitoring task management method based on the crowdsourcing model of this application.

[0026] In this embodiment, the crowdsourcing-based natural resource survey and monitoring task management method includes steps S10 to S70: Step S10: Receive the list of natural resource survey and monitoring tasks, and standardize and encapsulate each survey patch in the list to obtain a structured task dataset.

[0027] It should be noted that the structured task dataset includes spatial attribute information, technical attribute information, and management attribute information. Step S10 includes: receiving a list of survey patches issued by the superior system, and extracting the boundary coordinates and center point coordinates of each survey patch from the list to obtain spatial attribute information; determining the task type information, land use code information, preset collection element list information, and preset positioning accuracy threshold information for each survey patch based on the list to obtain technical attribute information; determining the task level information, committed completion time limit information, and basic remuneration information for each survey patch based on the list to obtain management attribute information; and generating a structured task dataset based on the spatial attribute information, technical attribute information, and management attribute information.

[0028] Specifically, firstly, a data interface connection is established with the superior natural resources survey system, and the system periodically polls or passively receives the survey patch list issued by the superior system. The survey patch list is usually stored in tabular file or geographic information database format and contains the original records of multiple plots to be surveyed. Then, the survey patch list is parsed one by one, and the boundary coordinate information (i.e., the latitude and longitude coordinate string of all vertices that constitute the plot outline) and the center point coordinate information (i.e., the single latitude and longitude coordinate of the geometric center of the plot) are extracted from each patch record. This spatial positioning information is summarized into spatial attribute information. This is done because it is necessary to calculate the spatial distance between the crowdsourcing personnel and the task based on the precise location and range of the patch, so as to support the assignment of tasks based on proximity and route planning. Secondly, the business classification field in the survey patch list is read again. Based on the preset mapping rules, the task type information (e.g., mapping business code 01 to annual change survey, code 02 to illegal land use verification), land use code information (e.g., converting the classification codes in the database to national standard land use codes such as cultivated land 0101, forest land 0301), preset collection element list information (e.g., automatically associating the specific list of panoramic photos, close-up photos, and feature photos to be collected according to the task type), and preset positioning accuracy threshold information (e.g., setting the positioning accuracy to be less than or equal to the preset positioning accuracy threshold (e.g., 5 meters) according to the task type) are summarized into technical attribute information. This is done so that each crowdsourcing task carries clear technical execution standards, so that non-professionals can clearly know what to shoot and what accuracy to achieve after arriving at the site. Then, the management fields in the survey patch list are read, and the task level information (e.g., classifying complex patches involving basic farmland or construction land as high-level tasks and ordinary land use patches as ordinary tasks) and basic reward information (e.g., setting a completion deadline of 24 hours or 72 hours based on the urgency and size of the task) and basic reward information (e.g., calculating the basic reward amount based on the task level, distance, and difficulty coefficient) are determined according to preset hierarchical rules. These management constraints are summarized into management attribute information. This is done so that tasks of appropriate difficulty can be assigned according to the personnel's qualification level during subsequent task matching, and hard constraints on completion time and reward incentives can be set. Finally, the spatial attribute information, technical attribute information, and management attribute information are linked and integrated according to a preset unified data structure to generate a complete task record for each survey patch, containing spatial location, technical specifications, and management constraints. Multiple records are combined to generate a structured task dataset. This is done so that the subsequent intelligent scheduling engine can read all dimensions of the task at once and perform multi-dimensional comprehensive matching calculations, avoiding repeated queries of scattered data sources during the scheduling process.

[0029] Step S20: Calculate the comprehensive matching score based on the structured task dataset and the status information of online crowdsourcing personnel using a multi-constraint matching algorithm.

[0030] It should be noted that step S20 includes: obtaining the real-time location information, qualification level information, historical credit score, current task load information, and land type proficiency information of the online crowdsourcing personnel set to obtain the status information of the online crowdsourcing personnel; calculating the spatiotemporal matching degree value based on the spatial attribute information of the structured task dataset and the real-time location information in the status information of the online crowdsourcing personnel; calculating the qualification compliance value, credit weighted score, load balancing factor value, and skill preference matching value based on the task level information in the management attribute information of the structured task dataset and the qualification level information, historical credit score, current task load information, and land type proficiency information in the status information of the online crowdsourcing personnel; performing a weighted summation operation on the spatiotemporal matching degree value, qualification compliance value, credit weighted score, load balancing factor value, and skill preference matching value to obtain the weighted summation result; and performing global optimization processing on the weighted summation result through a preset optimization algorithm to obtain the comprehensive matching degree score.

[0031] It's important to understand that the online crowdsourcing pool refers to the group of all socialized workers currently online and available to accept orders. Current task load information indicates the number of survey tasks that crowdsourcing workers have accepted but not yet completed. This information is used to avoid assigning new survey tasks to workers who are already at work. Land type proficiency information records the historical performance of crowdsourcing workers in different land type survey tasks. This information is used to prioritize matching workers with land type survey tasks they are skilled in, improving the standardization of work processes.

