Sighting telescope calibration method based on environmental perception

Abstract

The invention discloses a sighting telescope calibration method based on environmental perception, and relates to the technical field of data processing. The method comprises the steps that an environment perception integrator is interacted based on a risk mitigation time window, and an environment perception parameter sequence set is obtained; environment abnormity recognition is executed, and a construction environment abnormity coefficient is obtained; when a predefined environment anomaly threshold value is met, an aiming object recognition unit is activated, aiming object analysis and recognition are carried out, and a target aiming object is obtained; determining a target position feature of the target aiming object, and performing local amplification to obtain a target amplified view field; performing abnormal tracking on the target magnified view field to obtain a real-time tracking image; construction risk relief verification is executed based on the real-time tracking image, and calibration reset of the adjustable sighting telescope is controlled based on the verification result. The technical problem that in the prior art, it is difficult to accurately capture construction environment abnormity in building construction monitoring is solved, and the technical effect of improving the building construction monitoring level is achieved by integrating the sighting telescope in monitoring.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically to a method for calibrating a sight based on environmental perception. Background Technology

[0002] With the rapid development of the modern construction industry, the requirements for monitoring and management of construction sites are becoming increasingly stringent. Traditional construction monitoring methods, such as manual inspections and fixed surveillance cameras, while meeting basic safety monitoring needs to a certain extent, often fall short in terms of response speed, monitoring accuracy, and risk mitigation efficiency when faced with complex and ever-changing construction environments and rapidly evolving construction risks. This is especially true in large and complex construction projects, where the uncertainty of the construction environment and the diversity of construction risks make traditional monitoring methods inadequate. To address these technical challenges, an environmentally perceptive-based aiming scope calibration method is proposed. By integrating an adjustable aiming scope into the construction monitoring camera, it enables rapid location and magnification of the target object, providing high-definition real-time images. This helps construction managers more accurately assess construction risks and take appropriate mitigation measures. Summary of the Invention

[0003] This application provides a targeting scope calibration method based on environmental perception, which solves the technical problem in the prior art that it is difficult to accurately capture abnormalities in the construction environment during construction monitoring.

[0004] In view of the above problems, embodiments of this application provide a sight calibration method based on environmental awareness.

[0005] This application provides an environmental awareness-based method for calibrating a sight, the method comprising: A risk mitigation time window is pre-constructed. Data is retrieved from the interactive environment perception integrator based on this time window to obtain an environmental perception parameter sequence set. The environment perception integrator is pre-deployed in the construction monitoring space. Environmental anomaly identification is performed based on the environmental perception parameter sequence set to obtain a construction environment anomaly coefficient. A target identification unit is pre-constructed, communicatively connected to a construction monitoring camera deployed in the construction monitoring space. The construction monitoring camera integrates an adjustable aiming scope. When the construction environment anomaly coefficient meets a predefined environmental anomaly threshold, the target identification unit is activated. The target identification unit is based on… The risk mitigation time window interacts with the construction monitoring camera to retrieve historical images, and analyzes and identifies the target object based on the retrieval results to obtain the target target object; the target position characteristics of the target target object are determined based on the construction monitoring space, and the adjustable aiming scope is controlled to locally magnify the target target object according to the target position characteristics to obtain a magnified target field of view; the monitoring and tracking function of the construction monitoring camera is activated with the risk mitigation time window as a function start constraint to perform abnormal tracking on the magnified target field of view to obtain real-time tracking images; construction risk mitigation verification is performed based on the real-time tracking images, and the calibration reset of the adjustable aiming scope is controlled based on the verification results.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: First, a risk mitigation time window is pre-constructed. Based on this window, data is retrieved from an interactive environmental perception integrator to obtain a set of environmental perception parameter sequences. The environmental perception integrator is pre-deployed within the construction monitoring space. Next, environmental anomaly identification is performed based on the environmental perception parameter sequence set to obtain the construction environment anomaly coefficient. Then, a target identification unit is pre-constructed, which is communicatively connected to a construction monitoring camera deployed within the construction monitoring space. Each construction monitoring camera integrates an adjustable aiming scope. When the construction environment anomaly coefficient meets a predefined environmental anomaly threshold, the target identification unit is activated. This unit retrieves historical images from the construction monitoring camera based on the risk mitigation time window and analyzes and identifies the target based on the retrieved results, obtaining the target target. Further, the target position characteristics of the target target are determined based on the construction monitoring space. The adjustable aiming scope is then controlled to locally magnify the target target based on these characteristics, obtaining a magnified field of view. Finally, using the risk mitigation time window as a functional activation constraint, the monitoring and tracking function of the construction monitoring camera is activated to perform anomaly tracking on the magnified field of view, obtaining real-time tracking images. Finally, a construction risk mitigation verification is performed based on real-time tracking images, and the calibration reset of the adjustable sight is controlled based on the verification results. This solves the technical problem in existing technologies that make it difficult to accurately capture anomalies in the construction environment during construction monitoring, and achieves the technical effect of improving the level of construction monitoring by integrating a sight into the monitoring system. Attached Figure Description

[0007] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0008] Figure 1 A schematic diagram of the environmental awareness-based aiming scope calibration method provided in this application embodiment; Figure 2 This is a schematic diagram illustrating the process of environmental anomaly identification in the environmental perception-based aiming scope calibration method provided in the embodiments of this application. Detailed Implementation

[0009] This application provides an environmentally aware-based aiming scope calibration method, which solves the technical problem in the prior art of accurately capturing abnormal construction environment in building construction monitoring.