[0032] Specifically, firstly, the positioning module built into the crowdsourcing personnel's terminal periodically reports latitude and longitude coordinates at preset time intervals (e.g., 5 minutes). The system reads the most recently reported real-time location information of each online crowdsourcing personnel from the database, and simultaneously reads their qualification level information (e.g., junior investigator, senior investigator), historical credit score (a percentage score calculated from the quality of past task completion), current task load information (the number of tasks the personnel have undertaken but not completed), and land type proficiency information (the personnel's historical pass rate in various land type tasks). This multi-dimensional information is summarized into online crowdsourcing personnel status information. This is done so that the scheduling engine can grasp the real-time status and comprehensive capability profile of each available personnel, just like food delivery needs to know where the rider is, what their credit score is, how many orders they have, and what types of meals they are good at delivering. Secondly, spatial attribute information (i.e., center point coordinates and boundary coordinates of the patch) of each task to be assigned is extracted from the structured task dataset, and real-time location information of each person is extracted from the status information of online crowdsourcing personnel. The straight-line distance between the two or the estimated travel time based on the road network is calculated to obtain the spatiotemporal matching degree value. This is done in order to prioritize the assignment of tasks to personnel who are closer and can quickly reach the site, thereby reducing travel time. Then, task level information (e.g., advanced tasks require advanced investigator qualifications) is extracted from the management attribute information of the structured task dataset and compared with the qualification level information in the status information of online crowdsourcing personnel. If the qualifications meet the requirements, the qualification compliance value is a preset full score (1 in this embodiment); otherwise, it is a preset zero value (0 in this embodiment). At the same time, the historical credit score is read as the credit weighted score, and the current task load information is taken as the reciprocal or normalized to obtain the load balancing factor value (the more tasks, the lower the factor value, to avoid personnel overload). Then, based on the land type proficiency information, it is determined whether the personnel have a high historical pass rate in the target land type to obtain the skill preference matching value. This is done to quantify the suitability between personnel and tasks from four independent dimensions: qualification threshold, credit level, workload, and professional expertise. Then, the spatiotemporal matching value, the qualification compatibility value, the credit weighted score, the load balancing factor value, and the skill preference matching value are multiplied by preset weight coefficients (e.g., spatiotemporal matching weight 0.3, credit weighted score weight 0.25, load balancing factor weight 0.2, qualification compatibility weight 0.15, and skill preference weight 0.1). The weighted values ​​of each dimension are summed, and a weighted summation is performed to obtain the weighted summation result. This is done because a single dimension cannot fully reflect the matching quality. Through weighted fusion, multiple objectives such as spatial efficiency, quality assurance, load balancing, and professional matching can be unified into a preliminary quantitative indicator.Finally, the weighted summation results corresponding to multiple candidate crowdsourcing personnel are input into a preset optimization algorithm (such as a genetic algorithm or reinforcement learning model). The preset optimization algorithm searches globally for task allocation combinations that maximize overall matching efficiency or minimize overall completion time. It performs global optimization on the locally optimal weighted summation results to obtain a comprehensive matching score. This is done because simple weighted summation can only obtain a locally optimal match between a single task and a single person. In real-world scenarios, a person can complete multiple tasks along the way, and there are spatial clustering relationships between multiple tasks. Only through global optimization can the allocation scheme with the best overall efficiency be found, avoiding situations where task A is assigned to the nearest person, Xiao Wang, and task B is assigned to the nearest person, Xiao Li, but in reality, Xiao Wang can complete both A and B simultaneously along the way, while Xiao Li needs to make a special trip.

[0033] Step S30: Generate a task allocation scheme based on the comprehensive matching score.

[0034] It should be noted that step S30 includes: sorting candidate crowdsourcing personnel according to the comprehensive matching score to obtain a candidate personnel priority sequence; determining the task assignment method according to the candidate personnel priority sequence and the preset assignment mode rules, wherein the task assignment method includes system-forced assignment, market order grabbing, or a combination of guaranteed assignment and overflow order grabbing; generating a task allocation plan according to the task assignment method, and sending the task assignment instructions corresponding to the task allocation plan to the target crowdsourcing personnel's terminal.

[0035] Specifically, first, the overall matching score between all candidate crowdsourcing personnel and the tasks to be assigned is read, and the candidates are sorted in descending order of score to generate a priority sequence. This is done to ensure that the personnel with the highest matching score and the most suitable to perform the task are placed at the front of the sequence and have priority to obtain the task assignment opportunity. Secondly, the system reads the preset task assignment rules and determines the specific task assignment method based on the current task attributes and the priority sequence of the candidates. If the current task is an urgent verification task that must be completed, the system will directly lock and assign the task to the candidate ranked first in the priority sequence using a forced assignment method. If the current task is a normal patrol task and there are enough personnel, the system will simultaneously push the task information to the top preset number (5 in this embodiment) of candidates in the priority sequence using a market order-grabbing method, and the person who confirms the order first will receive the task. If the current task is a complex change investigation task, the system will use a mixed method of guaranteed assignment and overflow order grabbing. Guaranteed assignment instructions will be sent to the top preset number (3 in this embodiment) of candidates in the priority sequence in sequence and a response will be waited for. If all requests are rejected or do not respond within a preset time window (30 minutes in this embodiment), the task will be transferred to the order-grabbing pool and opened to a wider range of personnel. This is because different scenarios have different requirements for the certainty and efficiency of task completion. For example, for food delivery, orders in bad weather will be directly assigned to the nearest rider to ensure delivery, while orders in normal weather will be opened to grabbing to improve the overall order acceptance rate. Finally, a specific task allocation scheme is generated based on the determined distribution method, clearly recording the correspondence between each task and the target crowdsourcing personnel who receive the task. The task allocation scheme is then converted into a task distribution instruction and sent to the application interface of the target crowdsourcing personnel's terminal through a message push channel, triggering a pop-up window or ringtone reminder. This is done to transform the scheduling decision result into an executable operation instruction, ensuring that the selected personnel can perceive and respond to the task in a timely manner.

[0036] Step S40: Send a task assignment instruction to the crowdsourcing personnel terminal according to the task allocation scheme, so that the crowdsourcing personnel terminal can perform on-site evidence collection according to the task assignment instruction and obtain encrypted evidence data packets.

[0037] It should be noted that step S40 includes: generating a task dispatch instruction according to the task allocation scheme and sending the task dispatch instruction to the crowdsourcing personnel terminal, so that the crowdsourcing personnel terminal receives the task dispatch instruction and displays the augmented reality guidance interface; receiving real-time sensor data collected by the crowdsourcing personnel terminal, and performing location compliance verification, posture compliance verification, and element compliance verification based on the real-time sensor data collected by the sensor and the structured task dataset to obtain real-time compliance verification results, wherein the element compliance verification is to perform real-time ground feature recognition on the viewfinder using a preset lightweight visual recognition model; sending a shooting permission instruction to the crowdsourcing personnel terminal based on the real-time compliance verification results, so that the crowdsourcing personnel terminal performs image acquisition according to the shooting permission instruction to obtain evidence images and associated metadata information; receiving the evidence images and associated metadata information sent by the crowdsourcing personnel terminal, and encrypting and packaging the evidence images and associated metadata information to obtain an encrypted evidence data package.

[0038] It's important to understand that the augmented reality guidance interface is a visual interactive interface on the terminal that overlays job prompts onto real-time video. The interface intuitively presents guidance information such as the boundaries of image patches, shooting locations, and requirements for feature collection. The preset lightweight visual recognition model is an image intelligent recognition model, such as MobileNet, pre-built into the terminal and adapted for on-device operation. The model can quickly analyze the real-time view locally without relying on remote computing power. The shooting permission command is a control instruction issued to the terminal after all verifications are completed, allowing the execution of photo capture. This command serves as a compliance access control mechanism to prevent arbitrary shooting behavior that does not comply with regulations.

[0039] Specifically, firstly, based on the task identifier, target patch coordinates, and collection requirements in the task allocation scheme, a task dispatch instruction containing task details and navigation path is generated. This instruction is then sent to the target crowdsourcing personnel's terminal via a message push channel. Upon receiving the instruction, the crowdsourcing personnel's terminal displays an augmented reality guidance interface on the application screen. This interface overlays virtual patch boundaries, preset shooting point markers, and element prompts onto the real-time camera view. This is done to allow non-professionals to intuitively see the task target range and shooting requirements upon arrival at the site, much like how mobile navigation overlays arrows onto the real-world view, significantly reducing the difficulty for them to find the target and understand the task requirements. Secondly, the system receives real-time sensor data transmitted at a preset frequency from the crowdsourcing personnel's terminals. This real-time sensor data includes Global Navigation Satellite System (GNSS) positioning data, gyroscope attitude data, and accelerometer data. The GNSS positioning data is compared with the coordinates of the center points of the patches in the structured task dataset to calculate the distance. The system then determines whether the distance difference between the current device position and the target coordinates is less than or equal to a preset positioning accuracy threshold (5 meters in this embodiment), thus obtaining a positioning compliance verification result. Finally, the system fuses and calculates the gyroscope attitude data and accelerometer data to determine whether the device's pitch and roll angles are within a preset attitude compliance range, obtaining an attitude compliance verification result. The system invokes a pre-defined lightweight visual recognition model deployed on the terminal to perform real-time ground feature recognition on the viewfinder. The recognized ground features are then compared with the pre-defined collection feature list in the structured task dataset to determine whether the required ground features are included, thus obtaining the feature compliance verification result. The location compliance verification result, posture compliance verification result, and feature compliance verification result are summarized into a real-time compliance verification result. This is done to enforce the operation behavior of crowdsourcing personnel through technical means before taking photos, ensuring that they stand in the correct position and take photos containing the required content from the correct angle, and avoiding non-professionals from collecting unqualified data due to lack of experience. Then, it is determined whether all real-time compliance verification results are in a pass state. If so, a shooting permission instruction is sent to the crowdsourcing personnel's terminal, so that the crowdsourcing personnel's terminal can unlock the shooting button in the application interface, allowing the crowdsourcing personnel to perform image acquisition operations according to the shooting permission instruction, and obtain evidence images and associated metadata information. The associated metadata information includes shooting timestamp, shooting coordinates, positioning accuracy value and device attitude angle. This is done because taking pictures is only allowed when all compliance conditions are met at the same time, which technically prevents the generation of non-compliant data, just like a camera only allows the shutter to be pressed after successfully detecting a face in focus.Finally, the system receives the evidence images and associated metadata sent by the crowdsourcing personnel's terminals, encrypts the evidence images and associated metadata using a preset encryption algorithm, and then encapsulates the encrypted evidence images and associated metadata according to a preset data packet format to obtain an encrypted evidence data packet. This is done to protect the security and integrity of the investigation data during data transmission and storage, prevent the data from being tampered with or leaked during the return transmission, and ensure the reliability of the data in subsequent review and archiving stages.

[0040] Step S50: Receive the encrypted evidence data packet uploaded by the crowdsourcing personnel's terminal, perform automated initial screening and multi-source cross-validation on the encrypted evidence data packet, and obtain the quality review result.

[0041] It should be noted that multi-source cross-validation is a process of comparing and verifying the content of the evidence data package with multiple types of external reference data. Cross-validation can determine the authenticity and accuracy of on-site evidence information from different dimensions. The quality review result is the final disposal marker formed by integrating the automated initial screening conclusions and the multi-source cross-validation score. The review result can be used to determine whether the evidence data can be directly approved or requires manual review again.

[0042] Specifically, firstly, encrypted evidence data packets uploaded by crowdsourcing personnel's terminals are received via a secure communication link. These packets are then decrypted and restored to obtain evidence images and associated metadata. Automated initial screening is then performed, including verifying the file integrity of the data packets and the logical consistency between the shooting time, shooting coordinates, and task requirements. Data with missing transmissions or obvious out-of-bounds submissions is quickly excluded, yielding the initial screening results. Secondly, the latest time-series remote sensing image data, historical survey database data, and crowdsourced data of adjacent land parcels are simultaneously acquired. The features of ground objects in the evidence images are compared with changes detected in the remote sensing images, and compared with historical data. The survey records are compared for consistency and cross-validated with spatially adjacent map data collected by different personnel, including boundary connections and land category logic. The credibility of the evidence is assessed through triangulation of multi-source data, resulting in a cross-validation credibility score. Finally, based on the initial screening results and the cross-validation credibility score, data is marked as automatically passed, questionable and requiring manual review, or invalid and returned according to preset grading thresholds, resulting in a quality review result. This is done to automatically filter out credible data from massive crowdsourced data and directly store it in the database, while diverting questionable data to professional reviewers, avoiding the efficiency bottleneck caused by manually reviewing all data one by one.