[0010] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0011] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0012] Example 1 like Figure 1 As shown in the embodiments of this application, a method for calibrating a sight based on environmental awareness is provided, wherein the method includes: A risk mitigation time window is pre-constructed, and data is retrieved based on the risk mitigation time window through the interactive environment perception integrator to obtain a set of environmental perception parameter sequences. The environment perception integrator is pre-deployed in the construction monitoring space.

[0013] Based on the characteristics, risk types, and emergency response speed of the construction project, a risk mitigation time window is predefined. This window represents the maximum response time required from system identification of an environmental anomaly to the implementation of corresponding risk mitigation measures. Environmental sensing integrators are pre-deployed at key locations within the construction monitoring space to comprehensively and accurately perceive the state of the construction environment. These integrators include various sensors, such as temperature sensors, humidity sensors, and gas detectors. They continuously collect data from the construction environment and organize it into a set of environmental sensing parameter sequences, including but not limited to temperature, humidity, gas concentration, and light intensity. Based on the pre-built risk mitigation time window, the system interacts with the environmental sensing integrator to access the environmental sensing parameter sequence set. Pre-building the risk mitigation time window and interacting with the environmental sensing integrator to retrieve data based on it is an effective means of ensuring safety monitoring and risk management at the construction site.

[0014] Furthermore, the method for obtaining a set of environmental perception parameter sequences by calling data based on the risk mitigation time window interactive environment perception integrator includes: The environmental sensing integrator integrates a thermocouple, a capacitive humidity sensor, an integrated circuit microphone, an electrochemical sensor, and a laser particulate matter sensor; based on the risk mitigation time window, the environmental sensing integrator interacts with the data to retrieve the environmental sensing parameter sequence set, wherein the environmental sensing parameter sequence set includes a spatial limit temperature sequence, a spatial humidity change sequence, a spatial noise change sequence, a harmful air component sequence, and a spatial dust concentration sequence.

[0015] Within the construction monitoring space, the environmental sensing integrator can perceive and collect key data about the construction environment in real time. The integrator integrates multiple sensors to capture and record changes in various environmental parameters. Specifically, it integrates thermocouples, a capacitive humidity sensor, an integrated circuit microphone, an electrochemical sensor, and a laser particulate sensor. Thermocouples are used to measure the limiting temperature of the space and generate a sequence of limiting temperatures. Thermocouples can provide accurate temperature readings across different temperature ranges, thus obtaining the temperature limits and trends of the construction environment. The capacitive humidity sensor is used to detect changes in humidity and generate a sequence of humidity changes. The capacitive humidity sensor can quickly respond to minute changes in humidity, providing the system with real-time and accurate humidity data. The integrated circuit microphone is used to capture changes in noise at the construction site and generate a sequence of noise changes. The integrated circuit microphone can record the intensity and frequency distribution of noise to assess the impact of construction noise on workers and the environment. The electrochemical sensor is used to detect harmful components in the air, such as carbon monoxide, and generate a sequence of harmful air components. The electrochemical sensor has high sensitivity and selectivity, enabling accurate identification of harmful substances in the air. Laser particulate sensors are used to measure dust concentration in the space, generating a spatial dust concentration sequence. These sensors enable real-time monitoring of airborne particulate matter concentration, helping the system assess the impact of the construction environment on worker health. Based on a pre-built risk mitigation time window, the system interacts periodically or in real-time with the environmental sensing integrator, accessing the environmental sensing parameter sequence sets collected by these sensors. These environmental sensing parameter sequence sets include spatial limit temperature sequences, spatial humidity change sequences, spatial noise change sequences, hazardous air component sequences, and spatial dust concentration sequences. The spatial limit temperature sequence records the highest temperature in the space at each time point; the spatial humidity change sequence records real-time humidity data; the spatial noise change sequence records the intensity and frequency distribution of noise; the hazardous air component sequence records the concentration changes of hazardous components; and the spatial dust concentration sequence records dust concentration data.

[0016] Based on the set of environmental perception parameters, environmental anomaly identification is performed to obtain the construction environment anomaly coefficient.

[0017] Environmental anomaly identification is performed based on a set of environmentally sensed parameter sequences to determine whether any abnormalities exist in the construction environment, and an anomaly coefficient is calculated to quantify the degree of such anomalies. Specifically, according to construction environment standards and safety requirements, corresponding normal thresholds are set for each environmental parameter. These thresholds can be set based on historical data, industry standards, or expert recommendations. The set of environmentally sensed parameter sequences is compared with the set thresholds to identify abnormal data points that do not meet the thresholds. Based on the detected abnormal data points, the construction environment anomaly coefficient is calculated.