[0043] Step S60: Update the crowdsourcing personnel credit profiles based on the quality review results to obtain the updated crowdsourcing personnel credit profiles.

[0044] It should be noted that step S50 includes: calculating the task quality score and completion time score based on the automated verification results, cross-validation results, and manual verification results in the quality review results; performing a weighted summation of the task quality score and completion time score to obtain the task's overall score; updating the crowdsourcing personnel's long-term credit score based on the task's overall score, and updating the crowdsourcing personnel's proficiency rating for specific task types based on the task's overall score, to obtain the updated crowdsourcing personnel's credit profile.

[0045] Specifically, firstly, the automated verification results (including whether the initial screening passed and the number of passed items), cross-validation results (including multiple credibility scores obtained from comparing various data sources), and manual review results (including the pass or fail conclusions given by professional auditors for questionable data) are read from the quality review results. According to the preset scoring rules, the passed items in the automated verification results are assigned a basic quality score, and the credibility scores in the cross-validation results are converted into additional quality scores according to a preset ratio. Points are added or deducted based on whether the manual review results pass or fail, and the total quality score for this task is obtained. At the same time, the difference between the actual completion time and the promised completion time limit of the task is read, and the completion time score is calculated according to the preset timeliness scoring rules (e.g., bonus points for early completion, base score for on-time completion, and deduction points for late completion). This is done because data quality and completion timeliness are two core dimensions for measuring the work performance of crowdsourcing personnel, and they need to be quantified separately to fully reflect their actual performance in this task. Secondly, the task quality score is multiplied by a preset quality weighting coefficient (0.7 in this embodiment), and the completion timeliness score is multiplied by a preset timeliness weighting coefficient (0.3 in this embodiment). The two weighted values ​​are then summed to obtain the overall task score. This is done because, compared to simply pursuing speed while ignoring data accuracy, or excessively pursuing perfection and resulting in serious timeouts, it is more important to strike a balance between quality and timeliness. Weighting can guide crowdsourcing personnel to consider both aspects, just as a student's final grade is determined by a certain proportion of their homework and final exam scores. Finally, the overall score of this task is added to the crowdsourcing worker's historical long-term credit score with a preset decay coefficient (e.g., 0.1) to obtain the updated long-term credit score. At the same time, the land type corresponding to this task is identified, and the historical proficiency rating under this land type is adjusted according to the overall score of this task (e.g., if the score is higher than the preset upgrade threshold, it is upgraded by one level; if it is lower than the preset downgrade threshold, it is downgraded by one level) to obtain the updated proficiency rating for the specific task type. The updated long-term credit score and the updated proficiency rating for the specific task type are written into the corresponding record in the database to obtain the updated credit profile of the crowdsourcing worker. This is done so that the credit profile can retain the long-term accumulation of the worker's historical performance (avoiding complete negation due to a single mistake) and dynamically reflect the worker's true area of ​​expertise in the specific land type dimension, thereby achieving more accurate person-job matching in the next round of task matching.

[0046] Step S70: Adjust the task matching weight parameters according to the updated crowdsourcing personnel credit profile to obtain adaptive matching configuration information, thereby achieving hierarchical adaptive matching of tasks and personnel.

[0047] It's important to note that, firstly, the updated credit profiles of crowdsourcing personnel are read, extracting updated long-term credit scores and updated proficiency ratings for specific task types. Based on the changes in long-term credit scores relative to historical baselines, the weighting coefficients of the credit weighted score are dynamically adjusted (e.g., when a person's credit score consistently exceeds a preset excellent threshold, their weighting coefficient is increased, giving them a greater competitive advantage in subsequent matching; when their credit score falls below a preset warning threshold, their weighting coefficient is decreased, and the level of tasks they can accept is limited). Simultaneously, based on the rise and fall of the proficiency rating for specific task types, the weighting coefficients of the skill preference matching value in the corresponding geographical dimension are adjusted. These adjusted weighting coefficients are then aggregated into adaptive matching configuration information. This is done because changes in the credit profile directly reflect the actual fluctuations in a person's abilities. Dynamically adjusting the weighting parameters allows high-performing personnel to continuously obtain more high-quality task opportunities, while naturally reducing the allocation of high-value tasks to those with declining performance, forming a two-way adjustment mechanism of positive incentives and negative elimination, similar to how ride-hailing services dynamically adjust drivers' order acceptance priority and dispatch weight based on their service score and completion rate. Secondly, the adaptive matching configuration information is written into the task matching weight parameter library as the basis for real-time invocation during the next round of task matching. During the next round of multi-constraint matching algorithm execution, the various weight coefficients in the adaptive matching configuration information are directly read to participate in the calculation of the comprehensive matching score, instead of using fixed default weights. This ensures that the credit assessment results are truly fed back into the scheduling decision, preventing the credit file from becoming a static file and losing its actual constraint effect, thus forming a complete closed loop from task execution to quality assessment to scheduling optimization. Finally, based on the changes in weight coefficients and updates to personnel capability profiles in the adaptive matching configuration information, the next round of task allocation automatically implements hierarchical adaptive matching between tasks and personnel. For example, personnel with high credit and high proficiency are given priority for high-level, high-paying tasks, while personnel with low credit or new hires can only receive ordinary tasks or need to undergo more rounds of review. This allows the entire system to continuously self-optimize over time, identifying and strengthening high-quality data sources, filtering and eliminating low-quality data sources, and continuously improving the overall average data quality and execution efficiency of the system.

[0048] This embodiment constructs a structured dataset by standardizing and encapsulating natural resource survey map patch tasks. It then calculates the matching degree using a multi-constraint algorithm, combined with the status of crowdsourced personnel, and generates an allocation scheme. Instructions are issued to guide terminals to collect encrypted data packets for on-site evidence collection. Automated initial screening and multi-source cross-validation yield verification results. Personnel credit files are dynamically updated, and matching weights are adjusted to achieve tiered adaptive task assignment. This approach expands survey manpower, intelligently schedules resources, controls the quality of evidence collection throughout the entire process, forms a self-optimizing credit loop, and improves survey efficiency and data standardization.

[0049] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in Embodiment 1 above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 2 The crowdsourcing-based natural resource survey and monitoring task management method step S50 further includes steps S201~S204: Step S201: Receive the encrypted evidence data packet uploaded by the crowdsourcing personnel's terminal, and perform integrity verification and logical consistency verification on the encrypted evidence data packet to obtain the initial screening verification result.

[0050] It should be noted that, firstly, the encrypted evidence data packet uploaded by the crowdsourcing personnel's terminal is received through a preset secure communication link. The encrypted evidence data packet is then decrypted and restored to obtain the original data packet containing the evidence image file and associated metadata file. Next, an integrity check is performed on the original data packet, checking item by item whether the data packet simultaneously contains the evidence image file and associated metadata file, whether the image file can be opened normally and its size is not zero bytes, and whether all required fields in the metadata file are filled. If any item is missing or abnormal, the integrity check is deemed to have failed, and the integrity check result is obtained. Secondly, the shooting timestamp and shooting coordinates in the associated metadata information are read. The shooting timestamp is compared with the promised completion deadline corresponding to the task in the structured task dataset to determine whether the shooting time falls within the task's effective period. Simultaneously, the shooting coordinates are compared with the edges of the task patch. The spatial inclusion relationship is calculated using boundary coordinates to determine whether the shooting location is within the boundary range of the task patch. If either the time or space exceeds the task requirements, the logical consistency check is deemed to have failed, and the logical consistency check result is obtained. Finally, the integrity check result and the logical consistency check result are merged into the initial screening check result. If both pass, the initial screening check result is in a passed state; if either fails, the initial screening check result is in a failed state, and the specific anomaly is recorded. This is done so that before investing a lot of computing resources in deep cross-validation, lightweight rule-based validation can quickly filter out obviously unqualified data caused by network transmission interruption, terminal misoperation, or malicious submission. Just like a courier sorting center first uses a scanner to check whether the package label is complete and whether the address is within the delivery range. Only packages with qualified appearance and basic information will enter the subsequent weighing and security inspection stages.

[0051] Step S202: Obtain time-series remote sensing image data, historical survey database data, and crowdsourced data of adjacent map patches.

[0052] Specifically, firstly, a data request is sent to the satellite image data center through a preset remote sensing image service interface. This request carries the boundary coordinates of the task patch and the image resolution requirements. The latest time-series remote sensing image data is then received, including multiple satellite images covering the task patch area. This is done to obtain objective surface information of the same land parcel observed from an aerial perspective, for cross-comparison with ground-based evidence images from different viewpoints. Secondly, the historical survey database is accessed through a database query interface. Based on the unique identifier of the task patch, the survey records for that patch in the previous year or at the previous change point are retrieved. Corresponding historical evidence images, historical land use determination results, and historical boundary coordinate data are obtained, thus acquiring the historical survey database data. This is done to understand the past state of the land parcel and determine whether the current evidence data shows changes that are inconsistent with natural laws or management logic, much like a doctor needs to review a patient's medical records to determine if current symptoms indicate an abnormal condition. Finally, the spatial buffer range is calculated based on the boundary coordinates of the task patch. Encrypted evidence data packets of adjacent patches that have been evidenced and uploaded by other crowdsourcing personnel within the spatial buffer range are retrieved. Boundary coordinate information, land use determination results, and evidence image information of adjacent patches are extracted from them to obtain crowdsourcing data of adjacent patches. This is done to use adjacent plot data collected independently by different personnel to perform logical verification of spatial connection and land use transition, and to avoid data distortion caused by subjective misjudgment of a single person or systemic bias.

[0053] Step S203: Based on time-series remote sensing image data, historical survey database data, and crowdsourced data of adjacent map patches, perform multi-source cross-comparison of the evidence image information in the encrypted evidence data package to obtain the cross-validation credibility score.

[0054] Specifically, firstly, multiple satellite images corresponding to the task patch area are extracted from the time-series remote sensing image data. Then, the land cover features and boundary contour information of the plot in the multiple satellite images are extracted using a preset change detection algorithm. The land cover features and boundary contour information are then compared with the ground real-world feature types and spatial ranges reflected in the evidence images in the encrypted evidence data package. The first matching degree of the two in terms of land type consistency and change location coincidence is calculated to obtain the first credibility score. This is done because satellite images can objectively record the true state of the ground surface from a high altitude, just like looking down from an upper floor can clearly see the actual layout of the garden below. Ground shooting may cause misjudgment due to angle obstruction. Cross-verification of the sky and earth perspectives can eliminate misjudgment caused by shooting angle or partial obstruction. Secondly, the historical land use classification results and historical boundary coordinates of the map patch in the previous year or at the previous time are extracted from the historical survey database. The land use classification results in the current evidence image are compared with the historical land use classification results to determine whether the land use change conforms to a reasonable evolution pattern (for example, the conversion of cultivated land to construction land usually requires legal approval procedures. If the current evidence shows that it is construction land but there is no relevant approval record, there is an anomaly). At the same time, the current boundary coordinates are spatially superimposed and compared with the historical boundary coordinates to determine whether the boundary offset is within the preset reasonable threshold range, and a second credibility score is obtained. This is done because the state of natural resource plots usually has temporal continuity. Just as a person's physical examination report needs to be compared with data from previous years to determine whether a certain indicator has changed abnormally, abrupt changes that do not conform to logical patterns can be identified through historical longitudinal comparison. Then, the boundary coordinates and land use determination results of spatially adjacent land parcels collected by different crowdsourcers are extracted from the crowdsourced data of adjacent land parcels. The boundary coordinates of the current land parcel are spatially compared with the boundary coordinates of adjacent land parcels to check for gaps, overlaps, or misalignments. At the same time, the land use determination results of the current land parcel are compared with the land use determination results of adjacent land parcels to determine whether the land use transition between adjacent land parcels conforms to natural geographical laws (for example, the adjacent paddy fields and dry land are a reasonable transition, but there should be no abrupt change without buffer between paddy fields and construction land). A third credibility score is obtained. This is done because adjacent land parcels are collected independently by different people. If the land use and boundary determinations of the same boundary area by two people can corroborate each other, the data credibility is greatly improved. It is like two independent eyewitnesses describing the same car accident. If the details match, the testimony is highly credible. If they contradict each other, at least one of them has a problem.Finally, the first credibility score, the second credibility score, and the third credibility score are weighted and fused according to preset fusion weights (e.g., remote sensing image weight 0.4, historical data weight 0.3, and adjacent data weight 0.3) to obtain the cross-validation credibility score. This is done because the validation of a single data source has its own limitations. Satellite image resolution is limited and details may not be clear. Historical data may be too outdated. Adjacent data may be affected by the joint misjudgment of adjacent personnel. Only by combining the validation results of three independent sources can the most robust credibility assessment be formed.