[0018] Furthermore, such as Figure 2 As shown, environmental anomaly identification is performed based on the environmental perception parameter sequence set to obtain the construction environment anomaly coefficient. The method includes: An interactive method is used to obtain a set of construction qualification thresholds, and the environmental perception parameter sequence set is mapped using the construction qualification thresholds to perform data masking, thereby obtaining a set of environmental risk parameter sequences. Pre-set data extraction rules are used, and data collection and analysis are performed on the environmental risk parameter sequence set based on the data extraction rules to obtain multiple environmental risk anomaly arrays. A construction environment evaluation function is pre-constructed, and the multiple environmental risk anomaly arrays are synchronized to the construction environment evaluation function to perform environmental anomaly identification and obtain the construction environment anomaly coefficient.

[0019] Preferably, a construction qualification threshold set is determined through interaction with construction management personnel, safety experts, or relevant standards organizations. This set includes the qualification ranges or thresholds for various environmental sensing parameters (such as temperature, humidity, noise, harmful air components, dust concentration, etc.). Each data point in the environmental sensing parameter sequence set is compared with the construction qualification threshold set. If a data point exceeds its corresponding qualification threshold, it is marked as a risk data point; otherwise, it is marked as a normal data point. Normal data points are masked, and risk data points are retained. The sequence segments of all marked risk data points are extracted to form an environmental risk parameter sequence set. Data extraction rules are preset according to the characteristics and needs of the construction environment, including conditions such as time intervals, data fluctuation ranges, and the number of consecutive anomalies. For example, it can be set that if an environmental parameter exceeds its qualification threshold more than twice within three consecutive time units, the data within that time period is extracted into an environmental risk anomaly array. Based on the preset data extraction rules, data collection and analysis are performed on the environmental risk parameter sequence set. Environmental risk anomaly arrays that meet the conditions are extracted according to the data extraction rules. Each array contains consecutive environmental risk data points within a certain period. A construction environment evaluation function is pre-constructed based on the characteristics and needs of the construction environment. The construction environment evaluation function should be able to comprehensively consider multiple environmental risk anomaly arrays and provide the overall degree of anomaly in the construction environment. Multiple environmental risk anomaly arrays are synchronized to the construction environment evaluation function to perform environmental anomaly identification. The evaluation function then calculates the overall degree of anomaly in the construction environment and outputs the construction environment anomaly coefficient. The level of the construction environment anomaly coefficient reflects the magnitude of the construction environment risk.

[0020] Furthermore, the method further includes: pre-constructing a construction environment evaluation function, synchronizing the multiple environmental risk anomaly arrays to the construction environment evaluation function, performing environmental anomaly identification, and obtaining the construction environment anomaly coefficients; the method also includes: Based on the data extraction rules, data collection and analysis are performed on the spatial humidity change sequence to obtain a humidity risk anomaly array, wherein the humidity risk anomaly array includes abnormal humidity maximum value, abnormal humidity minimum value, abnormal humidity mean value, and humidity extreme value collection interval; and so on, based on the data extraction rules, data collection and analysis are performed on the environmental risk parameter sequence set to obtain the multiple environmental risk anomaly arrays.

[0021] The construction environment evaluation function is as follows: ; in, For the first The environmental risk anomaly coefficient of the various perception indicators For the first Maximum environmental risk of various perception indicators For the first Minimum environmental risk values ​​of various perception indicators For the first The average environmental risk of various perception indicators For the first The data collection time interval between the maximum and minimum environmental risk values ​​of various sensing indicators This represents the change in the difference between the maximum and minimum values ​​of environmental risk.

[0022] ; in, The construction environment anomaly coefficient is mentioned above. This refers to the environmental temperature anomaly coefficient. This refers to the environmental humidity anomaly coefficient. This represents the environmental noise anomaly coefficient. The abnormal coefficient of harmful components, The dust concentration anomaly coefficient is used; the multiple environmental risk anomaly arrays are synchronized to the construction environment evaluation function to perform environmental anomaly identification and obtain the construction environment anomaly coefficient.

[0023] Based on data extraction rules, the spatial humidity variation sequence is analyzed to identify abnormal humidity maxima, minimums, averages, and extreme value collection intervals. The extreme value collection interval refers to the time interval between the maximum and minimum values, for example, 37 seconds. Similarly, data collection and analysis are performed on other environmental risk parameter sequences (such as temperature, noise, hazardous air components, dust concentration, etc.), and their respective environmental risk anomaly arrays are obtained based on data extraction rules. For the i-th sensing indicator (such as temperature, humidity, noise, etc.), its environmental risk anomaly coefficient is calculated by the construction environment evaluation function. For the first The environmental risk anomaly coefficient of the various perception indicators For the first Maximum environmental risk of various perception indicators For the first Minimum environmental risk values ​​of various perception indicators For the first The average environmental risk of various perception indicators For the first The data collection time interval between the maximum and minimum environmental risk values ​​of various sensing indicators The difference between the maximum and minimum environmental risk values ​​is the measure of change, reflecting the dynamic trend of fluctuations in construction environmental parameters. The Construction Environmental Anomaly Coefficient (ED) is a weighted combination of the environmental risk anomaly coefficients of various perceived indicators. The construction environment anomaly coefficient is mentioned above. This refers to the environmental temperature anomaly coefficient. This refers to the environmental humidity anomaly coefficient. This represents the environmental noise anomaly coefficient. The abnormal coefficient of harmful components, This represents the dust concentration anomaly coefficient. Multiple environmental risk anomaly arrays are synchronized to the construction environment assessment function to calculate the construction environment anomaly coefficient.