[0055] Step S204: Based on the initial screening and verification results and the cross-validation credibility score, the encrypted evidence data packet is graded and marked to obtain the quality review results.

[0056] It should be noted that the quality review results include an automatic pass mark, a questionable review mark, or an invalid return mark. The automatic pass mark is a special identifier assigned to evidence data that meets compliance and credibility standards. Evidence data with this mark can be directly included in the investigation results database without manual intervention. The questionable review mark is a special identifier assigned to evidence data with contradictory information or whose credibility is in a critical range. Evidence data with this mark needs to be sent to professionals for secondary manual verification. The invalid return mark is a special identifier for evidence data that fails verification and whose credibility is significantly lower than the standard. Evidence data with this mark is directly deemed invalid and will not be included in subsequent investigation results.

[0057] Specifically, firstly, the initial screening verification result and cross-validation credibility score are read to determine whether the initial screening verification result is in a passed state and whether the cross-validation credibility score is higher than the preset automatic pass threshold (95 in this embodiment). If both conditions are met, the encrypted evidence data packet is marked as automatically passed and directly written into the results database for archiving. Secondly, if the initial screening verification result is in a passed state but the cross-validation credibility score is between the lower limit of the preset doubtful interval (70 in this embodiment) and the preset automatic pass threshold, the encrypted evidence data packet is marked as doubtful for review, and the doubtful data packet is sent to the message push channel. The verification mark and corresponding evidence data are pushed to the professional reviewer's terminal, triggering the manual verification process. Finally, if the initial screening result is unsuccessful, or the cross-validation credibility score is lower than the preset invalid return threshold, the encrypted evidence data packet is marked as invalid and returned. The task is automatically unbound from the crowdsourcing personnel and the task is re-entered into the task pool to trigger the re-verification process. This is done to automatically distribute massive crowdsourcing data according to its credibility level. Credible data is directly stored in the database to reduce the pressure of manual review, suspicious data enters the professional verification channel to ensure seriousness, and invalid data is returned in time to avoid contaminating the database.

[0058] This embodiment receives encrypted evidence data packets uploaded by crowdsourced personnel's terminals and first performs integrity and logical consistency checks to obtain initial screening results. Then, it acquires time-series remote sensing image data, historical survey database data, and crowdsourced data from adjacent image patches, performing multi-source cross-comparison of the evidence image information to obtain a credibility score. Combining these two types of results, the evidence data packets are categorized and labeled, resulting in three types of quality review results: automatically approved, questionable for review, and invalid and returned. This enables automated batch initial review and multi-source verification of evidence data, reducing the burden of manual review and ensuring the standardization and reliability of evidence data in natural resource surveys.

[0059] Based on the first embodiment of this application, this application also provides a natural resource survey and monitoring task management device based on a crowdsourcing model. Please refer to... Figure 3 The device includes: The task encapsulation module 10 is used to receive the list of natural resource survey and monitoring tasks, and to encapsulate each survey patch in the list of natural resource survey and monitoring tasks in a standardized manner to obtain a structured task dataset, wherein the structured task dataset includes spatial attribute information, technical attribute information and management attribute information.

[0060] The calculation module 20 is used to calculate the comprehensive matching score based on the structured task dataset and the status information of online crowdsourcing personnel through a multi-constraint matching algorithm.

[0061] The scheme generation module 30 is used to generate a task allocation scheme based on the comprehensive matching score.

[0062] The task dispatch module 40 is used to send task dispatch instructions to the crowdsourcing personnel terminal according to the task allocation scheme, so that the crowdsourcing personnel terminal can perform on-site evidence collection according to the task dispatch instructions and obtain encrypted evidence data packets.

[0063] The quality review module 50 is used to receive encrypted evidence data packets uploaded by crowdsourcing personnel's terminals, perform automated initial screening and multi-source cross-validation on the encrypted evidence data packets, and obtain quality review results.

[0064] The credit update module 60 is used to update the credit profiles of crowdsourcing personnel based on the quality review results, and obtain the updated credit profiles of crowdsourcing personnel.

[0065] The weight adjustment module 70 is used to adjust the task matching weight parameters according to the updated crowdsourcing personnel credit profile, obtain adaptive matching configuration information, and realize hierarchical adaptive matching of tasks and personnel.

[0066] The crowdsourcing-based natural resource survey and monitoring task management device provided in this application, employing the crowdsourcing-based natural resource survey and monitoring task management method described in the above embodiments, can solve the technical problem of how to absorb non-professional social manpower and standardize the entire process of high-precision natural resource survey and monitoring. Compared with the prior art, the beneficial effects of the crowdsourcing-based natural resource survey and monitoring task management device provided in this application are the same as those of the crowdsourcing-based natural resource survey and monitoring task management method provided in the above embodiments, and other technical features in the crowdsourcing-based natural resource survey and monitoring task management device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0067] This application provides a crowdsourcing-based natural resource survey and monitoring task management device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the crowdsourcing-based natural resource survey and monitoring task management method in Embodiment 1 above.

[0068] The following is for reference. Figure 4 The diagram illustrates a structural schematic of a crowdsourcing-based natural resource survey and monitoring task management device suitable for implementing embodiments of this application. The crowdsourcing-based natural resource survey and monitoring task management device in this application embodiment may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), and vehicle-mounted terminals (e.g., vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 4 The crowdsourcing-based natural resource survey and monitoring task management device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0069] like Figure 4As shown, the crowdsourcing-based natural resource survey and monitoring task management device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the crowdsourcing-based natural resource survey and monitoring task management device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the crowdsourcing-based natural resource survey and monitoring task management equipment to communicate wirelessly or wiredly with other devices to exchange data. Although various crowdsourcing-based natural resource survey and monitoring task management devices are shown in the figures, it should be understood that implementation or possession of all of them is not required. More or fewer devices may be implemented alternatively.

[0070] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0071] The crowdsourcing-based natural resource survey and monitoring task management equipment provided in this application, employing the crowdsourcing-based natural resource survey and monitoring task management method described in the above embodiments, can solve the technical problem of how to absorb non-professional social manpower and standardize the completion of the entire process of high-precision natural resource survey and monitoring. Compared with the prior art, the beneficial effects of the crowdsourcing-based natural resource survey and monitoring task management equipment provided in this application are the same as those of the crowdsourcing-based natural resource survey and monitoring task management method provided in the above embodiments, and other technical features of this crowdsourcing-based natural resource survey and monitoring task management equipment are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0072] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0073] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0074] This application provides a computer-readable medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the crowdsourcing-based natural resource survey and monitoring task management method in the above embodiments.

[0075] The computer-readable medium provided in this application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor devices, or any combination thereof. More specific examples of computer-readable media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable medium may be any tangible medium containing or storing a program that can be executed by instructions, used by a device, or used in conjunction with it. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0076] The aforementioned computer-readable medium may be included in a crowdsourcing-based natural resource survey and monitoring task management device; or it may exist independently and not be assembled into a crowdsourcing-based natural resource survey and monitoring task management device.

[0077] The aforementioned computer-readable medium carries one or more programs that, when executed by a crowdsourcing-based natural resource survey and monitoring task management device, enable the crowdsourcing-based natural resource survey and monitoring task management device to write computer program code for performing the operations of this application in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0078] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of this application. In this regard, all blocks in the flowcharts or block diagrams may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that all blocks in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using dedicated hardware-based implementations that perform the specified functions or operations, or using a combination of dedicated hardware and computer instructions.

[0079] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0080] The readable medium provided in this application is a computer-readable medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-described crowdsourcing-based natural resource survey and monitoring task management method. This solves the technical problem of how to absorb non-professional human resources from the community and standardize the entire process of high-precision natural resource survey and monitoring. Compared with the prior art, the beneficial effects of the computer-readable medium provided in this application are the same as those of the crowdsourcing-based natural resource survey and monitoring task management method provided in the above embodiments, and will not be repeated here.

[0081] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the crowdsourcing-based natural resource survey and monitoring task management method described above.

[0082] The computer program product provided in this application can solve the technical problem of how to absorb non-professional social manpower and standardize the entire process of high-precision natural resource survey and monitoring. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the crowdsourcing-based natural resource survey and monitoring task management method provided in the above embodiments, and will not be repeated here.

[0083] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for managing natural resource survey and monitoring tasks based on a crowdsourcing model, characterized in that, The method includes: Receive a list of natural resource survey and monitoring tasks, and standardize and encapsulate each survey patch in the list to obtain a structured task dataset, wherein the structured task dataset includes spatial attribute information, technical attribute information and management attribute information; The comprehensive matching score is calculated based on the structured task dataset and the status information of online crowdsourcing personnel using a multi-constraint matching algorithm. A task allocation scheme is generated based on the comprehensive matching score; According to the task allocation scheme, a task assignment instruction is sent to the crowdsourcing personnel terminal so that the crowdsourcing personnel terminal can perform on-site evidence collection according to the task assignment instruction and obtain an encrypted evidence data packet. Receive the encrypted evidence data packet uploaded by the crowdsourcing personnel's terminal, perform automated initial screening and multi-source cross-validation on the encrypted evidence data packet, and obtain the quality review result; The crowdsourcing personnel credit profiles are updated based on the quality review results to obtain the updated crowdsourcing personnel credit profiles. Based on the updated crowdsourcing personnel credit profiles, the task matching weight parameters are adjusted to obtain adaptive matching configuration information, thereby achieving hierarchical adaptive matching of tasks and personnel.

2. The method as described in claim 1, characterized in that, The steps of receiving the natural resource survey and monitoring task list and standardizing and encapsulating each survey patch in the natural resource survey and monitoring task list to obtain a structured task dataset include: Receive the list of survey patches issued by the superior system, and extract the boundary coordinate information and center point coordinate information of each survey patch from the list of survey patches to obtain spatial attribute information; Based on the survey patch list, determine the task type information, land category code information, preset collection element list information, and preset positioning accuracy threshold information for each survey patch to obtain technical attribute information; Based on the list of survey patches, determine the task level information, promised completion time limit information, and basic reward information for each survey patch to obtain management attribute information; A structured task dataset is generated based on the spatial attribute information, the technical attribute information, and the management attribute information.

3. The method as described in claim 1, characterized in that, The step of calculating the comprehensive matching score based on the structured task dataset and the online crowdsourcing personnel status information using a multi-constraint matching algorithm includes: Obtain real-time location information, qualification level information, historical credit score, current task load information, and land type proficiency information of the online crowdsourcing personnel to obtain the status information of the online crowdsourcing personnel; The spatiotemporal matching degree value is calculated based on the spatial attribute information of the structured task dataset and the real-time location information in the online crowdsourcing personnel status information; Based on the task level information in the management attribute information of the structured task dataset and the qualification level information, historical credit score, current task load information and land type proficiency information in the status information of the online crowdsourcing personnel, calculate the qualification compliance value, credit weighted score, load balancing factor value and skill preference matching value. The weighted summation result is obtained by performing a weighted summation operation on the spatiotemporal matching degree value, the qualification conformity degree value, the credit weighted score, the load balancing factor value, and the skill preference matching value. The weighted summation result is globally optimized using a preset optimization algorithm to obtain a comprehensive matching score.