[0024] A pre-built target identification unit is provided, wherein the target identification unit is communicatively connected to a construction monitoring camera, the construction monitoring camera is deployed in the construction monitoring space, and the construction monitoring camera integrates an adjustable aiming scope.

[0025] A pre-built target identification unit is incorporated and communicates with the construction monitoring camera. This unit receives video streams from the camera in real time and automatically identifies risky behaviors within the monitored construction area. The monitoring cameras are strategically deployed to ensure coverage of the areas requiring monitoring. They collect video data in real time, capturing visual information about construction activities. Each camera integrates an adjustable scope, where magnification, focal length, illumination, aiming point, and horizontal and vertical positions are adjustable to accurately identify targets at distances or in detail. The adjustable scope works in conjunction with the camera's control system to enable rapid adjustments to focal length and viewing angle.

[0026] Furthermore, the method for pre-constructing the target object recognition unit includes: The system interactively obtains a set of construction types within the construction monitoring space, including multiple construction types. Based on these multiple construction types, it interactively obtains multiple sample risk behavior image sets and performs recognition time and recognition space analysis on these image sets to obtain standard behavior duration and standard behavior persistence space. Based on the standard behavior duration, it constructs a video segmentation processing layer, and based on the standard behavior persistence space, it constructs a video block processing layer. Based on the multiple construction types and the multiple sample risk behavior image sets, it constructs multiple construction risk identification branches, and by connecting these branches in parallel, it generates a target detection layer. Finally, it cascades the video segmentation processing layer, the video block processing layer, and the target detection layer to obtain the target identification unit.

[0027] The target object recognition unit can perform video analysis and processing based on a set of construction types and a set of sample risk behavior images to identify risky behaviors in the construction monitoring space. The set of construction types in the monitoring space is obtained interactively. This set includes various construction types, such as civil engineering, electrical engineering, and pipeline installation. For each construction type, multiple sets of sample risk behavior images are interactively obtained, containing various risky behaviors that may occur within that construction type. Time and spatial analysis are performed on each sample risk behavior image set, analyzing the time required for each risky behavior to occur and end (standard behavior duration) and the spatial range of the behavior (standard behavior duration space). Based on the standard behavior duration, a video segmentation processing layer is constructed. This layer divides the continuous video stream captured by the monitoring camera into multiple time segments, each corresponding to a possible time period for a risky behavior. Based on the standard behavior duration space, a video block processing layer is constructed. This layer divides the video frames within each time segment into multiple regions or blocks to more accurately identify the specific location of the risky behavior. Multiple construction risk recognition branches are constructed based on multiple construction types and multiple sets of sample risk behavior images. Each branch is responsible for identifying risky behaviors in a specific construction type. Multiple construction risk identification branches are connected in parallel to generate the target object detection layer. This layer can simultaneously handle risk behavior identification tasks across multiple construction types. The video segmentation processing layer, video block processing layer, and target object detection layer are cascaded to form a complete target object identification unit. This unit receives video stream input from construction monitoring cameras, processes it, and outputs the identified risk behaviors.

[0028] When the construction environment anomaly coefficient meets the predefined environment anomaly threshold, the target object identification unit is activated. The target object identification unit interacts with the construction monitoring camera to retrieve historical images based on the risk mitigation time window, and performs target object analysis and identification based on the retrieval results to obtain the target target object.

[0029] When the construction environment anomaly coefficient meets the predefined environmental anomaly threshold, an anomaly is determined to exist in the construction environment, and the target identification unit is activated. After activation, the target identification unit interacts with the construction monitoring camera through a preset communication protocol according to the risk mitigation time window, and retrieves historical images within the risk mitigation time window. The retrieved historical images are analyzed and processed to obtain the target object.

[0030] Furthermore, the target identification unit interacts with the construction monitoring camera to retrieve historical images based on the risk mitigation time window, and performs target identification analysis based on the retrieval results to obtain the target target. The method includes: The target identification unit interacts with the construction monitoring camera based on the risk mitigation time window to retrieve historical images and obtain historical monitoring videos. Based on the video segmentation processing layer of the target identification unit, the historical monitoring videos are split to obtain N time-connected monitoring video slices, and a first monitoring video slice with a time-series front end is retrieved from the N monitoring video slices. Based on the video block processing layer of the target identification unit, M candidate target objects are located in the first monitoring video slice based on target detection, and the M candidate target objects are selected in the first monitoring video slice based on the standard behavior persistence space to obtain M candidate target video blocks. The multiple construction risk identification branches in the target object detection layer of the target identification unit are traversed one by one using the M candidate target video blocks to identify risk behaviors and obtain a first target object set. Similarly, the N monitoring video slices are traversed through the video block processing layer and the target identification unit to obtain N target object sets. The target target object is obtained by counting the frequency of occurrence of candidate target objects and retrieving extreme values ​​from the N target object sets.

[0031] When the construction environment anomaly coefficient meets a predefined environmental anomaly threshold, the target object identification unit is activated. Based on the risk mitigation time window, the target object identification unit interacts with the construction monitoring camera, calling up historical monitoring videos within the risk mitigation time window. The video segmentation processing layer of the target object identification unit splits the called historical monitoring video into N time-series connected monitoring video slices. From these N monitoring video slices, the first monitoring video slice at the time front end is called for analysis. The video block processing layer of the target object identification unit locates M candidate target objects in the first monitoring video slice based on a target detection algorithm. Based on the standard behavior persistence space, these M candidate target objects are selected in the video slice, forming M candidate target video blocks. The target object detection layer of the target object identification unit contains multiple construction risk identification branches, each branch corresponding to the identification of risk behaviors in different construction types. The M candidate target video blocks are used to traverse these identification branches one by one to identify risk behaviors. After identification, a first target set is obtained, containing all possible risky objects identified in the first surveillance video slice, such as construction workers exhibiting risky construction behavior within a certain time period. Similarly, the target identification unit continues to traverse the remaining N-1 surveillance video slices, performing similar video segmentation, target detection, and risky behavior identification on each slice to obtain N target sets. The frequency of occurrence of candidate target objects in these N target sets is counted. If a candidate target object is identified in multiple video slices and its frequency exceeds a certain set threshold, then this object is more likely to be the actual target target. In other words, if a construction worker exhibits risky construction behavior in multiple time periods, it indicates that the risky construction behavior is not unintentional but real. Based on the frequency count results, the candidate target object with the highest frequency is selected for extreme value analysis to further confirm whether it is the actual target target, thus obtaining the target target.

[0032] Based on the construction monitoring space, the target position characteristics of the target aiming object are determined, and the adjustable aiming scope is controlled to locally magnify the target aiming object according to the target position characteristics to obtain a magnified target field of view.

[0033] Based on the target object, determine its positional characteristics within the construction monitoring space, including its spatial coordinates (e.g., X, Y, Z coordinates) and physical dimensions (e.g., length, width, height). The spatial coordinates can be determined using image recognition techniques (e.g., feature point matching, target tracking). Physical dimensions can be estimated using pixel dimensions in the image and known camera parameters (e.g., focal length, sensor size). Once the target positional characteristics are determined, the adjustable sight can be controlled to locally magnify the target object. Based on the target's spatial coordinates, the direction and angle the sight needs to be adjusted to ensure the target object is centered within the sight's field of view. Based on the target's physical dimensions and the required magnification, the sight's focal length or magnification can be adjusted to obtain a suitable magnified field of view. After these steps, the adjustable sight will provide a magnified field of view where the target object is locally magnified and clearly displayed within the field of view.

[0034] Furthermore, the method involves determining the target position characteristics of the target object based on the construction monitoring space, and controlling the adjustable sight to locally magnify the target object according to the target position characteristics to obtain a magnified target field of view. Based on the construction monitoring space, the target position characteristics of the target aiming object are determined, wherein the target position characteristics include target spatial coordinates and target physical dimensions; risk type extraction is performed based on the target aiming object to obtain the target risk type; a risk type-magnification level table and a physical dimension-magnification level table are pre-constructed; based on the target risk type, the risk type-magnification level table is traversed to obtain a first magnification level, and based on the target physical dimensions, the physical dimension-magnification level table is traversed to obtain a second magnification level, and a target magnification level is generated based on the first magnification level and the second magnification level; the standard behavior continuous space is used as the target field of view constraint, and the target field of view constraint, target spatial coordinates, and target magnification level are synchronized to a pre-constructed aiming scope adjustment analysis model for adjustment analysis to obtain a target adjustment array; based on the target adjustment array, the adjustable aiming scope is controlled to locally magnify the target aiming object to obtain the target magnified field of view.

[0035] Within the construction monitoring space, the target's location characteristics, including its spatial coordinates and physical dimensions, are determined. Based on the target's characteristics, environmental conditions, and construction activities, the target risk type is determined. For example, if the target is a fragile building structure, the risk type might be structural collapse. Pre-built risk type-magnification level tables and physical size-magnification level tables are used to determine the magnification level based on the risk type and physical size. The risk type-magnification level table associates different risk types with specific magnification levels for more detailed observation of high-risk targets. The physical size-magnification level table determines the magnification level based on the target's physical dimensions to maintain sufficient detail during magnification. The first magnification level is found in the risk type-magnification level table based on the target risk type, and the second magnification level is found in the physical size-magnification level table based on the target physical dimensions. Combining the first and second magnification levels, the highest magnification level is selected to generate a target magnification level. The standard behavior persistence space is a predefined area or range used to limit the scope's adjustment range. The standard behavior continuous space is used as the target field of view constraint, and the target field of view constraint, target spatial coordinates, and target magnification level are synchronized to a pre-built scope adjustment analysis model. The scope adjustment analysis model calculates appropriate scope adjustment parameters based on the input parameters. Using the target adjustment array obtained from the adjustment analysis model, the adjustable scope is controlled to locally magnify the target object, thereby obtaining a magnified target field of view. Through this process, the adjustable scope provides a magnified field of view, enabling construction personnel to observe and analyze the target object more clearly, facilitating more accurate decision-making and operations.