4. The method as described in claim 1, characterized in that, The step of generating a task allocation scheme based on the comprehensive matching score includes: Candidate crowdsourcing personnel are ranked according to the comprehensive matching score to obtain a candidate priority sequence; The task assignment method is determined based on the candidate priority sequence and the preset assignment mode rules, wherein the task assignment method includes system forced assignment, market order grabbing, or a combination of guaranteed assignment and overflow order grabbing. A task allocation scheme is generated based on the task assignment method, and the task assignment instruction corresponding to the task allocation scheme is sent to the target crowdsourcing personnel's terminal.

5. The method as described in claim 1, characterized in that, The step of sending a task assignment instruction to the crowdsourcing personnel terminal according to the task allocation scheme, so that the crowdsourcing personnel terminal can perform on-site evidence collection according to the task assignment instruction and obtain an encrypted evidence data packet, includes: A task assignment instruction is generated according to the task allocation scheme, and the task assignment instruction is sent to the crowdsourcing personnel terminal so that the crowdsourcing personnel terminal receives the task assignment instruction and displays an augmented reality guidance interface; The system receives real-time sensor data sent by the crowdsourcing personnel's terminal, and performs location compliance verification, posture compliance verification, and element compliance verification based on the real-time sensor data and the structured task dataset to obtain real-time compliance verification results. The element compliance verification is performed by identifying ground features in the viewfinder using a preset lightweight visual recognition model. Based on the real-time compliance verification result, a shooting permission instruction is sent to the crowdsourcing personnel terminal, so that the crowdsourcing personnel terminal can perform image acquisition according to the shooting permission instruction to obtain evidence images and associated metadata information; The system receives the evidence image and associated metadata information sent by the crowdsourcing personnel's terminal, and encrypts and packages the evidence image and associated metadata information to obtain an encrypted evidence data packet.

6. The method as described in claim 1, characterized in that, The steps of receiving the encrypted evidence data packet uploaded by the crowdsourcing personnel's terminal, performing automated initial screening and multi-source cross-validation on the encrypted evidence data packet, and obtaining quality review results include: The system receives the encrypted evidence data packet uploaded by the crowdsourcing personnel's terminal and performs integrity and logical consistency checks on the encrypted evidence data packet to obtain the initial screening results. Acquire time-series remote sensing image data, historical survey database data, and crowdsourced data of adjacent map features; Based on the time-series remote sensing image data, the historical survey database data, and the crowdsourced data of adjacent image patches, the evidence image information in the encrypted evidence data package is cross-referenced from multiple sources to obtain a cross-validation credibility score. The encrypted evidence data packet is graded and marked according to the initial screening and verification results and the cross-validation credibility score to obtain the quality review results, wherein the quality review results include automatic pass mark, doubtful review mark or invalid return mark.

7. The method as described in claim 1, characterized in that, The step of updating the crowdsourcing personnel credit profile based on the quality review results to obtain the updated crowdsourcing personnel credit profile includes: The quality score and completion time score for this task are calculated based on the automated verification results, cross-validation results, and manual verification results in the quality review results. The overall score for this task is obtained by weighted summation of the task quality score and completion time score. The crowdsourcing personnel's long-term credit score is updated based on the overall score of this task, and their proficiency rating for specific task types is also updated based on the overall score of this task, resulting in an updated credit profile for the crowdsourcing personnel.

8. A natural resource survey and monitoring task management device based on a crowdsourcing model, characterized in that, The device includes: The task encapsulation module is used to receive a list of natural resource survey and monitoring tasks, and to encapsulate each survey patch in the list of natural resource survey and monitoring tasks in a standardized manner to obtain a structured task dataset, wherein the structured task dataset includes spatial attribute information, technical attribute information and management attribute information. The calculation module is used to calculate the comprehensive matching score based on the structured task dataset and the status information of online crowdsourcing personnel using a multi-constraint matching algorithm. The scheme generation module is used to generate a task allocation scheme based on the comprehensive matching score; The task dispatch module is used to send a task dispatch instruction to the crowdsourcing personnel terminal according to the task allocation scheme, so that the crowdsourcing personnel terminal can perform on-site evidence collection according to the task dispatch instruction and obtain an encrypted evidence data packet. The quality review module is used to receive the encrypted evidence data packet uploaded by the crowdsourcing personnel's terminal, perform automated initial screening and multi-source cross-validation on the encrypted evidence data packet, and obtain the quality review result. The credit update module is used to update the credit profiles of crowdsourcing personnel based on the quality review results, and obtain the updated credit profiles of crowdsourcing personnel. The weight adjustment module is used to adjust the task matching weight parameters according to the updated crowdsourcing personnel credit profile, so as to obtain adaptive matching configuration information and realize hierarchical adaptive matching of tasks and personnel.

9. A crowdsourcing-based natural resource survey and monitoring task management device, characterized in that, The device includes: a memory, a processor, and a crowdsourcing-based natural resource survey and monitoring task management program stored on the memory and running on the processor, the crowdsourcing-based natural resource survey and monitoring task management program being configured to implement the steps of the crowdsourcing-based natural resource survey and monitoring task management method as described in any one of claims 1-7.

10. A storage medium, characterized in that, The storage medium stores a crowdsourcing-based natural resource survey and monitoring task management program. When the crowdsourcing-based natural resource survey and monitoring task management program is executed by the processor, it implements the steps of the crowdsourcing-based natural resource survey and monitoring task management method as described in any one of claims 1-7.

Images