[0036] Furthermore, the methods also include: The system interactively obtains a set of sample field-of-view constraints, a set of sample spatial coordinates, a set of sample magnification levels, and a set of sample adjustment arrays. Based on the operator model library, it randomly calls target operators to construct the scope adjustment analysis model. Based on preset data partitioning rules, it partitions the set of sample field-of-view constraints, the set of sample spatial coordinates, the set of sample magnification levels, and the set of sample adjustment arrays to obtain a model training set, a model test set, and a model validation set. Based on the model training set, the model test set, and the model validation set, it trains the scope adjustment analysis model until the output accuracy meets preset requirements.

[0037] Preferably, a sample set of field-of-view constraints, a sample set of spatial coordinates, a sample set of magnification levels, and a sample set of adjustment arrays are obtained interactively. These data represent the input and output values ​​of scope adjustment under different conditions. Based on a predefined operator model library, target operators (machine learning models) are randomly invoked to construct a scope adjustment analysis model. The scope adjustment analysis model predicts a suitable adjustment array based on the input field-of-view constraints, spatial coordinates, and magnification levels. The sample data is divided into three subsets according to preset data partitioning rules (such as random sampling, stratified sampling, etc.): a model training set, a model test set, and a model validation set. The model training set is used to train the scope adjustment analysis model, i.e., to adjust the model parameters to minimize the prediction error; the model test set is used to initially evaluate the performance of the trained model, but not for adjusting model parameters; the model validation set is used to finally verify the model's generalization ability, ensuring that the model performs well on new data. Training the scope adjustment analysis model using the model training set typically involves multiple iterations, with each iteration adjusting the model parameters based on the model's prediction error. The training process continues until the model's output accuracy meets the preset requirements or reaches the predetermined number of iterations. The model test set is used to initially evaluate the performance of the trained model. Error metrics, such as mean squared error and accuracy, are calculated by comparing the model's predicted output with the actual output in the test set. If the model's performance on the test set meets the requirements, a model validation set is used for final validation. If the model also performs well on the validation set, it can be considered that the model has been successfully trained.

[0038] Using the risk mitigation time window as a function activation constraint, the monitoring and tracking function of the construction monitoring camera is activated to perform abnormal tracking on the magnified field of view of the target and obtain real-time tracking images.

[0039] Setting the risk mitigation time window as the activation constraint for the construction monitoring camera's tracking function means that the camera's tracking function will be activated within the risk mitigation time window. When the tracking function is triggered within this window, the camera will begin capturing images of the target's magnified field of view and perform anomaly tracking, continuously capturing images of the target's magnified field of view to obtain real-time tracking images. In this way, the construction monitoring camera can automatically activate its tracking function within the risk mitigation time window, perform anomaly tracking of the target's magnified field of view, and provide monitoring personnel with real-time tracking images for timely detection and response to potential anomalies.

[0040] Based on the real-time tracking images, a construction risk mitigation verification is performed, and based on the verification results, the calibration reset of the adjustable sight is controlled.

[0041] The construction risk mitigation verification is performed based on real-time tracking images. This involves determining whether the target object is still within the construction monitoring space based on the real-time tracking images. If it is not, it indicates that the construction risk has been mitigated, and the verification result is obtained. Based on the verification result, the calibration of the adjustable sight is reset, returning it to the sight parameter settings used for full-space monitoring of the construction monitoring space.

[0042] Furthermore, the method of performing construction risk mitigation verification based on the real-time tracking image and controlling the calibration reset of the adjustable sight based on the verification result includes: The real-time tracking image is input into the aiming object recognition unit to obtain an updated aiming object; the updated aiming object and the target aiming object are compared by feature to obtain an object comparison result; if the object comparison result is consistent, a construction risk warning is generated; if the object comparison result is inconsistent, the adjustable aiming scope is controlled to perform calibration reset.

[0043] Real-time tracking images captured by construction monitoring cameras are input into the target object recognition unit. The unit analyzes the images, identifies the target object, and uses it as the updated target object. Features of the updated target object are extracted and compared with those of the target target object. Based on the comparison results, it is determined whether the updated and target targets are identical. If the comparison shows they are identical, it indicates that the monitored target object has not changed, but other risk factors (such as abnormal behavior or environmental changes) may have emerged. In this case, a construction risk alert is generated to notify relevant personnel of the potential risk and to implement appropriate risk control measures. If the comparison shows they are inconsistent, it indicates that the target target object does not exhibit any construction risk behavior, the risk has been mitigated, and there is no need for magnified field-of-view monitoring. The adjustable scope is then calibrated and reset, returning to the parameter settings for full-space monitoring of the construction area.

[0044] In summary, the embodiments of this application have at least the following technical effects: First, a risk mitigation time window is pre-constructed. Based on this window, data is retrieved from an interactive environmental perception integrator to obtain a set of environmental perception parameter sequences. The environmental perception integrator is pre-deployed within the construction monitoring space. Next, environmental anomaly identification is performed based on the environmental perception parameter sequence set to obtain the construction environment anomaly coefficient. Then, a target identification unit is pre-constructed, which is communicatively connected to a construction monitoring camera deployed within the construction monitoring space. Each construction monitoring camera integrates an adjustable aiming scope. When the construction environment anomaly coefficient meets a predefined environmental anomaly threshold, the target identification unit is activated. This unit retrieves historical images from the construction monitoring camera based on the risk mitigation time window and analyzes and identifies the target based on the retrieved results, obtaining the target target. Further, the target position characteristics of the target target are determined based on the construction monitoring space. The adjustable aiming scope is then controlled to locally magnify the target target based on these characteristics, obtaining a magnified field of view. Finally, using the risk mitigation time window as a functional activation constraint, the monitoring and tracking function of the construction monitoring camera is activated to perform anomaly tracking on the magnified field of view, obtaining real-time tracking images. Finally, a construction risk mitigation verification is performed based on real-time tracking images, and the calibration reset of the adjustable sight is controlled based on the verification results. This solves the technical problem in existing technologies that make it difficult to accurately capture anomalies in the construction environment during construction monitoring, and achieves the technical effect of improving the level of construction monitoring by integrating a sight into the monitoring system.

[0045] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0046] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0047] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A sight calibration method based on environmental perception, characterized in that, The method includes: A risk mitigation time window is pre-constructed, and data is called based on the risk mitigation time window interactive environment perception integrator to obtain an environmental perception parameter sequence set, wherein the environment perception integrator is pre-deployed in the construction monitoring space; Based on the environmental perception parameter sequence set, perform environmental anomaly identification to obtain the construction environment anomaly coefficient; A pre-built target recognition unit is provided, wherein the target recognition unit is communicatively connected to a construction monitoring camera, the construction monitoring camera is deployed in the construction monitoring space, and the construction monitoring camera integrates an adjustable aiming scope; When the construction environment anomaly coefficient meets the predefined environment anomaly threshold, the target object identification unit is activated. The target object identification unit interacts with the construction monitoring camera to retrieve historical images based on the risk mitigation time window, and performs target object analysis and identification based on the retrieval results to obtain the target target object. Based on the construction monitoring space, the target position characteristics of the target aiming object are determined, and the adjustable aiming scope is controlled to locally magnify the target aiming object according to the target position characteristics to obtain a magnified target field of view; Using the risk mitigation time window as a function activation constraint, the monitoring and tracking function of the construction monitoring camera is activated to perform abnormal tracking on the magnified field of view of the target and obtain real-time tracking images; Based on the real-time tracking images, a construction risk mitigation verification is performed, and based on the verification results, the calibration reset of the adjustable sight is controlled.

2. The aiming scope calibration method based on environmental perception as described in claim 1, characterized in that, The method further includes retrieving data from the risk mitigation time window interactive environment perception integrator to obtain an environment perception parameter sequence set, and also includes: The environmental sensing integrator integrates a thermocouple, a capacitive humidity sensor, an integrated circuit microphone, an electrochemical sensor, and a laser particulate sensor. Based on the risk mitigation time window, the environmental sensing integrator is interacted with to retrieve data and obtain the environmental sensing parameter sequence set, which includes spatial limit temperature sequence, spatial humidity change sequence, spatial noise change sequence, harmful air component sequence, and spatial dust concentration sequence.

3. The aiming scope calibration method based on environmental perception as described in claim 2, characterized in that, Based on the environmental perception parameter sequence set, environmental anomaly identification is performed to obtain the construction environment anomaly coefficient. The method further includes: The construction qualification threshold set is obtained interactively, and the construction qualification threshold set is used to map the environmental perception parameter sequence set to perform data masking, thereby obtaining the environmental risk parameter sequence set; Preset data extraction rules, and perform data collection and analysis on the environmental risk parameter sequence set based on the data extraction rules to obtain multiple environmental risk anomaly arrays; A construction environment evaluation function is pre-constructed, and the multiple environmental risk anomaly arrays are synchronized to the construction environment evaluation function to perform environmental anomaly identification and obtain the construction environment anomaly coefficient.

4. The aiming scope calibration method based on environmental perception as described in claim 3, characterized in that, The method further includes: pre-constructing a construction environment evaluation function, synchronizing the multiple environmental risk anomaly arrays to the construction environment evaluation function, performing environmental anomaly identification, and obtaining the construction environment anomaly coefficients. Based on the data extraction rules, data collection and analysis are performed on the spatial humidity change sequence to obtain a humidity risk anomaly array, wherein the humidity risk anomaly array includes abnormal humidity maximum value, abnormal humidity minimum value, abnormal humidity mean value, and humidity extreme value collection interval. Similarly, based on the data extraction rules, data collection and analysis are performed on the environmental risk parameter sequence set to obtain the multiple environmental risk anomaly arrays; The construction environment evaluation function is as follows: ； Among them, is the first The environmental risk anomaly coefficient of the various perception indicators For the first Maximum environmental risk of various perception indicators For the first Minimum environmental risk values ​​of various perception indicators For the first The average environmental risk of various perception indicators For the first The data collection time interval between the maximum and minimum environmental risk values ​​of a certain sensing indicator; ； in, The construction environment anomaly coefficient is mentioned above. This refers to the environmental temperature anomaly coefficient. This refers to the environmental humidity anomaly coefficient. This represents the environmental noise anomaly coefficient. The abnormal coefficient of harmful components, This is the dust concentration anomaly coefficient; The multiple environmental risk anomaly arrays are synchronized to the construction environment evaluation function to perform environmental anomaly identification and obtain the construction environment anomaly coefficient.

5. The aiming scope calibration method based on environmental perception as described in claim 1, characterized in that, The method further includes: pre-constructing an object recognition unit; The construction type set of the construction monitoring space is obtained interactively, wherein the construction type set includes multiple construction types; Multiple sample risk behavior image sets are obtained based on the interaction of the various construction types, and recognition time and recognition space analysis are performed based on the multiple sample risk behavior image sets to obtain the standard behavior duration and standard behavior duration space. A video segmentation processing layer is constructed based on the standard behavior duration, and a video block processing layer is constructed based on the standard behavior duration space. Based on the various construction types and multiple sample risk behavior image sets, multiple construction risk identification branches are constructed, and the target detection layer is generated by connecting the multiple construction risk identification branches in parallel. The video segmentation processing layer, the video block processing layer, and the target detection layer are cascaded together to obtain the target recognition unit.

6. The aiming scope calibration method based on environmental perception as described in claim 5, characterized in that, The target identification unit interacts with the construction monitoring camera to retrieve historical images based on the risk mitigation time window, and analyzes and identifies the target based on the retrieval results to obtain the target target. The method further includes: The target identification unit interacts with the construction monitoring camera based on the risk mitigation time window to retrieve historical images and obtain historical monitoring videos. Based on the target identification unit, the video segmentation processing layer splits the historical surveillance video to obtain N time-connected surveillance video slices, and calls the first surveillance video slice of the time front end from the N surveillance video slices. Based on the target identification unit, the video block processing layer locates M candidate target objects in the first monitoring video slice based on target detection, and then selects the M candidate target objects in the first monitoring video slice based on the standard behavior persistence space to obtain M candidate target video blocks. The first target set is obtained by sequentially traversing the multiple construction risk identification branches in the target detection layer of the target identification unit using the M candidate target video blocks to identify risk behaviors. Similarly, the video segmentation processing layer and the target identification unit traverse the N surveillance video segments to obtain N target sets; The target target is obtained by counting the frequency of occurrence of candidate target targets and calling extreme values ​​for the N target target sets.

7. The aiming scope calibration method based on environmental perception as described in claim 6, characterized in that, Based on the construction monitoring space, the target position characteristics of the target aiming object are determined, and the adjustable aiming scope is controlled to locally magnify the target aiming object according to the target position characteristics to obtain a magnified target field of view. The method further includes: The target location features of the target aiming object are determined based on the construction monitoring space, wherein the target location features include target spatial coordinates and target physical dimensions; Based on the target object, risk type extraction is performed to obtain the target risk type; Pre-constructed risk type - scale-up level table and physical size - scale-up level table; Based on the target risk type, the first amplification level is obtained by traversing the risk type-amplification level table; based on the target physical size, the second amplification level is obtained by traversing the physical size-amplification level table; and a target amplification level is generated based on the first amplification level and the second amplification level. The standard behavior persistent space is used as the target field of view constraint. The target field of view constraint, target spatial coordinates and target magnification level are synchronized to the pre-constructed scope adjustment analysis model for adjustment analysis to obtain the target adjustment array. Based on the target adjustment array, the adjustable sight is controlled to locally magnify the target object, thereby obtaining the magnified field of view of the target.

8. The aiming scope calibration method based on environmental perception as described in claim 7, characterized in that, The method further includes: Interactively obtain the sample field of view constraint set, sample spatial coordinate set, sample magnification level set, and sample adjustment array set; The aiming scope adjustment analysis model is constructed by randomly calling target operators based on the operator model library; Based on preset data partitioning rules, the sample field of view constraint set, sample spatial coordinate set, sample magnification level set, and sample adjustment array set are divided to obtain the model training set, model test set, and model validation set. The aiming scope adjustment analysis model is trained based on the model training set, model test set, and model verification set until the output accuracy meets the preset requirements.

9. The aiming scope calibration method based on environmental perception as described in claim 7, characterized in that, The method further includes performing construction risk mitigation verification based on the real-time tracking image, and controlling the calibration reset of the adjustable sight based on the verification result. The real-time tracking image is input into the target recognition unit to obtain the updated target; The updated targeting object and the target targeting object are compared by feature comparison to obtain object comparison results; If the comparison results of the objects are consistent, a construction risk warning will be generated; If the object comparison results are inconsistent, the adjustable sight will be controlled to perform a calibration reset.

Images